Re: Review Request 12747: Fix configuration of Jetty-based execution so that CloudStack can execute scripts properly

[Hugo Trippaers]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12747/#review23488
---

Ship it!


Ship It!

- Hugo Trippaers


On July 18, 2013, 9:34 p.m., Donal Lafferty wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12747/
> ---
> 
> (Updated July 18, 2013, 9:34 p.m.)
> 
> 
> Review request for cloudstack, Chip Childers, Devdeep Singh, and Hugo 
> Trippaers.
> 
> 
> Bugs: CLOUDSTACK-3650
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CloudStack scripts fail when running on Jetty because the executable flag is 
> not set on their permissions.
> 
> 
> Diffs
> -
> 
>   client/pom.xml 32ab94a3cc80d1136d962cad6493461f1d366e29 
> 
> Diff: https://reviews.apache.org/r/12747/diff/
> 
> 
> Testing
> ---
> 
> Deployed sample XenServer cloud, verified that error message no longer 
> appears.
> 
> Inspected file system as well.
> 
> 
> Thanks,
> 
> Donal Lafferty
> 
>

Re: Review Request 12233: CLOUDSTACK-3324: Fix/Improve upgrade system vm templates 41 to 42 (Stable master Patch)

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12233/#review23489
---


Commit 85e32552a33bc93d00bdfc9297fad5daac7f0a27 in branch refs/heads/4.2 from 
Harikrishna Patnala
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=85e3255 ]

CLOUDSTACK-3324: Fix/Improve upgrade system vm template 41 to 42

Signed-off-by: Abhinandan Prateek 


- ASF Subversion and Git Services


On July 2, 2013, 4:21 p.m., Harikrishna Patnala wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12233/
> ---
> 
> (Updated July 2, 2013, 4:21 p.m.)
> 
> 
> Review request for cloudstack, Abhinandan Prateek and Kishan Kavala.
> 
> 
> Bugs: CLOUDSTACK-3324
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3324: Fix/Improve upgrade system vm template 41 to 42
> 
> During upgrade system vm template from 4.1 to 4.2 there are case sensitive 
> checks happening which needs to be fixed and improve by generalizing the code 
> for all hypervisors 
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java 90a311b 
> 
> Diff: https://reviews.apache.org/r/12233/diff/
> 
> 
> Testing
> ---
> 
> Tested locally
> 
> 
> Thanks,
> 
> Harikrishna Patnala
> 
>

RE: code formatting for enums

[Donal Lafferty]

> -Original Message-
> From: Frank Zhang [mailto:frank.zh...@citrix.com]
> Sent: 19 July 2013 12:19 AM
> To: dev@cloudstack.apache.org
> Subject: RE: code formatting for enums
> 
> Then the most important factor of name
> convention is user experience where Running is more user friendly than
> IS_RUNNING or whatever all capitalized sentence splitting by underscore.
> 
[Donal Lafferty] 
User experience?  Tell me about it.

Consider how much easier it is to use HypervisorType.HyperV rather than 
HypervisorType.Hyperv

E.g. try running 'grep -R Hyperv *' some time :)

I'm in favour of *anything* that changes the 'Hyperv' hypervisor type to a 
unique sequence.

reviewer wanted

[Daan Hoogland]

H,

Can someone have a quick look at review request 12685 , please?

thanks,
Daan

New user - Introduction

[René Diepstraten]

Hi everyone.

Since I have joined the mailing list, I'd like to say hello and introduce
myself.
My name is Rene Diepstraten, am a system administrator and the company I
work for ( PCextreme ) uses Apache CloudStack.

I'd like to contribute here and there with my skills in python and bash,
and perhaps learn some java.
A few small patches that I wrote have been accepted into Master, and I'd
like to contribute more to this great project in the future.

Best regards,

Rene Diepstraten

Re: Query String Request Authentication(QSRA) support by S3 providers

[Thomas O'Dowd]

Hi Min,

I will have time next week to check through the bugs and verify
everything. Thanks for checking on me.

Tom.

On Thu, 2013-07-18 at 23:37 +, Min Chen wrote:
> Thanks Jessica. Tom, did you still see the issue?
> 
> -min
> 
> On 7/8/13 1:20 PM, "Jessica Wang"  wrote:
> 
> >Min, 
> >
> >> would you please take a look at this to see if UI can disable decoding
> >>in displaying this download template url
> >> returned from API?
> >
> >I just changed UI to not decode the URL returned in extractTemplate,
> >extractIso API.
> >
> >Jessica
> >
> >
> >-Original Message-
> >From: Min Chen 
> >Sent: Wednesday, July 03, 2013 5:53 PM
> >To: dev@cloudstack.apache.org; Thomas O'Dowd
> >Cc: Jessica Wang
> >Subject: Re: Query String Request Authentication(QSRA) support by S3
> >providers
> >
> >Jessica, would you please take a look at this to see if UI can disable
> >decoding in displaying this download template url returned from API?
> >
> >Thanks
> >-min
> >
> >On 7/3/13 5:38 PM, "Min Chen"  wrote:
> >
> >>By examining further what returned from extractTemplateCmd api, I
> >>realized
> >>that the URL returned from API is different from what is displayed from
> >>pop-up dialog from UI. Directly using the link returned from API (with /
> >>encoded as %2F can successfully download the template. So the issue may
> >>not be that bad to upgrade Amazon SDK, but a simple UI issue. That is, is
> >>it possible for UI not to decode URLEncoded string in this case?
> >>
> >>Thanks
> >>-min
> >>
> >>On 7/3/13 5:22 PM, "Min Chen"  wrote:
> >>
> >>>Hi Tom,
> >>>
> >>>   I can reproduce this issue using Cloudian, after investigation, I
> >>>realized that this is a bug in Amazon SDK we have used, based on this
> >>>thread: 
> >>>http://stackoverflow.com/questions/15473582/amazon-s3-presigned-urls-esc
> >>>a
> >>>p
> >>>e
> >>>-the-slashes-in-the-key. When generatePresignedUrl is called it takes
> >>>the
> >>>entire key and escapes it, and then creates a signature using the
> >>>escaped
> >>>key. You cannot use the signature from the escaped key and combine it
> >>>with
> >>>the unescaped key in the URL. See the bug code here:
> >>>
> >>>   String resourcePath = "/" +
> >>>   ((bucketName != null) ? bucketName + "/" : "") +
> >>>   ((key != null) ? ServiceUtils.urlEncode(key) : "") +
> >>>   ((subResource != null) ? "?" + subResource : "");
> >>>
> >>>We have two options to fix this:
> >>>   1. Either upgrade Amazon SDK to use 1.4.3 version, someone in that
> >>>thread
> >>>claimed that it is fixed in that version, but I haven't checked that.
> >>>Currently CloudStack is using 1.3.21. Not sure if this will break
> >>>CloudStack cloud_bridge.
> >>>   2. Workaround by creating customized AmazonS3Client to change the
> >>>internal implementation on this.
> >>>
> >>>   Thanks
> >>>   -min
> >>>
> >>>
> >>>
> >>>On 7/2/13 11:31 PM, "Thomas O'Dowd"  wrote:
> >>>
> Excellent. The link is there now. Thank you Min. I verified that bug
> and
> closed it.
> 
> However - now that I can finally click the download link... I ran into
> the issue that the link doesn't work on AWS or Cloudian. Please see
> this
> bug for details (latest 4.2 updates included in my test).
> 
> https://issues.apache.org/jira/browse/CLOUDSTACK-3341
> 
> Tom.
> 
> On Tue, 2013-07-02 at 22:54 +, Min Chen wrote:
> > Hi Tom,
> > I investigated this issue through the db dump you provided in 
> > the
> >bug,
> > this is an issue with our db view template_view creation script, and
> >it
> > has been fixed in resolving
> > another bug (https://issues.apache.org/jira/browse/CLOUDSTACK-3314).
> >I
> > have verified the fix using your db dump on my local setup. Please
> >check
> > out latest 4.2 or master code to try again.
> > 
> > Thanks
> > -min
> > 
> > On 7/2/13 2:18 PM, "Min Chen"  wrote:
> > 
> > >Tom, this seems like an issue with entry stored in our DB. I will
> >take
> >a
> > >look at this bug and update you. Just to clarify, this symptom only
> > >happens when you register these templates to Amazon S3, not for
> >Cloudian
> > >or RiakCS S3, right?
> > >
> > >Thanks
> > >-min
> > >
> > >On 7/1/13 7:27 PM, "Thomas O'Dowd"  wrote:
> > >
> > >>Yes thanks Jessica. I re-opened the bug again. I know its not a gui
> > >>problem per-say in that the template is not ready to show the
> >download
> > >>link. However, it never becomes ready is the actual problem. What
> >sets
> > >>the "isready" property to true? As far as I can see, the objects in
> >the
> > >>S3 stores (AWS or Cloudian) are complete and from my perspective
> >"ready"
> > >>to download/use. It sounds like a bug when registering the
> >template.
> > >>
> > >>Tom.
> > >>
> > >>On Mon, 2013-07-01 at 18:54 +, Jessica Wang wrote:
> 

Re: documentation links to epub/pdf broken

[Sebastien Goasguen]

On Jul 19, 2013, at 2:54 AM, Jörgen Maas  wrote:

> Hi,
> 
> Some links for the 4.0.2 pdf's are also invalid:
> - api developers guide
> - installation guide
> 
> Can someone take a look at this please?

fixed

> 
> Thanks!
> 
> 
> 
> On Wed, Jul 17, 2013 at 7:01 PM, Jörgen Maas  wrote:
> 
>> No sir, thank you for fixing this! :)
>> 
>> 
>> On Wed, Jul 17, 2013 at 6:35 PM, Sebastien Goasguen wrote:
>> 
>>> 
>>> On Jul 17, 2013, at 12:21 PM, Sebastien Goasguen 
>>> wrote:
>>> 
 
 On Jul 17, 2013, at 4:47 AM, Jörgen Maas  wrote:
 
> Hi guys,
> 
> Tried to download the documentation PDF's for version 4.1.0; but the
>>> links
> to the pdf files are almost all incorrect (they all point to the
>>> release
> notes).
> 
> Can someone take a look at this please?
 
 Looking into it now. I can confirm for the pdfs. the epubs seem fine.
 
 I will get it fixed
 
 -sebastien
>>> 
>>> Jorgen, this should be fixed now (clear your cache just in case). thanks
>>> for reporting it.
>>> 
 
> 
> Thanks!
> 
> --
> Grtz,
> Jörgen Maas
 
>>> 
>>> 
>> 
>> 
>> --
>> Grtz,
>> Jörgen Maas
>> 
> 
> 
> 
> -- 
> Grtz,
> Jörgen Maas

Re: code formatting for enums

[Daan Hoogland]

all,

I am for adding a convention on enums to the coding conventions, thought I
don't like Johns best, I'm fine with it and with Alex'. Let's add this one
and my proposal on method length and number of methods in classes to the
conventions and keep in mind that a convention is not a law. We won't lynch
anybody over them. Also migration to adherence does not have to take place
today, but may well wait till 5.0 and/or 6.0. In cases it isn't harmful a
newbee knows he can quite safely propose a change and get a feel for the
process this way. By adding them we do state the necessity and our wish for
those styles of coding and that is the first step we should take.

my € 0,02,
Daan


On Fri, Jul 19, 2013 at 9:28 AM, Donal Lafferty
wrote:

> > -Original Message-
> > From: Frank Zhang [mailto:frank.zh...@citrix.com]
> > Sent: 19 July 2013 12:19 AM
> > To: dev@cloudstack.apache.org
> > Subject: RE: code formatting for enums
> >
> > Then the most important factor of name
> > convention is user experience where Running is more user friendly than
> > IS_RUNNING or whatever all capitalized sentence splitting by underscore.
> >
> [Donal Lafferty]
> User experience?  Tell me about it.
>
> Consider how much easier it is to use HypervisorType.HyperV rather than
> HypervisorType.Hyperv
>
> E.g. try running 'grep -R Hyperv *' some time :)
>
> I'm in favour of *anything* that changes the 'Hyperv' hypervisor type to a
> unique sequence.
>
>

Re: Demo Lab - To rent or To buy

[Sebastien Goasguen]

On Jul 18, 2013, at 7:56 PM, Ian Duffy  wrote:

> Hi,
> 
> I'm looking for peoples personal opinions.
> 
> I'm thinking about undertaking a project for a 2 month college assignment
> sometime in the future to create an API wrapper and web front end with the
> end goal of producing something like the rackspace or linode web-ui. I do
> realise 'basic cloud' is a solved problem, but I'm not seeing any *open
> source projects* that allow for user sign up, select a package offering,
> billing, ticketing etc. along with basic instance/vm management.
> 

It's kinda of nice, but I think you have proven that you can tackle a much more 
challenging problem than this one.

tackle this instead:

https://github.com/Netflix/Cloud-Prize/wiki


> The cost of setting up a demo lab seems somewhat expensive.
> 
> My idea was: 1 xenserver with 16gb+ ram make use of local storage, 1
> cloudstack manager/secondary storage 2gb+ of ram, finally something to host
> the wrapper and web app.
> 
> What hardware would you recommend for creating a budget, portable and quiet
> demo lab?
> Do you think I would be better of renting? If so are their any companies
> you would suggestion?
> 




> Thanks,
> Ian

Re: New user - Introduction

[Wei ZHOU]

Welcome!


2013/7/19 René Diepstraten 

> Hi everyone.
>
> Since I have joined the mailing list, I'd like to say hello and introduce
> myself.
> My name is Rene Diepstraten, am a system administrator and the company I
> work for ( PCextreme ) uses Apache CloudStack.
>
> I'd like to contribute here and there with my skills in python and bash,
> and perhaps learn some java.
> A few small patches that I wrote have been accepted into Master, and I'd
> like to contribute more to this great project in the future.
>
> Best regards,
>
> Rene Diepstraten
>

Review Request 12757: CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

[Harikrishna Patnala]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12757/
---

Review request for cloudstack, Jayapal Reddy and Kishan Kavala.


Bugs: CLOUDSTACK-3634


Repository: cloudstack-git


Description
---

CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

Presently during system vm template upgrade from 4.1 to 4.2 takes care of 
inserting these parameters.
Adding now in 4.1 to 4.2 sql file so that normal developer upgrade does not 
fail in deploying router.


Diffs
-

  engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java e54a699 
  setup/db/db/schema-410to420.sql 921d5fa 

Diff: https://reviews.apache.org/r/12757/diff/


Testing
---


Thanks,

Harikrishna Patnala

Re: Review Request 12757: [4.2]CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

[Harikrishna Patnala]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12757/
---

(Updated July 19, 2013, 8:52 a.m.)


Review request for cloudstack, Jayapal Reddy and Kishan Kavala.


Summary (updated)
-

[4.2]CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade 
setup 


Bugs: CLOUDSTACK-3634


Repository: cloudstack-git


Description
---

CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

Presently during system vm template upgrade from 4.1 to 4.2 takes care of 
inserting these parameters.
Adding now in 4.1 to 4.2 sql file so that normal developer upgrade does not 
fail in deploying router.


Diffs
-

  engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java e54a699 
  setup/db/db/schema-410to420.sql 921d5fa 

Diff: https://reviews.apache.org/r/12757/diff/


Testing
---


Thanks,

Harikrishna Patnala

Re: New user - Introduction

[Prasanna Santhanam]

Welcome Ren??. You can comb through JIRA for anything that interests
you.

On Fri, Jul 19, 2013 at 10:19:14AM +0200, Ren? Diepstraten wrote:
> Hi everyone.
> 
> Since I have joined the mailing list, I'd like to say hello and introduce
> myself.
> My name is Rene Diepstraten, am a system administrator and the company I
> work for ( PCextreme ) uses Apache CloudStack.
> 
> I'd like to contribute here and there with my skills in python and bash,
> and perhaps learn some java.
> A few small patches that I wrote have been accepted into Master, and I'd
> like to contribute more to this great project in the future.
> 
> Best regards,
> 
> Rene Diepstraten

-- 
Prasanna.,


Powered by BigRock.com

Re: Is it possible for reviewer to add other reviewers in the reviewboard?

[Daan Hoogland]

I'll have a quick look on rbtool.

I don't agree at all that more ASFBot noise should be produced. I'd like to
take the time to learn lua, but the dev channel is not inviting to me
because of it. I would propose to remove the present day message from it
instead. They get send to the mail list which is good enough, isn't it? Now
that we stopped #meeting , lets make sure discussion have all space
available on #dev...???

daan


On Thu, Jul 18, 2013 at 6:18 PM, Prasanna Santhanam  wrote:

> On Thu, Jul 18, 2013 at 06:07:48PM +0200, Daan Hoogland wrote:
> > this has been kind of bugging me too. Along with unanswered questions on
> > teh list by newbees like me. As we all depend on volunteers and
> > conculeagues I don't really see a solution but reporting on outstanding
> > reviews and maybe unanswered questions. The latter can only be done
> > manually though, as it is really hard to automatically determine that a
> > mail requires reply. I kept track of the unanswered mails for one week
> > after ccc13. It were six that I didn't have time to gain knowledge to
> > answer. This is not extreme, but still may be a waste as some of the
> posing
> > people might have been thrown of the cloudstack track by them. Hugo said
> he
> > had a script querying the review board for old reviews and an automated
> > report on that would be easier. I know all you guru's do your best but a
> > weekly report, keeping us all conscious might help.
> >
> Yes - Rohit wrote the script and I send it sometimes before things
> like freeze/deadlines to alert the community but everyone's like ...
> meh.
>
> https://github.com/vogxn/RBTool (that's the tool)
>
> We should also have IRC alerts via ASFBot for every rb request posted
> and merged.  I was going to work on this with Humbeedoh (INFRA) but am
> yet to get to it. Feel free to pick it up - ASFBot is written in Lua.
>
> If you can't convince the people, try write tools around it eh?
>
>
> > As you might have guessed this is me volunteering to keep track of
> > unanswered questions for a few weeks. As we go along good ideas on how to
> > improve our way might spring up. /me is an optimist at rare occasions.
>
> I rant and go do the postive thing hoping for it to be picked up.
> Guess that makes me an optimist :)
>
> --
> Prasanna.,
>
> 
> Powered by BigRock.com
>
>

Re: New user - Introduction

[Jayapal Reddy Uradi]

Welcome Rene!

On 19-Jul-2013, at 1:49 PM, René Diepstraten 
 wrote:

> Hi everyone.
> 
> Since I have joined the mailing list, I'd like to say hello and introduce
> myself.
> My name is Rene Diepstraten, am a system administrator and the company I
> work for ( PCextreme ) uses Apache CloudStack.
> 
> I'd like to contribute here and there with my skills in python and bash,
> and perhaps learn some java.
> A few small patches that I wrote have been accepted into Master, and I'd
> like to contribute more to this great project in the future.
> 
> Best regards,
> 
> Rene Diepstraten

Re: Review Request 12757: [4.2]CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12757/#review23490
---


Commit 7c664728f7d9d82c32fae09f62f98d77b74634b5 in branch refs/heads/4.2 from 
Harikrishna Patnala
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=7c66472 ]

CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

Signed-off-by: Jayapal 


- ASF Subversion and Git Services


On July 19, 2013, 8:52 a.m., Harikrishna Patnala wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12757/
> ---
> 
> (Updated July 19, 2013, 8:52 a.m.)
> 
> 
> Review request for cloudstack, Jayapal Reddy and Kishan Kavala.
> 
> 
> Bugs: CLOUDSTACK-3634
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade 
> setup
> 
> Presently during system vm template upgrade from 4.1 to 4.2 takes care of 
> inserting these parameters.
> Adding now in 4.1 to 4.2 sql file so that normal developer upgrade does not 
> fail in deploying router.
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java e54a699 
>   setup/db/db/schema-410to420.sql 921d5fa 
> 
> Diff: https://reviews.apache.org/r/12757/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Harikrishna Patnala
> 
>

Re: Review Request 12757: [4.2]CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade setup

[Jayapal Reddy]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12757/#review23491
---

Ship it!


Ship It!

- Jayapal Reddy


On July 19, 2013, 8:52 a.m., Harikrishna Patnala wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12757/
> ---
> 
> (Updated July 19, 2013, 8:52 a.m.)
> 
> 
> Review request for cloudstack, Jayapal Reddy and Kishan Kavala.
> 
> 
> Bugs: CLOUDSTACK-3634
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3634: Adding router.template.xen/kvm/hyperv/kvm/lxc in upgrade 
> setup
> 
> Presently during system vm template upgrade from 4.1 to 4.2 takes care of 
> inserting these parameters.
> Adding now in 4.1 to 4.2 sql file so that normal developer upgrade does not 
> fail in deploying router.
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java e54a699 
>   setup/db/db/schema-410to420.sql 921d5fa 
> 
> Diff: https://reviews.apache.org/r/12757/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Harikrishna Patnala
> 
>

Re: Review Request 12752: Async commands can inject the job id

[Murali Reddy]

Ryan,

You should open a bug and add details on what the use case you are trying
to solve. Since CloudStack is not in control of UUID generation, you
should at least check that uniqueness of passed in UUID.

Also please introduce a new smoke test for 'injected job id' instead of
piggy backing the test for deploy VM.

On 19/07/13 6:02 AM, "Ryan Dietrich"  wrote:

>
>
>> On July 18, 2013, 11:34 p.m., Marcus Sorensen wrote:
>> > server/src/com/cloud/async/AsyncJobVO.java, line 141
>> > 
>>>>
>> >
>> > I'm not immediately certain why this was pulled out.
>
>Nothing is using it.  I did a scan to see if anyone is using it, and I
>didn't see anything.  Dead could should be removed, right?  Git will
>remember for us if we need it later.
>
>
>- Ryan
>
>
>---
>This is an automatically generated e-mail. To reply, visit:
>https://reviews.apache.org/r/12752/#review23457
>---
>
>
>On July 18, 2013, 11:28 p.m., Ryan Dietrich wrote:
>> 
>> ---
>> This is an automatically generated e-mail. To reply, visit:
>> https://reviews.apache.org/r/12752/
>> ---
>> 
>> (Updated July 18, 2013, 11:28 p.m.)
>> 
>> 
>> Review request for cloudstack and Marcus Sorensen.
>> 
>> 
>> Repository: cloudstack-git
>> 
>> 
>> Description
>> ---
>> 
>> I added "injectedjobid" to the BaseAsyncCmd class as a parameter.
>> If set, it will allow you to tell Cloudstack what the job id instead of
>>it choosing one.
>> A basic string length test is done to verify the variable passed in is
>>actually a UUID.
>> If it is not valid, it is ignored and the job generates it's own.
>> 
>> 
>> Diffs
>> -
>> 
>>   api/src/org/apache/cloudstack/api/BaseAsyncCmd.java 0e6f95d
>>   server/src/com/cloud/api/ApiServer.java 95f17af
>>   server/src/com/cloud/async/AsyncJobVO.java 41eccb4
>>   test/integration/smoke/test_deploy_vm.py 425aeb7
>>   tools/marvin/marvin/codegenerator.py 632b8c6
>>   tools/marvin/marvin/integration/lib/base.py 161d03c
>> 
>> Diff: https://reviews.apache.org/r/12752/diff/
>> 
>> 
>> Testing
>> ---
>> 
>> Updated marvin, updated the deploy vm test.  Ran multiple async
>>commands manually, with and without injectedjobid present, no issues
>>detected.
>> 
>> 
>> Thanks,
>> 
>> Ryan Dietrich
>> 
>>
>
>

Re: New user - Introduction

[Wido den Hollander]

Hi Rene!

For everybody here: Rene works for the company I work for to as well. I 
encouraged him to start working on CloudStack. Fix what's broken as you 
encounter things :)


Wido

On 07/19/2013 10:19 AM, René Diepstraten wrote:

Hi everyone.

Since I have joined the mailing list, I'd like to say hello and introduce
myself.
My name is Rene Diepstraten, am a system administrator and the company I
work for ( PCextreme ) uses Apache CloudStack.

I'd like to contribute here and there with my skills in python and bash,
and perhaps learn some java.
A few small patches that I wrote have been accepted into Master, and I'd
like to contribute more to this great project in the future.

Best regards,

Rene Diepstraten

Re: Review Request 12427: Fix for CLOUDSTACK-3365 and CLOUDSTACK-2536

[Koushik Das]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12427/#review23492
---


The changes related to CLOUDSTACK-2536 is not required. updateCluster needs to 
be fixed as updateConfiguration is a generic approach and shouldn't be broken 
for handling a special scenario.

- Koushik Das


On July 17, 2013, 7:37 a.m., Harikrishna Patnala wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12427/
> ---
> 
> (Updated July 17, 2013, 7:37 a.m.)
> 
> 
> Review request for cloudstack, Abhinandan Prateek, Koushik Das, and Nitin 
> Mehta.
> 
> 
> Bugs: CLOUDSTACK-2536 and CLOUDSTACK-3365
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3365: cluster level parameters 
> cluster.(cpu/memory).allocated.capacity.notificationthreshold is not 
> considering overcommit value
> 
> CLOUDSTACK-2536:  parameters (cpu/memory)overcommit ratio and 
> (cpu/memory).overprosioning.factor are redundant(cluster level)
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/alert/AlertManagerImpl.java 9b7cd27 
>   server/src/com/cloud/configuration/Config.java e77f6ce 
>   server/src/com/cloud/vm/VirtualMachineManagerImpl.java 827e233 
>   server/test/com/cloud/vm/VirtualMachineManagerImplTest.java 44e22e2 
> 
> Diff: https://reviews.apache.org/r/12427/diff/
> 
> 
> Testing
> ---
> 
> Tested locally both with actual usage and over committed values for memory. 
> 
> 
> Thanks,
> 
> Harikrishna Patnala
> 
>

Re: Review Request 12685: CLOUDSTACK-1532 added utility functions to scan URIs

[Wido den Hollander]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12685/#review23493
---


It seems good to me? Code-wise I can't find anything that isn't right.

I see that Spark404 is also mentioned here, did he get a chance to look at this 
yet?

- Wido den Hollander


On July 17, 2013, 3:46 p.m., daan Hoogland wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12685/
> ---
> 
> (Updated July 17, 2013, 3:46 p.m.)
> 
> 
> Review request for cloudstack and Hugo Trippaers.
> 
> 
> Bugs: CLOUDSTACK-1532
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> the review for the complete patch was open to long and no longer applicable, 
> so I am starting with smaller patches as this is really needed to implement 
> CLOUDSTACK-1532
> 
> 
> Diffs
> -
> 
>   api/src/com/cloud/network/Networks.java 5aede05 
>   api/test/com/cloud/network/NetworksTest.java PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/12685/diff/
> 
> 
> Testing
> ---
> 
> unit testing
> 
> 
> Thanks,
> 
> daan Hoogland
> 
>

Review Request 12758: Cloustack - 3452: Resolved ssh issue for basic zone. Added security group for vms created in basic zone to allow ssh

[Gaurav Aradhye]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12758/
---

Review request for cloudstack and Prasanna Santhanam.


Repository: cloudstack-git


Description
---

Resolved issue CLOUDSTACK - 3452: 
https://issues.apache.org/jira/browse/CLOUDSTACK-3452

Added basic security group if not present, each vm created in basic zone will 
be added to this security group.
This group has been authorized ingress traffic on port 22 (TCP connection).

Now SSH connection error is removed for basic zone.

For advanced zone, working remains the same as before.


Diffs
-

  tools/marvin/marvin/integration/lib/base.py 161d03c 

Diff: https://reviews.apache.org/r/12758/diff/


Testing
---

Tested below test cases:

integration.smoke.test_vm_life_cycle.TestVMLifeCycle.test_10_attachAndDetach_iso
 
integration.smoke.test_volumes.TestVolumes.test_02_attach_volume 
integration.smoke.test_volumes.TestCreateVolume.test_01_create_volume


Thanks,

Gaurav Aradhye

Re: Review Request 12685: CLOUDSTACK-1532 added utility functions to scan URIs

[Hugo Trippaers]

> On July 19, 2013, 9:41 a.m., Wido den Hollander wrote:
> > It seems good to me? Code-wise I can't find anything that isn't right.
> > 
> > I see that Spark404 is also mentioned here, did he get a chance to look at 
> > this yet?

Yeah, i'm ok with this.

I'm actually abusing this patch to test jenkins-reviewboard integration, thats 
why i haven't committed it.


- Hugo


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12685/#review23493
---


On July 17, 2013, 3:46 p.m., daan Hoogland wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12685/
> ---
> 
> (Updated July 17, 2013, 3:46 p.m.)
> 
> 
> Review request for cloudstack and Hugo Trippaers.
> 
> 
> Bugs: CLOUDSTACK-1532
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> the review for the complete patch was open to long and no longer applicable, 
> so I am starting with smaller patches as this is really needed to implement 
> CLOUDSTACK-1532
> 
> 
> Diffs
> -
> 
>   api/src/com/cloud/network/Networks.java 5aede05 
>   api/test/com/cloud/network/NetworksTest.java PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/12685/diff/
> 
> 
> Testing
> ---
> 
> unit testing
> 
> 
> Thanks,
> 
> daan Hoogland
> 
>

Re: Review Request 12427: Fix for CLOUDSTACK-3365: cluster level parameters cluster.(cpu/memory).allocated.capacity.notificationthreshold is not considering overcommit value

[Harikrishna Patnala]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12427/
---

(Updated July 19, 2013, 10:14 a.m.)


Review request for cloudstack, Abhinandan Prateek, Koushik Das, and Nitin Mehta.


Changes
---

removed the fix for 2536.


Summary (updated)
-

Fix for CLOUDSTACK-3365: cluster level parameters 
cluster.(cpu/memory).allocated.capacity.notificationthreshold is not 
considering overcommit value


Bugs: CLOUDSTACK-3365


Repository: cloudstack-git


Description (updated)
---

CLOUDSTACK-3365: cluster level parameters 
cluster.(cpu/memory).allocated.capacity.notificationthreshold is not 
considering overcommit value


Diffs (updated)
-

  server/src/com/cloud/alert/AlertManagerImpl.java 9b7cd27 

Diff: https://reviews.apache.org/r/12427/diff/


Testing
---

Tested locally both with actual usage and over committed values for memory. 


Thanks,

Harikrishna Patnala

Re: Review Request 12222: CLOUDSTACK-2510: Getting NPE while executing listStoragePools API command.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/1/#review23497
---


Commit 6e63f10a4f4f77c1b57f1e804b5a443c6b75 in branch refs/heads/master 
from Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=6e63f10 ]

CLOUDSTACK-2510: Getting NPE while executinng listStoragePools API command.


- ASF Subversion and Git Services


On July 2, 2013, 5:26 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/1/
> ---
> 
> (Updated July 2, 2013, 5:26 a.m.)
> 
> 
> Review request for cloudstack, Devdeep Singh, edison su, and Nitin Mehta.
> 
> 
> Bugs: CLOUDSTACK-2510
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-2510: Getting NPE while executinng listStoragePools API command.
> 
> Tried to add a primary storage to empty xen cluster; after this event not 
> able to list primary storage, getting NPE . 
> 
> 
> Diffs
> -
> 
>   
> plugins/storage/volume/default/src/org/apache/cloudstack/storage/datastore/lifecycle/CloudStackPrimaryDataStoreLifeCycleImpl.java
>  2e0ff66 
> 
> Diff: https://reviews.apache.org/r/1/diff/
> 
> 
> Testing
> ---
> 
> Tests:
> 1. Add an empty cluster (without any host).
> 2. Add a cluster wide primary storage to this cluster (It will fail with an 
> exception that "Fail to add data store").
> 3. Verified that listStoragePools API is working properly and its not listing 
> the storage pool that admin tried to add in step 2.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12544: CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting updated if detached volume is deleted.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12544/#review23495
---


Commit 3771e8a0e61385aa6bda0fd73afa961b6a6cb091 in branch refs/heads/master 
from Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=3771e8a ]

CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting 
updated if detached volume is deleted


- ASF Subversion and Git Services


On July 15, 2013, 9:16 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12544/
> ---
> 
> (Updated July 15, 2013, 9:16 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3496
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting 
> updated if detached volume is deleted.
> 
> Issue caused because of the change in storage design as a part of 
> object_store branch, fixed the check accordingly.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/storage/VolumeManagerImpl.java 72d27bc 
> 
> Diff: https://reviews.apache.org/r/12544/diff/
> 
> 
> Testing
> ---
> 
> Tests:
> 
> 1. Create a volume.
> 2. Check the primaryStorageTotal value by calling listAccounts&id= .
> 3. Attach this volume to an instance.
> 4. Detach this volume and delete it.
> 5. Again check the primayStorageTotal.
> 
> Verified the fix by following the above mentioned steps.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12629: CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating the primary storage incorretly

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12629/#review23496
---


Commit 539eb38908ede53a36d4bb680b848caf9fe469c4 in branch refs/heads/master 
from Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=539eb38 ]

CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating 
the primary storage incorretly.

The issue is that while calculating the used primary storage size, the 
updateResourceCount
API is also calculating the disk size of virtual router VM, created for that 
account and
because of this, the API is returning the incorrect result.


- ASF Subversion and Git Services


On July 17, 2013, 5:11 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12629/
> ---
> 
> (Updated July 17, 2013, 5:11 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3492
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating 
> the primary storage incorretly.
> 
> Steps: 
> 
> 1. Deploy instance with user account1 
> 2. Get the primary storage is usage using ListAccounts- It is 10 GB same as 
> ROOT disk size of this instance 
> 3. Now tried Update Resource Count for this account. With this when verified 
> the usage using listAccounts it increased to 11 GB 
> 
> 4. Added one DATA disk of 5 GB . With which usage has become 16 GB 
> 
> 5. Then tried Update Resource Count for this account. With this when verified 
> the usage using listAccounts it decreased to 11 GB 
> 
> Observation : 
> Update Resource Count for an account is updating the primary storage 
> incorrectly (Attached API log) 
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/storage/dao/VolumeDao.java fb7dc70 
>   engine/schema/src/com/cloud/storage/dao/VolumeDaoImpl.java ba85466 
>   engine/schema/src/com/cloud/vm/dao/VMInstanceDao.java e564052 
>   engine/schema/src/com/cloud/vm/dao/VMInstanceDaoImpl.java ec04085 
>   server/src/com/cloud/resourcelimit/ResourceLimitManagerImpl.java 5a9d360 
> 
> Diff: https://reviews.apache.org/r/12629/diff/
> 
> 
> Testing
> ---
> 
> Verified locally by following the steps mentioned in description.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12629: CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating the primary storage incorretly

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12629/#review23499
---


Commit 5de96db81d666a680f99e0a719ef79794c6358ff in branch refs/heads/4.2 from 
Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=5de96db ]

CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating 
the primary storage incorretly.

The issue is that while calculating the used primary storage size, the 
updateResourceCount
API is also calculating the disk size of virtual router VM, created for that 
account and
because of this, the API is returning the incorrect result.


- ASF Subversion and Git Services


On July 17, 2013, 5:11 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12629/
> ---
> 
> (Updated July 17, 2013, 5:11 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3492
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating 
> the primary storage incorretly.
> 
> Steps: 
> 
> 1. Deploy instance with user account1 
> 2. Get the primary storage is usage using ListAccounts- It is 10 GB same as 
> ROOT disk size of this instance 
> 3. Now tried Update Resource Count for this account. With this when verified 
> the usage using listAccounts it increased to 11 GB 
> 
> 4. Added one DATA disk of 5 GB . With which usage has become 16 GB 
> 
> 5. Then tried Update Resource Count for this account. With this when verified 
> the usage using listAccounts it decreased to 11 GB 
> 
> Observation : 
> Update Resource Count for an account is updating the primary storage 
> incorrectly (Attached API log) 
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/storage/dao/VolumeDao.java fb7dc70 
>   engine/schema/src/com/cloud/storage/dao/VolumeDaoImpl.java ba85466 
>   engine/schema/src/com/cloud/vm/dao/VMInstanceDao.java e564052 
>   engine/schema/src/com/cloud/vm/dao/VMInstanceDaoImpl.java ec04085 
>   server/src/com/cloud/resourcelimit/ResourceLimitManagerImpl.java 5a9d360 
> 
> Diff: https://reviews.apache.org/r/12629/diff/
> 
> 
> Testing
> ---
> 
> Verified locally by following the steps mentioned in description.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12222: CLOUDSTACK-2510: Getting NPE while executing listStoragePools API command.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/1/#review23500
---


Commit 6770c5d92dae37fc48f5b63900936616218815ea in branch refs/heads/4.2 from 
Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=6770c5d ]

CLOUDSTACK-2510: Getting NPE while executinng listStoragePools API command.


- ASF Subversion and Git Services


On July 2, 2013, 5:26 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/1/
> ---
> 
> (Updated July 2, 2013, 5:26 a.m.)
> 
> 
> Review request for cloudstack, Devdeep Singh, edison su, and Nitin Mehta.
> 
> 
> Bugs: CLOUDSTACK-2510
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-2510: Getting NPE while executinng listStoragePools API command.
> 
> Tried to add a primary storage to empty xen cluster; after this event not 
> able to list primary storage, getting NPE . 
> 
> 
> Diffs
> -
> 
>   
> plugins/storage/volume/default/src/org/apache/cloudstack/storage/datastore/lifecycle/CloudStackPrimaryDataStoreLifeCycleImpl.java
>  2e0ff66 
> 
> Diff: https://reviews.apache.org/r/1/diff/
> 
> 
> Testing
> ---
> 
> Tests:
> 1. Add an empty cluster (without any host).
> 2. Add a cluster wide primary storage to this cluster (It will fail with an 
> exception that "Fail to add data store").
> 3. Verified that listStoragePools API is working properly and its not listing 
> the storage pool that admin tried to add in step 2.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12544: CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting updated if detached volume is deleted.

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12544/#review23498
---


Commit a98c0f5b89533e9e8d3ef671c994ff9528aa5665 in branch refs/heads/4.2 from 
Sanjay Tripathi
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=a98c0f5 ]

CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting 
updated if detached volume is deleted


- ASF Subversion and Git Services


On July 15, 2013, 9:16 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12544/
> ---
> 
> (Updated July 15, 2013, 9:16 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3496
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting 
> updated if detached volume is deleted.
> 
> Issue caused because of the change in storage design as a part of 
> object_store branch, fixed the check accordingly.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/storage/VolumeManagerImpl.java 72d27bc 
> 
> Diff: https://reviews.apache.org/r/12544/diff/
> 
> 
> Testing
> ---
> 
> Tests:
> 
> 1. Create a volume.
> 2. Check the primaryStorageTotal value by calling listAccounts&id= .
> 3. Attach this volume to an instance.
> 4. Detach this volume and delete it.
> 5. Again check the primayStorageTotal.
> 
> Verified the fix by following the above mentioned steps.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12222: CLOUDSTACK-2510: Getting NPE while executing listStoragePools API command.

[Devdeep Singh]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/1/#review23501
---

Ship it!


Ship It!

- Devdeep Singh


On July 2, 2013, 5:26 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/1/
> ---
> 
> (Updated July 2, 2013, 5:26 a.m.)
> 
> 
> Review request for cloudstack, Devdeep Singh, edison su, and Nitin Mehta.
> 
> 
> Bugs: CLOUDSTACK-2510
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-2510: Getting NPE while executinng listStoragePools API command.
> 
> Tried to add a primary storage to empty xen cluster; after this event not 
> able to list primary storage, getting NPE . 
> 
> 
> Diffs
> -
> 
>   
> plugins/storage/volume/default/src/org/apache/cloudstack/storage/datastore/lifecycle/CloudStackPrimaryDataStoreLifeCycleImpl.java
>  2e0ff66 
> 
> Diff: https://reviews.apache.org/r/1/diff/
> 
> 
> Testing
> ---
> 
> Tests:
> 1. Add an empty cluster (without any host).
> 2. Add a cluster wide primary storage to this cluster (It will fail with an 
> exception that "Fail to add data store").
> 3. Verified that listStoragePools API is working properly and its not listing 
> the storage pool that admin tried to add in step 2.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12544: CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting updated if detached volume is deleted.

[Devdeep Singh]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12544/#review23503
---

Ship it!


Ship It!

- Devdeep Singh


On July 15, 2013, 9:16 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12544/
> ---
> 
> (Updated July 15, 2013, 9:16 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3496
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3496: [Regression]Primary Storage resource usage is not getting 
> updated if detached volume is deleted.
> 
> Issue caused because of the change in storage design as a part of 
> object_store branch, fixed the check accordingly.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/storage/VolumeManagerImpl.java 72d27bc 
> 
> Diff: https://reviews.apache.org/r/12544/diff/
> 
> 
> Testing
> ---
> 
> Tests:
> 
> 1. Create a volume.
> 2. Check the primaryStorageTotal value by calling listAccounts&id= .
> 3. Attach this volume to an instance.
> 4. Detach this volume and delete it.
> 5. Again check the primayStorageTotal.
> 
> Verified the fix by following the above mentioned steps.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: Review Request 12629: CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating the primary storage incorretly

[Devdeep Singh]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12629/#review23502
---

Ship it!


Ship It!

- Devdeep Singh


On July 17, 2013, 5:11 a.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12629/
> ---
> 
> (Updated July 17, 2013, 5:11 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: CLOUDSTACK-3492
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3492: [Regression]Update Resource Count for an account is updating 
> the primary storage incorretly.
> 
> Steps: 
> 
> 1. Deploy instance with user account1 
> 2. Get the primary storage is usage using ListAccounts- It is 10 GB same as 
> ROOT disk size of this instance 
> 3. Now tried Update Resource Count for this account. With this when verified 
> the usage using listAccounts it increased to 11 GB 
> 
> 4. Added one DATA disk of 5 GB . With which usage has become 16 GB 
> 
> 5. Then tried Update Resource Count for this account. With this when verified 
> the usage using listAccounts it decreased to 11 GB 
> 
> Observation : 
> Update Resource Count for an account is updating the primary storage 
> incorrectly (Attached API log) 
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/storage/dao/VolumeDao.java fb7dc70 
>   engine/schema/src/com/cloud/storage/dao/VolumeDaoImpl.java ba85466 
>   engine/schema/src/com/cloud/vm/dao/VMInstanceDao.java e564052 
>   engine/schema/src/com/cloud/vm/dao/VMInstanceDaoImpl.java ec04085 
>   server/src/com/cloud/resourcelimit/ResourceLimitManagerImpl.java 5a9d360 
> 
> Diff: https://reviews.apache.org/r/12629/diff/
> 
> 
> Testing
> ---
> 
> Verified locally by following the steps mentioned in description.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: New user - Introduction

[Nitin Mehta]

Welcome aboard Rene. Feel free to browse the wiki, JIRA bugs, code and ask
us questions :)

On 19/07/13 3:04 PM, "Wido den Hollander"  wrote:

>Hi Rene!
>
>For everybody here: Rene works for the company I work for to as well. I
>encouraged him to start working on CloudStack. Fix what's broken as you
>encounter things :)
>
>Wido
>
>On 07/19/2013 10:19 AM, René Diepstraten wrote:
>> Hi everyone.
>>
>> Since I have joined the mailing list, I'd like to say hello and
>>introduce
>> myself.
>> My name is Rene Diepstraten, am a system administrator and the company I
>> work for ( PCextreme ) uses Apache CloudStack.
>>
>> I'd like to contribute here and there with my skills in python and bash,
>> and perhaps learn some java.
>> A few small patches that I wrote have been accepted into Master, and I'd
>> like to contribute more to this great project in the future.
>>
>> Best regards,
>>
>> Rene Diepstraten
>>

Re: 41-42 db upgrade

[Harikrishna Patnala]

Hi
During system vm template upgrade from 41 to 42 takes care of insertion of 
router.template.xen/* parameters.
But checked in my part so that it does nt break any upgrade by developer.

Thanks
Harikrishna

On 18-Jul-2013, at 2:39 AM, Min Chen 
mailto:min.c...@citrix.com>> wrote:

Checked in my part into 4.2 and master branch.

Thanks
-min

From: Alena Prokharchyk 
mailto:alena.prokharc...@citrix.com>>
Date: Wednesday, July 17, 2013 1:26 PM
To: Harikrishna Patnala 
mailto:harikrishna.patn...@citrix.com>>, Min 
Chen mailto:min.c...@citrix.com>>, Rajesh Battala 
mailto:rajesh.batt...@citrix.com>>, Edison Su 
mailto:edison...@citrix.com>>, Sateesh Chodapuneedi 
mailto:sateesh.chodapune...@citrix.com>>, Hugo 
Trippaers 
mailto:htrippa...@schubergphilis.com>>, mice xia 
mailto:mice_...@tcloudcomputing.com>>
Cc: Animesh Chaturvedi 
mailto:animesh.chaturv...@citrix.com>>, Chandan 
Purushothama 
mailto:chandan.purushoth...@citrix.com>>, 
Sudha Ponnaganti 
mailto:sudha.ponnaga...@citrix.com>>, 
"dev@cloudstack.apache.org" 
mailto:dev@cloudstack.apache.org>>
Subject: 41-42 db upgrade

There is a bug in 41-42 db upgrade process – some global configs are missing in 
upgraded setup:

https://issues.apache.org/jira/browse/CLOUDSTACK-3601

Can please everyone from below list, insert them in as a part of 41-42 db 
upgrade sql script, as it is a part of your feature(s)?

  *   Harikrishna: "router.template.hyperv" and the rest of template related 
configuration
  *   Min: "s3.rrs.enabled"
  *   Rajesh: "healthcheck.update.interval"
  *   Edison: "snapshot.backup.rightafter", "storage.cache*'
  *   Sateesh: ""vmware.ports.per.dvportgroup", "vmware.use.dvswitch"
  *   Hugo: "vmware.nested.virtualization"
  *   Mice: "vmsnapshot.max", "vmsnapshot.create.wait"

Thanks!
-Alena.

Re: Review Request 12720: CLOUDSTACK: 3382 Unable to Migrate VM's If the hosts are implicitly or explicitly dedicated.

[Saksham Srivastava]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12720/
---

(Updated July 19, 2013, 11:52 a.m.)


Review request for cloudstack and Devdeep Singh.


Changes
---

Updated diff.


Bugs: 3382


Repository: cloudstack-git


Description
---

Allow root admin to migrate VMs across dedicated hosts.
Generate alerts for each migration performed between dedicated hosts.
Alerts are generated for VM migration between:
1) Source host is dedicated and destination host is not.
2) Source host is not dedicated and destination host is dedicated.
3) Both hosts are dedicated.
Alerts are now generated for both explicit and implicit dedication.


Diffs (updated)
-

  server/src/com/cloud/deploy/dao/PlannerHostReservationDao.java 69118f1 
  server/src/com/cloud/deploy/dao/PlannerHostReservationDaoImpl.java 41e0964 
  server/src/com/cloud/vm/UserVmManagerImpl.java 9968690 

Diff: https://reviews.apache.org/r/12720/diff/


Testing
---

Migrated VMs across explicitly dedicated hosts and alerts are generated.
Migrated VMs across implicitly dedicated hosts and alerts are generated.
 


Thanks,

Saksham Srivastava

Re: Review Request 12720: CLOUDSTACK: 3382 Unable to Migrate VM's If the hosts are implicitly or explicitly dedicated.

[Saksham Srivastava]

> On July 18, 2013, 11:09 a.m., Prasanna Santhanam wrote:
> > Do you think you can add a integration test to test_explicit_dedication.py 
> > for this failure? That would be really useful to catch this part of tricky 
> > code

Prasanna the changes I have made are specifically related to alert generation, 
the original functionality is kept unchanged.
Do you suggest to additional test that checks for alert generation ?


- Saksham


---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12720/#review23372
---


On July 19, 2013, 11:52 a.m., Saksham Srivastava wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12720/
> ---
> 
> (Updated July 19, 2013, 11:52 a.m.)
> 
> 
> Review request for cloudstack and Devdeep Singh.
> 
> 
> Bugs: 3382
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Allow root admin to migrate VMs across dedicated hosts.
> Generate alerts for each migration performed between dedicated hosts.
> Alerts are generated for VM migration between:
> 1) Source host is dedicated and destination host is not.
> 2) Source host is not dedicated and destination host is dedicated.
> 3) Both hosts are dedicated.
> Alerts are now generated for both explicit and implicit dedication.
> 
> 
> Diffs
> -
> 
>   server/src/com/cloud/deploy/dao/PlannerHostReservationDao.java 69118f1 
>   server/src/com/cloud/deploy/dao/PlannerHostReservationDaoImpl.java 41e0964 
>   server/src/com/cloud/vm/UserVmManagerImpl.java 9968690 
> 
> Diff: https://reviews.apache.org/r/12720/diff/
> 
> 
> Testing
> ---
> 
> Migrated VMs across explicitly dedicated hosts and alerts are generated.
> Migrated VMs across implicitly dedicated hosts and alerts are generated.
>  
> 
> 
> Thanks,
> 
> Saksham Srivastava
> 
>

Re: Demo Lab - To rent or To buy

[Ian Duffy]

@IIya - Thanks for the response.

You dont need a server, a nicely loaded desktop or even large laptop would
> do the job.


I've currently been trying to do this on a late 2011 macbook pro with i7
processor, 16gb of ram, OS on SSD. I was running VMware fusion with an
xenserver VM and a Cloudstack Manager VM. It was tight for ram but did work
and VMware fusion did pass the VT technology through to xenserver providing
HVM capabilities.

What is the budget like?


Somewhere between 1000 - 1200 eur. I looked at renting with hetzner but it
was coming out at being 500eur for setup costs and 300eur every month. I
have found two second hand dell servers with an Intel Xeon 2.6Ghz (4 dual
core processors with HT installed) and 32gb of ram(total of 83.2ghz and
64gb of ram) but storage for them is posing to be an issue. (noisy/loud,
electricity cost, internet connection etc.)

@Sebastien - That looks pretty interesting! Will take a better read of it
tonight.

On 19 July 2013 09:39, Sebastien Goasguen  wrote:

>
> On Jul 18, 2013, at 7:56 PM, Ian Duffy  wrote:
>
> > Hi,
> >
> > I'm looking for peoples personal opinions.
> >
> > I'm thinking about undertaking a project for a 2 month college assignment
> > sometime in the future to create an API wrapper and web front end with
> the
> > end goal of producing something like the rackspace or linode web-ui. I do
> > realise 'basic cloud' is a solved problem, but I'm not seeing any *open
> > source projects* that allow for user sign up, select a package offering,
> > billing, ticketing etc. along with basic instance/vm management.
> >
>
> It's kinda of nice, but I think you have proven that you can tackle a much
> more challenging problem than this one.
>
> tackle this instead:
>
> https://github.com/Netflix/Cloud-Prize/wiki
>
>
> > The cost of setting up a demo lab seems somewhat expensive.
> >
> > My idea was: 1 xenserver with 16gb+ ram make use of local storage, 1
> > cloudstack manager/secondary storage 2gb+ of ram, finally something to
> host
> > the wrapper and web app.
> >
> > What hardware would you recommend for creating a budget, portable and
> quiet
> > demo lab?
> > Do you think I would be better of renting? If so are their any companies
> > you would suggestion?
> >
>
>
>
>
> > Thanks,
> > Ian
>
>

Re: Demo Lab - To rent or To buy

[David Nalley]

So it's not ideal, particularly from a RAM perspective. But my
personal, local demo lab is using Mac Mini Servers. Mac Mini's are
small, quiet, don't draw a ton of power and can be repopulated with up
to 16GB of RAM. The UEFI is somewhat painful for installing things
that aren't OSX, but not impossible.

--David

On Fri, Jul 19, 2013 at 8:00 AM, Ian Duffy  wrote:
> @IIya - Thanks for the response.
>
> You dont need a server, a nicely loaded desktop or even large laptop would
>> do the job.
>
>
> I've currently been trying to do this on a late 2011 macbook pro with i7
> processor, 16gb of ram, OS on SSD. I was running VMware fusion with an
> xenserver VM and a Cloudstack Manager VM. It was tight for ram but did work
> and VMware fusion did pass the VT technology through to xenserver providing
> HVM capabilities.
>
> What is the budget like?
>
>
> Somewhere between 1000 - 1200 eur. I looked at renting with hetzner but it
> was coming out at being 500eur for setup costs and 300eur every month. I
> have found two second hand dell servers with an Intel Xeon 2.6Ghz (4 dual
> core processors with HT installed) and 32gb of ram(total of 83.2ghz and
> 64gb of ram) but storage for them is posing to be an issue. (noisy/loud,
> electricity cost, internet connection etc.)
>
> @Sebastien - That looks pretty interesting! Will take a better read of it
> tonight.
>
> On 19 July 2013 09:39, Sebastien Goasguen  wrote:
>
>>
>> On Jul 18, 2013, at 7:56 PM, Ian Duffy  wrote:
>>
>> > Hi,
>> >
>> > I'm looking for peoples personal opinions.
>> >
>> > I'm thinking about undertaking a project for a 2 month college assignment
>> > sometime in the future to create an API wrapper and web front end with
>> the
>> > end goal of producing something like the rackspace or linode web-ui. I do
>> > realise 'basic cloud' is a solved problem, but I'm not seeing any *open
>> > source projects* that allow for user sign up, select a package offering,
>> > billing, ticketing etc. along with basic instance/vm management.
>> >
>>
>> It's kinda of nice, but I think you have proven that you can tackle a much
>> more challenging problem than this one.
>>
>> tackle this instead:
>>
>> https://github.com/Netflix/Cloud-Prize/wiki
>>
>>
>> > The cost of setting up a demo lab seems somewhat expensive.
>> >
>> > My idea was: 1 xenserver with 16gb+ ram make use of local storage, 1
>> > cloudstack manager/secondary storage 2gb+ of ram, finally something to
>> host
>> > the wrapper and web app.
>> >
>> > What hardware would you recommend for creating a budget, portable and
>> quiet
>> > demo lab?
>> > Do you think I would be better of renting? If so are their any companies
>> > you would suggestion?
>> >
>>
>>
>>
>>
>> > Thanks,
>> > Ian
>>
>>

[discuss] add code style checker rules to master?

[Donal Lafferty]

For automated code style checking, I created a rules file for CheckStyle based 
on the ACS coding conventions, and blogged on how to use it with Eclipse ( 
http://dlafferty.blogspot.co.uk/2013/07/apache-cloudstack-java-coding.html ).

However, the rules are not 100% correct, and it is difficult for the community 
to correct them.

Would people be interested in adding it to master under 
./tools/eclipse/acs_checkstyle.xml ?

Review Request 12759: CLOUDSTACK-3669: Bringing down the dynamic scalable check (for VM whether it has xstools or vmware tools in it) to the management layer.

[Harikrishna Patnala]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12759/
---

Review request for cloudstack, Abhinandan Prateek and Nitin Mehta.


Bugs: CLOUDSTACK-3669


Repository: cloudstack-git


Description
---

CLOUDSTACK-3669:  Bringing down the dynamic scalable check (for VM whether it 
has xstools or vmware tools in it) to the management layer. 


Diffs
-

  core/src/com/cloud/agent/api/ScaleVmCommand.java 83cdcac 
  
plugins/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceBase.java
 f80d4b6 
  server/src/com/cloud/vm/UserVmManagerImpl.java adcf475 
  server/src/com/cloud/vm/VirtualMachineManagerImpl.java 7a4bf50 
  server/test/com/cloud/vm/VirtualMachineManagerImplTest.java 1e6d005 

Diff: https://reviews.apache.org/r/12759/diff/


Testing
---


Thanks,

Harikrishna Patnala

RE: Does the default Xen SystemVM Template understand S3?

[Donal Lafferty]

FYI:  the default XenServer System VM does not use the latest JSON 
serialisation design.  Must be some old code in there.

See 
https://issues.apache.org/jira/browse/CLOUDSTACK-3377?focusedCommentId=13713542&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13713542
 

> -Original Message-
> From: Chip Childers [mailto:chip.child...@sungard.com]
> Sent: 12 July 2013 8:51 PM
> To: dev@cloudstack.apache.org
> Cc: cloudstack-...@incubator.apache.org
> Subject: Re: Does the default Xen SystemVM Template understand S3?
> 
> AFAIK, yes.  That code is loaded when the agent software is updated by the
> mgmt server.
> 
> On Fri, Jul 12, 2013 at 05:02:12PM +, Donal Lafferty wrote:
> > AFAIK, the database deploy script in Master sets the Xen SystemVM URL
> > to
> > http://download.cloud.com/templates/acton/acton-systemvm-
> 02062012.vhd.
> > bz2
> >
> > Is this SystemVM image aware of S3 object store?
> >
> > If now, can someone recommend a Xen image that both works and works
> with S3?
> >
> >
> > DL
> >
> >

Re: Does the default Xen SystemVM Template understand S3?

[Prasanna Santhanam]

See CLOUDSTACK-3480. if using object store the templates are going to
be wrong since the storage code will get the templates as in the
template table. There is no "seeding" the template when using the
object store feature. Is this related to what you are seeing?

On Fri, Jul 19, 2013 at 12:17:25PM +, Donal Lafferty wrote:
> FYI:  the default XenServer System VM does not use the latest JSON 
> serialisation design.  Must be some old code in there.
> 
> See 
> https://issues.apache.org/jira/browse/CLOUDSTACK-3377?focusedCommentId=13713542&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13713542
>  
> 
> > -Original Message-
> > From: Chip Childers [mailto:chip.child...@sungard.com]
> > Sent: 12 July 2013 8:51 PM
> > To: dev@cloudstack.apache.org
> > Cc: cloudstack-...@incubator.apache.org
> > Subject: Re: Does the default Xen SystemVM Template understand S3?
> > 
> > AFAIK, yes.  That code is loaded when the agent software is updated by the
> > mgmt server.
> > 
> > On Fri, Jul 12, 2013 at 05:02:12PM +, Donal Lafferty wrote:
> > > AFAIK, the database deploy script in Master sets the Xen SystemVM URL
> > > to
> > > http://download.cloud.com/templates/acton/acton-systemvm-
> > 02062012.vhd.
> > > bz2
> > >
> > > Is this SystemVM image aware of S3 object store?
> > >
> > > If now, can someone recommend a Xen image that both works and works
> > with S3?
> > >
> > >
> > > DL
> > >
> > >

-- 
Prasanna.,


Powered by BigRock.com

Re: [rant] stupid test cases

[Prasanna Santhanam]

My problem is that we have tests that already check cleanup of
accounts. A test should do the most crystallized set of steps to
achieve the scenario and not try to put everything and the kitchen
sink in to it. If we see ourselves doing that - we need to break down
our tests into smaller blocks. They'll still be
system/integration/live tests only.

As to the simulator - yes you can run these tests on a simulator
today.

On Fri, Jul 19, 2013 at 01:12:14AM +, Alex Huang wrote:
> I disagree.  Error handling should be part of our testing.
> 
> We should incorporate the simulator into the BVT and regression
> tests.  On testcases that really is to test the business logic
> rather than the provisioning code, the test case should perform all
> of the provisioning on the simulator instead.  Then simulator can be
> programmed to simulate VM stopped failure etc and see how the
> business responds to these problems.
> 
> --Alex
> 
> > -Original Message-
> > From: Anthony Xu [mailto:xuefei...@citrix.com]
> > Sent: Thursday, July 18, 2013 3:02 PM
> > To: dev@cloudstack.apache.org
> > Subject: RE: [rant] stupid test cases
> > 
> > +1   VM can be in "Stopped" state
> > 
> > 
> > Anthony
> > 
> > -Original Message-
> > From: Marcus Sorensen [mailto:shadow...@gmail.com]
> > Sent: Wednesday, July 17, 2013 10:47 PM
> > To: dev@cloudstack.apache.org
> > Subject: Re: [rant] stupid test cases
> > 
> > I can understand that we may want to test that everything related to the
> > domain gets cleaned up properly. We have run into all sorts of things when
> > deleting accounts, for example where resources won't clean up because the
> > account is gone and we throw null pointers because a bunch of code looks up
> > account when deleting. However, to your point, VMs can be created in a
> > "stopped" state and that wouldn't incur the overhead of deployment.
> > On Jul 17, 2013 11:33 PM, "Prasanna Santhanam"  wrote:
> > 
> > > I was just going through one of the automated test cases and I find it
> > > really silly that there's the following test:
> > >
> > > def test_forceDeleteDomain(self):
> > > """ Test delete domain force option"""
> > >
> > > # Steps for validations
> > > # 1. create a domain DOM
> > > # 2. create 2 users under this domain
> > > # 3. deploy 1 VM into each of these user accounts
> > > # 4. create PF / FW rules for port 22 on these VMs for their
> > > #respective accounts
> > > # 5. delete the domain with force=true option
> > > # Validate the following
> > > # 1. listDomains should list the created domain
> > > # 2. listAccounts should list the created accounts
> > > # 3. listvirtualmachines should show the Running VMs
> > > # 4. PF and FW rules should be shown in listFirewallRules
> > > # 5. domain should delete successfully and above three list calls
> > > #should show all the resources now deleted. listRouters should
> > > #not return any routers in the deleted accounts/domains
> > >
> > > Why would one need the overhead of creating VMs in a domain deletion
> > test?
> > > Do
> > > we not understand that the basics accounts/domains/ etc that
> > > cloudstack has nothing to do with the virtual machines? This kind of a
> > > test slows down other useful tests that we could be running. More over
> > > when this fails in the VM creation step I'd have to go in and analyse
> > > logs to realize that deletedomain was perhaps fine but vm creation has
> > > failed.
> > >
> > > That's a pointless effort. I'm sure there are others in the automated
> > > tests that do this kind of wasteful testing. So please please pleaes
> > > &*()#@()# please review test plans before automating them!
> > >
> > > I'm not going to be looking at this forever to fix these issues when
> > > we want to see pretty metrics and numbers.
> > >
> > > --
> > > Prasanna.,
> > >
> > > 
> > > Powered by BigRock.com
> > >
> > >

-- 
Prasanna.,


Powered by BigRock.com

Re: Review Request 12758: Cloustack - 3452: Added check if security groups feature is enabled in basic zone before creating any sec group

[Gaurav Aradhye]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12758/
---

(Updated July 19, 2013, 1:07 p.m.)


Review request for cloudstack and Prasanna Santhanam.


Summary (updated)
-

Cloustack - 3452: Added check if security groups feature is enabled in basic 
zone before creating any sec group


Repository: cloudstack-git


Description (updated)
---

Resolved issue CLOUDSTACK - 3452: 
https://issues.apache.org/jira/browse/CLOUDSTACK-3452

Addded check if security groups feature is enabled for basic zone before 
creating security group while creating VM.


Diffs (updated)
-

  tools/marvin/marvin/integration/lib/base.py 13507bc 

Diff: https://reviews.apache.org/r/12758/diff/


Testing (updated)
---


Thanks,

Gaurav Aradhye

[DOCS] Known issues in release notes

[Chip Childers]

Hi all,

I've been working on the release notes for 4.1.1, and when I got to the
known issues section I took a look at past releases.

For the 4.0.x series, we used a query that checked for unresolved bugs
that had a fix version of the release that the notes were for.  This
doesn't seem quite right, since in reality, the know errors are all open
bugs that aren't limited to *future* releases.

While it's unfortunate to have such a long list, isn't a query like:

project = CLOUDSTACK AND resolution = Unresolved and type = Bug AND
affectedVersion != '4.2.0' ORDER BY priority DESC

really the best query to represent the know issues with 4.1.1?  And
actually, there is a problem with that query.  Since affectedVersion is
actually a list field, I'm still trying to figure out how to exclude
only the bugs that are ONLY affecting 4.2.0.

Thoughts?

Re: [DOCS] Known issues in release notes

[Chip Childers]

On Fri, Jul 19, 2013 at 09:09:26AM -0400, Chip Childers wrote:
> Hi all,
> 
> I've been working on the release notes for 4.1.1, and when I got to the
> known issues section I took a look at past releases.
> 
> For the 4.0.x series, we used a query that checked for unresolved bugs
> that had a fix version of the release that the notes were for.  This
> doesn't seem quite right, since in reality, the know errors are all open
> bugs that aren't limited to *future* releases.
> 
> While it's unfortunate to have such a long list, isn't a query like:
> 
> project = CLOUDSTACK AND resolution = Unresolved and type = Bug AND
> affectedVersion != '4.2.0' ORDER BY priority DESC
> 
> really the best query to represent the know issues with 4.1.1?  And
> actually, there is a problem with that query.  Since affectedVersion is
> actually a list field, I'm still trying to figure out how to exclude
> only the bugs that are ONLY affecting 4.2.0.
> 
> Thoughts?

Joe and I discussed this on IRC.

Here's the query that we're suggesting:

project = CLOUDSTACK AND resolution = Unresolved AND type = Bug AND
affectedVersion in releasedVersions() ORDER BY priority DESC

I'll use that for 4.1.1 unless anyone objects.

Re: code formatting for enums

[John Burwell]

Frank,

I expect that renaming an enum value to be a minor change (i.e. once the 
codebase compiles, the change is good).  However, i can say I am surprised to 
learn that such an operation could break the API and/or corrupt that database.  
To my mind, this issue has nothing to do with developer choice.  Instead, our 
architecture needs to recognize that an enumerated value may have multiple 
representations dependent on the context (e.g. CloudStack HTTP API, AWS API, 
UI, code, database, etc), and provide mechanisms to perform transformations 
between them when they need to differ.  From a persistence perspective, we need 
to should be keying off of a value that is unlikely to change (e.g. an 
artificial ID).  I would submit that Java enumeration value names, as code 
symbols, are subject to an unacceptable level of change to be used for 
persistence.

Thanks,
-John

On Jul 18, 2013, at 8:49 PM, Frank Zhang  wrote:

> Yes it's brittle.  I assume "law of least astonishment" here means reducing 
> learning curve of 
> developers and the changes they made should not surprise user. To achieve 
> this the code must
> provide very flexibility to developer. However, I am thinking of such kind of 
> flexibility is really needed, isn't it?
> Currently "Convention over configuration" which sacrifices flexibility for 
> simplicity is getting more and more
> popular. If we can state clearly to developers on constructing data structure 
> from database tables to API responses, we are 
> achieving great simplicity and overall coding convention.
> 
> Sometimes, it's better off offering only choice to developer and don't make 
> them think. Anyway, a [DISCUSS] thread is necessary.
> 
>> -Original Message-
>> From: John Burwell [mailto:jburw...@basho.com]
>> Sent: Thursday, July 18, 2013 4:39 PM
>> To: dev@cloudstack.apache.org
>> Subject: Re: code formatting for enums
>> 
>> Alex and Frank,
>> 
>> In terms of conventions, the APIs exposed Java SDK and many other common
>> APIs following this convention.  Hence, the reason for my suggestion.
>> 
>> The notion that changing an enum key can break the API and/or database
>> persistence feels a bit brittle.  In particular, it doesn't conform to the 
>> Law of
>> Least Surprise.  It feels like we should have a transformation mechanism from
>> the API endpoint to an enum value, and employ foreign, artificial keys to 
>> code
>> tables in schema.  This conversation has expanded a bit, and it seems
>> appropriate to open a new [DISCUSS] thread to delve into it further.  Do 
>> y'all
>> agree?
>> 
>> Thanks,
>> -John
>> 
>> On Jul 18, 2013, at 7:19 PM, Frank Zhang  wrote:
>> 
>>> Those enums cannot be simply considered as internal data structures where
>> code convention applies to, they should be considered in API level.
>>> 
>>> Most CloudStack API responses retrieve its fields from xxxVO classes
>>> which represent database table. In our example, enum State directly
>>> maps to UserVmResponse. state. Then the most important factor of name
>> convention is user experience where Running is more user friendly than
>> IS_RUNNING or whatever all capitalized sentence splitting by underscore.
>>> 
>>> And any changes to those existing enum should be thought twice, as it 
>>> relates
>> to API compatibility.
>>> 
>>> Though we can introduce some mapping layer between internal enum and
>>> API response, I don't see any benefits if the only reason is to follow some
>> name convention.
>>> 
 -Original Message-
 From: Alex Huang [mailto:alex.hu...@citrix.com]
 Sent: Thursday, July 18, 2013 3:55 PM
 To: dev@cloudstack.apache.org
 Subject: RE: code formatting for enums
 
 Actually, that's more of a C/C++ coding convention.  (Speaking of
 which, please don't use "I" to start interfaces.)
 
 I prefer to have enums as follows
 
 Public class Vm {
 enum State {
 IsRunning,
 Stopped,
 }
 }
 
 I generally like to write Vm.State.IsRunning  in the code.  It's readable 
 and
>> clear.
 
 As opposed to Vm.State.IS_RUNNING which is a little less readable.
 
 But the thing I've seen people do is just using IS_RUNNING or
 State.IsRunning which often becomes confusing.  I'm more against that
 then all caps and underscore.
 
 My $.02.  I will caution that any change to existing enums, we have
 to think about how it maps to the database.  If the VO object stores
 the enum, you'll have to either upgrade the database or add methods
 to the enum so that when storing it, it becomes the same.
 
 --Alex
 
> -Original Message-
> From: John Burwell [mailto:jburw...@basho.com]
> Sent: Thursday, July 18, 2013 12:33 PM
> To: dev@cloudstack.apache.org
> Subject: Re: code formatting for enums
> 
> All,
> 
> Another thing I have noticed is that enum values are not capitalized.
> General coding convention is that enum values

Re: Demo Lab - To rent or To buy

[Kelly Hair]

@Ian - which version of Fusion were you using?

According to
http://www.virtualizationadmin.com/blogs/lowe/news/vmware-fusion-5-enable-vt-xept-inside-a-virtual-machine-288.html,
it is supported on version 5.

I had a similar problem in Workstation when running nested ESXi VMs until I
upgraded.  Mine was due to Ivy Bridge incompatibilities so it was slightly
different but with asking...
On Jul 19, 2013 8:07 AM, "David Nalley"  wrote:

> So it's not ideal, particularly from a RAM perspective. But my
> personal, local demo lab is using Mac Mini Servers. Mac Mini's are
> small, quiet, don't draw a ton of power and can be repopulated with up
> to 16GB of RAM. The UEFI is somewhat painful for installing things
> that aren't OSX, but not impossible.
>
> --David
>
> On Fri, Jul 19, 2013 at 8:00 AM, Ian Duffy  wrote:
> > @IIya - Thanks for the response.
> >
> > You dont need a server, a nicely loaded desktop or even large laptop
> would
> >> do the job.
> >
> >
> > I've currently been trying to do this on a late 2011 macbook pro with i7
> > processor, 16gb of ram, OS on SSD. I was running VMware fusion with an
> > xenserver VM and a Cloudstack Manager VM. It was tight for ram but did
> work
> > and VMware fusion did pass the VT technology through to xenserver
> providing
> > HVM capabilities.
> >
> > What is the budget like?
> >
> >
> > Somewhere between 1000 - 1200 eur. I looked at renting with hetzner but
> it
> > was coming out at being 500eur for setup costs and 300eur every month. I
> > have found two second hand dell servers with an Intel Xeon 2.6Ghz (4 dual
> > core processors with HT installed) and 32gb of ram(total of 83.2ghz and
> > 64gb of ram) but storage for them is posing to be an issue. (noisy/loud,
> > electricity cost, internet connection etc.)
> >
> > @Sebastien - That looks pretty interesting! Will take a better read of it
> > tonight.
> >
> > On 19 July 2013 09:39, Sebastien Goasguen  wrote:
> >
> >>
> >> On Jul 18, 2013, at 7:56 PM, Ian Duffy  wrote:
> >>
> >> > Hi,
> >> >
> >> > I'm looking for peoples personal opinions.
> >> >
> >> > I'm thinking about undertaking a project for a 2 month college
> assignment
> >> > sometime in the future to create an API wrapper and web front end with
> >> the
> >> > end goal of producing something like the rackspace or linode web-ui.
> I do
> >> > realise 'basic cloud' is a solved problem, but I'm not seeing any
> *open
> >> > source projects* that allow for user sign up, select a package
> offering,
> >> > billing, ticketing etc. along with basic instance/vm management.
> >> >
> >>
> >> It's kinda of nice, but I think you have proven that you can tackle a
> much
> >> more challenging problem than this one.
> >>
> >> tackle this instead:
> >>
> >> https://github.com/Netflix/Cloud-Prize/wiki
> >>
> >>
> >> > The cost of setting up a demo lab seems somewhat expensive.
> >> >
> >> > My idea was: 1 xenserver with 16gb+ ram make use of local storage, 1
> >> > cloudstack manager/secondary storage 2gb+ of ram, finally something to
> >> host
> >> > the wrapper and web app.
> >> >
> >> > What hardware would you recommend for creating a budget, portable and
> >> quiet
> >> > demo lab?
> >> > Do you think I would be better of renting? If so are their any
> companies
> >> > you would suggestion?
> >> >
> >>
> >>
> >>
> >>
> >> > Thanks,
> >> > Ian
> >>
> >>
>

Re: Review Request 12752: Async commands can inject the job id

[Marcus Sorensen]

This is maybe a tangent, but does cloudstack check for uniqueness when
generating a UUID? I haven't seen it anywhere, I've always seen it
do UUID.randomUUID() and no checking. However, I guess the point is that CS
is in control of making sure it's randomly generated, with this patch CS
could be fed duplicate UUIDs by a misbehaving program.
On Jul 19, 2013 3:23 AM, "Murali Reddy"  wrote:

> Ryan,
>
> You should open a bug and add details on what the use case you are trying
> to solve. Since CloudStack is not in control of UUID generation, you
> should at least check that uniqueness of passed in UUID.
>
> Also please introduce a new smoke test for 'injected job id' instead of
> piggy backing the test for deploy VM.
>
> On 19/07/13 6:02 AM, "Ryan Dietrich"  wrote:
>
> >
> >
> >> On July 18, 2013, 11:34 p.m., Marcus Sorensen wrote:
> >> > server/src/com/cloud/async/AsyncJobVO.java, line 141
> >> >
> >><
> https://reviews.apache.org/r/12752/diff/1/?file=323558#file323558line141
> >>>
> >> >
> >> > I'm not immediately certain why this was pulled out.
> >
> >Nothing is using it.  I did a scan to see if anyone is using it, and I
> >didn't see anything.  Dead could should be removed, right?  Git will
> >remember for us if we need it later.
> >
> >
> >- Ryan
> >
> >
> >---
> >This is an automatically generated e-mail. To reply, visit:
> >https://reviews.apache.org/r/12752/#review23457
> >---
> >
> >
> >On July 18, 2013, 11:28 p.m., Ryan Dietrich wrote:
> >>
> >> ---
> >> This is an automatically generated e-mail. To reply, visit:
> >> https://reviews.apache.org/r/12752/
> >> ---
> >>
> >> (Updated July 18, 2013, 11:28 p.m.)
> >>
> >>
> >> Review request for cloudstack and Marcus Sorensen.
> >>
> >>
> >> Repository: cloudstack-git
> >>
> >>
> >> Description
> >> ---
> >>
> >> I added "injectedjobid" to the BaseAsyncCmd class as a parameter.
> >> If set, it will allow you to tell Cloudstack what the job id instead of
> >>it choosing one.
> >> A basic string length test is done to verify the variable passed in is
> >>actually a UUID.
> >> If it is not valid, it is ignored and the job generates it's own.
> >>
> >>
> >> Diffs
> >> -
> >>
> >>   api/src/org/apache/cloudstack/api/BaseAsyncCmd.java 0e6f95d
> >>   server/src/com/cloud/api/ApiServer.java 95f17af
> >>   server/src/com/cloud/async/AsyncJobVO.java 41eccb4
> >>   test/integration/smoke/test_deploy_vm.py 425aeb7
> >>   tools/marvin/marvin/codegenerator.py 632b8c6
> >>   tools/marvin/marvin/integration/lib/base.py 161d03c
> >>
> >> Diff: https://reviews.apache.org/r/12752/diff/
> >>
> >>
> >> Testing
> >> ---
> >>
> >> Updated marvin, updated the deploy vm test.  Ran multiple async
> >>commands manually, with and without injectedjobid present, no issues
> >>detected.
> >>
> >>
> >> Thanks,
> >>
> >> Ryan Dietrich
> >>
> >>
> >
> >
>
>
>

Re: Review Request 11582: CLOUDSTACK-2053: fixed if/else logic to address redundant entires, fixed boto/cloud-init support by adding metadata rewrite rule

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/11582/#review23513
---


Commit 8f4b9bccfed63a37762907bdd058506f4e7b6e6d in branch refs/heads/4.0 from 
ilya musayev
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=8f4b9bc ]

CLOUDSTACK-2053: Fixed userdata.sh redundancy and cloud-init support


- ASF Subversion and Git Services


On June 12, 2013, 11:54 p.m., ilya musayev wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/11582/
> ---
> 
> (Updated June 12, 2013, 11:54 p.m.)
> 
> 
> Review request for cloudstack, Chip Childers and Chiradeep Vittal.
> 
> 
> Bugs: CLOUDSTACK-2053
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Resubmitting for the 3rd time, as review board crashed on my submissions 
> previously
> 
> Rewrote the bash if/else check to fix redundant entries
> Added RewriteRules to support boto/cloud-init - as per Chiradeep - this would 
> only affect
> VMWare deployments, as this userdata.sh script is only used with vmware 
> system router vms
> 
> 
> Diffs
> -
> 
>   patches/systemvm/debian/config/root/userdata.sh e5d170b 
> 
> Diff: https://reviews.apache.org/r/11582/diff/
> 
> 
> Testing
> ---
> 
> 4.1 with vSphere
> 
> 
> Thanks,
> 
> ilya musayev
> 
>

Re: Review Request 12685: CLOUDSTACK-1532 added utility functions to scan URIs

[Alex Huang]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12685/#review23515
---

Ship it!


master: 2d4464d

One review comment is if you are checking for the colon to see if the value has 
the schema already, then shouldn't it also check if the schema matches what's 
declared?  But that shouldn't block this review.  The code itself is 
technically sound.

- Alex Huang


On July 17, 2013, 3:46 p.m., daan Hoogland wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12685/
> ---
> 
> (Updated July 17, 2013, 3:46 p.m.)
> 
> 
> Review request for cloudstack and Hugo Trippaers.
> 
> 
> Bugs: CLOUDSTACK-1532
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> the review for the complete patch was open to long and no longer applicable, 
> so I am starting with smaller patches as this is really needed to implement 
> CLOUDSTACK-1532
> 
> 
> Diffs
> -
> 
>   api/src/com/cloud/network/Networks.java 5aede05 
>   api/test/com/cloud/network/NetworksTest.java PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/12685/diff/
> 
> 
> Testing
> ---
> 
> unit testing
> 
> 
> Thanks,
> 
> daan Hoogland
> 
>

RE: Does the default Xen SystemVM Template understand S3?

[Donal Lafferty]

Followed "Maven procedure developed for Cloudstack 4.1.0 and later" on 
http://cloudstack.apache.org/develop/environment.html 

I'll try seeding as per  "CloudStack Installation Guide" 
(https://cloudstack.apache.org/docs/en-US/Apache_CloudStack/4.0.2/html/Installation_Guide/management-server-install-flow.html#prepare-system-vm-template).

Here's hoping that will patch it!

DL


> -Original Message-
> From: Prasanna Santhanam [mailto:t...@apache.org]
> Sent: 19 July 2013 1:22 PM
> To: dev@cloudstack.apache.org
> Subject: Re: Does the default Xen SystemVM Template understand S3?
> 
> See CLOUDSTACK-3480. if using object store the templates are going to be
> wrong since the storage code will get the templates as in the template table.
> There is no "seeding" the template when using the object store feature. Is
> this related to what you are seeing?
> 
> On Fri, Jul 19, 2013 at 12:17:25PM +, Donal Lafferty wrote:
> > FYI:  the default XenServer System VM does not use the latest JSON
> serialisation design.  Must be some old code in there.
> >
> > See
> > https://issues.apache.org/jira/browse/CLOUDSTACK-
> 3377?focusedCommentId
> >
> =13713542&page=com.atlassian.jira.plugin.system.issuetabpanels:comment
> > -tabpanel#comment-13713542
> >
> > > -Original Message-
> > > From: Chip Childers [mailto:chip.child...@sungard.com]
> > > Sent: 12 July 2013 8:51 PM
> > > To: dev@cloudstack.apache.org
> > > Cc: cloudstack-...@incubator.apache.org
> > > Subject: Re: Does the default Xen SystemVM Template understand S3?
> > >
> > > AFAIK, yes.  That code is loaded when the agent software is updated
> > > by the mgmt server.
> > >
> > > On Fri, Jul 12, 2013 at 05:02:12PM +, Donal Lafferty wrote:
> > > > AFAIK, the database deploy script in Master sets the Xen SystemVM
> > > > URL to
> > > > http://download.cloud.com/templates/acton/acton-systemvm-
> > > 02062012.vhd.
> > > > bz2
> > > >
> > > > Is this SystemVM image aware of S3 object store?
> > > >
> > > > If now, can someone recommend a Xen image that both works and
> > > > works
> > > with S3?
> > > >
> > > >
> > > > DL
> > > >
> > > >
> 
> --
> Prasanna.,
> 
> 
> Powered by BigRock.com

[DISCUSS] Bylaw changes for new committer / new PMC member votes

[Chip Childers]

As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
know we didn't define that term correctly as of now, but I think that's
a different discussion).  We currently have that term defined as:

> Lazy Consensus - Lazy consensus requires 3 binding +1 votes and no
> binding -1 votes.

I'd like to propose that we change the PMC and committer voting rule to
use the Lazy 2/3 Majority approach defined as: 

> Lazy 2/3 majority votes requires at least 3 binding votes and twice as 
> many binding +1 votes as binding -1 votes.

Are there any objections to me starting a VOTE on this change?

Re: Review Request 11582: CLOUDSTACK-2053: fixed if/else logic to address redundant entires, fixed boto/cloud-init support by adding metadata rewrite rule

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/11582/#review23511
---


Commit 88b36edf87140692861f00da05f3fa8955340c80 in branch refs/heads/4.2 from 
ilya musayev
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=88b36ed ]

CLOUDSTACK-2053: Fixed userdata.sh redundancy and cloud-init support


- ASF Subversion and Git Services


On June 12, 2013, 11:54 p.m., ilya musayev wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/11582/
> ---
> 
> (Updated June 12, 2013, 11:54 p.m.)
> 
> 
> Review request for cloudstack, Chip Childers and Chiradeep Vittal.
> 
> 
> Bugs: CLOUDSTACK-2053
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Resubmitting for the 3rd time, as review board crashed on my submissions 
> previously
> 
> Rewrote the bash if/else check to fix redundant entries
> Added RewriteRules to support boto/cloud-init - as per Chiradeep - this would 
> only affect
> VMWare deployments, as this userdata.sh script is only used with vmware 
> system router vms
> 
> 
> Diffs
> -
> 
>   patches/systemvm/debian/config/root/userdata.sh e5d170b 
> 
> Diff: https://reviews.apache.org/r/11582/diff/
> 
> 
> Testing
> ---
> 
> 4.1 with vSphere
> 
> 
> Thanks,
> 
> ilya musayev
> 
>

RE: code formatting for enums

[Alex Huang]

John,

Just to be clear.  Not doing enum change correctly can cause problems with 
existing databases.  That's a known problem with enum persistence in Java 
because enums by default using the name of the constant to represent the string 
of the constant.  I was just cautioning anyone who wants to take that up to 
make sure they're doing the right thing.  It doesn't require some huge (or even 
minor) framework/architectural change in CloudStack in order for enum changes 
to work with db.

--Alex

> -Original Message-
> From: John Burwell [mailto:jburw...@basho.com]
> Sent: Friday, July 19, 2013 6:33 AM
> To: dev@cloudstack.apache.org
> Subject: Re: code formatting for enums
> 
> Frank,
> 
> I expect that renaming an enum value to be a minor change (i.e. once the
> codebase compiles, the change is good).  However, i can say I am surprised to
> learn that such an operation could break the API and/or corrupt that
> database.  To my mind, this issue has nothing to do with developer choice.
> Instead, our architecture needs to recognize that an enumerated value may
> have multiple representations dependent on the context (e.g. CloudStack
> HTTP API, AWS API, UI, code, database, etc), and provide mechanisms to
> perform transformations between them when they need to differ.  From a
> persistence perspective, we need to should be keying off of a value that is
> unlikely to change (e.g. an artificial ID).  I would submit that Java 
> enumeration
> value names, as code symbols, are subject to an unacceptable level of
> change to be used for persistence.
> 
> Thanks,
> -John
> 
> On Jul 18, 2013, at 8:49 PM, Frank Zhang  wrote:
> 
> > Yes it's brittle.  I assume "law of least astonishment" here means
> > reducing learning curve of developers and the changes they made should
> > not surprise user. To achieve this the code must provide very flexibility to
> developer. However, I am thinking of such kind of flexibility is really 
> needed,
> isn't it?
> > Currently "Convention over configuration" which sacrifices flexibility
> > for simplicity is getting more and more popular. If we can state
> > clearly to developers on constructing data structure from database tables
> to API responses, we are achieving great simplicity and overall coding
> convention.
> >
> > Sometimes, it's better off offering only choice to developer and don't make
> them think. Anyway, a [DISCUSS] thread is necessary.
> >
> >> -Original Message-
> >> From: John Burwell [mailto:jburw...@basho.com]
> >> Sent: Thursday, July 18, 2013 4:39 PM
> >> To: dev@cloudstack.apache.org
> >> Subject: Re: code formatting for enums
> >>
> >> Alex and Frank,
> >>
> >> In terms of conventions, the APIs exposed Java SDK and many other
> >> common APIs following this convention.  Hence, the reason for my
> suggestion.
> >>
> >> The notion that changing an enum key can break the API and/or
> >> database persistence feels a bit brittle.  In particular, it doesn't
> >> conform to the Law of Least Surprise.  It feels like we should have a
> >> transformation mechanism from the API endpoint to an enum value, and
> >> employ foreign, artificial keys to code tables in schema.  This
> >> conversation has expanded a bit, and it seems appropriate to open a
> >> new [DISCUSS] thread to delve into it further.  Do y'all agree?
> >>
> >> Thanks,
> >> -John
> >>
> >> On Jul 18, 2013, at 7:19 PM, Frank Zhang  wrote:
> >>
> >>> Those enums cannot be simply considered as internal data structures
> >>> where
> >> code convention applies to, they should be considered in API level.
> >>>
> >>> Most CloudStack API responses retrieve its fields from xxxVO classes
> >>> which represent database table. In our example, enum State directly
> >>> maps to UserVmResponse. state. Then the most important factor of
> >>> name
> >> convention is user experience where Running is more user friendly
> >> than IS_RUNNING or whatever all capitalized sentence splitting by
> underscore.
> >>>
> >>> And any changes to those existing enum should be thought twice, as
> >>> it relates
> >> to API compatibility.
> >>>
> >>> Though we can introduce some mapping layer between internal enum
> and
> >>> API response, I don't see any benefits if the only reason is to
> >>> follow some
> >> name convention.
> >>>
>  -Original Message-
>  From: Alex Huang [mailto:alex.hu...@citrix.com]
>  Sent: Thursday, July 18, 2013 3:55 PM
>  To: dev@cloudstack.apache.org
>  Subject: RE: code formatting for enums
> 
>  Actually, that's more of a C/C++ coding convention.  (Speaking of
>  which, please don't use "I" to start interfaces.)
> 
>  I prefer to have enums as follows
> 
>  Public class Vm {
>  enum State {
>  IsRunning,
>  Stopped,
>  }
>  }
> 
>  I generally like to write Vm.State.IsRunning  in the code.  It's
>  readable and
> >> clear.
> 
>  As opposed to Vm.State.IS_RUNNING which is a little less readable.
> >

Re: Demo Lab - To rent or To buy

[Joe Brockmeier]

On Thu, Jul 18, 2013, at 08:27 PM, Musayev, Ilya wrote:
> Specs would really depend on budget, in US I can build this setup for
> about $800, more or less depends on the componets.

One of the things I've usually done is go to Geeks.com and buy a
second-hand workstation. Typically you can pick up a pretty beefy
machine for cheap. For example, they have a quad-core Xeon machine with
12GB of RAM for ~$350. (The inventory varies quite a bit, though. If you
don't see something you like, check back in a few days. If you do see
something you like, it might be gone in a few days if you wait to make a
decision...)

Best,

jzb
-- 
Joe Brockmeier
j...@zonker.net
Twitter: @jzb
http://www.dissociatedpress.net/

RE: New user - Introduction

[Alex Huang]

Welcome Rene!

--Alex

> -Original Message-
> From: René Diepstraten [mailto:r...@renediepstraten.nl]
> Sent: Friday, July 19, 2013 1:19 AM
> To: dev@cloudstack.apache.org
> Subject: New user - Introduction
> 
> Hi everyone.
> 
> Since I have joined the mailing list, I'd like to say hello and introduce 
> myself.
> My name is Rene Diepstraten, am a system administrator and the company I
> work for ( PCextreme ) uses Apache CloudStack.
> 
> I'd like to contribute here and there with my skills in python and bash, and
> perhaps learn some java.
> A few small patches that I wrote have been accepted into Master, and I'd like
> to contribute more to this great project in the future.
> 
> Best regards,
> 
> Rene Diepstraten

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[Chiradeep Vittal]

There's several places in the by laws that call for Lazy Consensus. Are we
discussing modifying all of them or just new committer votes?

On 7/19/13 9:02 PM, "Chip Childers"  wrote:

>As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
>know we didn't define that term correctly as of now, but I think that's
>a different discussion).  We currently have that term defined as:
>
>> Lazy Consensus - Lazy consensus requires 3 binding +1 votes and no
>> binding -1 votes.
>
>I'd like to propose that we change the PMC and committer voting rule to
>use the Lazy 2/3 Majority approach defined as:
>
>> Lazy 2/3 majority votes requires at least 3 binding votes and twice as
>> many binding +1 votes as binding -1 votes.
>
>Are there any objections to me starting a VOTE on this change?

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[Chip Childers]

On Fri, Jul 19, 2013 at 04:29:07PM +, Chiradeep Vittal wrote:
> There's several places in the by laws that call for Lazy Consensus. Are we
> discussing modifying all of them or just new committer votes?

New committer and PMC membership.

sorry, I think the email could be more clear.  This is per the $subject:
new committer / new PMC member votes only.

> 
> On 7/19/13 9:02 PM, "Chip Childers"  wrote:
> 
> >As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
> >know we didn't define that term correctly as of now, but I think that's
> >a different discussion).  We currently have that term defined as:
> >
> >> Lazy Consensus - Lazy consensus requires 3 binding +1 votes and no
> >> binding -1 votes.
> >
> >I'd like to propose that we change the PMC and committer voting rule to
> >use the Lazy 2/3 Majority approach defined as:
> >
> >> Lazy 2/3 majority votes requires at least 3 binding votes and twice as
> >> many binding +1 votes as binding -1 votes.
> >
> >Are there any objections to me starting a VOTE on this change?
> 
> 

Re: New user - Introduction

[Joe Brockmeier]

On Fri, Jul 19, 2013, at 03:19 AM, René Diepstraten wrote:
> A few small patches that I wrote have been accepted into Master, and I'd
> like to contribute more to this great project in the future.

Welcome René! 

Best,

jzb
-- 
Joe Brockmeier
j...@zonker.net
Twitter: @jzb
http://www.dissociatedpress.net/

Re: Review Request 11582: CLOUDSTACK-2053: fixed if/else logic to address redundant entires, fixed boto/cloud-init support by adding metadata rewrite rule

[ASF Subversion and Git Services]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/11582/#review23507
---


Commit eb3ffef95ae493df94df97639fdf8c1a4cd8410f in branch refs/heads/master 
from ilya musayev
[ https://git-wip-us.apache.org/repos/asf?p=cloudstack.git;h=eb3ffef ]

CLOUDSTACK-2053: Fixed userdata.sh redundancy and cloud-init support


- ASF Subversion and Git Services


On June 12, 2013, 11:54 p.m., ilya musayev wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/11582/
> ---
> 
> (Updated June 12, 2013, 11:54 p.m.)
> 
> 
> Review request for cloudstack, Chip Childers and Chiradeep Vittal.
> 
> 
> Bugs: CLOUDSTACK-2053
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> Resubmitting for the 3rd time, as review board crashed on my submissions 
> previously
> 
> Rewrote the bash if/else check to fix redundant entries
> Added RewriteRules to support boto/cloud-init - as per Chiradeep - this would 
> only affect
> VMWare deployments, as this userdata.sh script is only used with vmware 
> system router vms
> 
> 
> Diffs
> -
> 
>   patches/systemvm/debian/config/root/userdata.sh e5d170b 
> 
> Diff: https://reviews.apache.org/r/11582/diff/
> 
> 
> Testing
> ---
> 
> 4.1 with vSphere
> 
> 
> Thanks,
> 
> ilya musayev
> 
>

Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP

[Chip Childers]

On Thu, Jul 18, 2013 at 10:01:02PM +, Animesh Chaturvedi wrote:
> [Animesh>] Yes this seems late for 4.2, but it makes it easier to add support 
> now. If there is objection we can track it for 4.3.  

Are you sure it's easier to add support now?  That basically puts us in
a position where it's *possible* to attempt it.  That means we have to
test it, because if things go wrong then we are in a mess for 4.3.

Am I getting it wrong?

Re: Reviewboard and jenkins

[Chip Childers]

Freaking awesome.

I saw that hit the ML, and was just about to ask about it.


On Fri, Jul 19, 2013 at 12:51 PM, Hugo Trippaers  wrote:

> Hey all,
>
> I've been working on some scripting to get patches submitted on review
> board automagically tested with Jenkins.
>
> The way it works is that a script will check if the jenkins user on review
> board (catofong) has commented on the review. If it hasn't it will submit a
> job to jenkins.cloudstack.org with a patch parameter. When the job it
> ready it will add a comment to the review with the result of the build and
> the url of that particular build. If the build succeeds cato will set the
> ship it flag to indicate that it's ok.
>
> I'm finishing up the tests with this script now, so before i move it to
> jenkins.cloudstack.org i would like to feedback if we think this would
> use useful.
>
> An result of the test run can be seen here:
> https://reviews.apache.org/r/12762/
>
> Cheers,
>
> Hugo
>
>
>

Re: Reviewboard and jenkins

[Daan Hoogland]

good stuff, does it run the risk of running while another instance is busy?


On Fri, Jul 19, 2013 at 6:57 PM, Prasanna Santhanam  wrote:

> On Fri, Jul 19, 2013 at 06:51:29PM +0200, Hugo Trippaers wrote:
> > Hey all,
> >
> > I've been working on some scripting to get patches submitted on review
> board automagically tested with Jenkins.
> >
> > The way it works is that a script will check if the jenkins user on
> > review board (catofong) has commented on the review. If it hasn't it
> > will submit a job to jenkins.cloudstack.org with a patch parameter.
> > When the job it ready it will add a comment to the review with the
> > result of the build and the url of that particular build. If the
> > build succeeds cato will set the ship it flag to indicate that it's
> > ok.
> >
> > I'm finishing up the tests with this script now, so before i move it
> > to jenkins.cloudstack.org i would like to feedback if we think this
> > would use useful.
> >
> > An result of the test run can be seen here:
> > https://reviews.apache.org/r/12762/
> >
> > Cheers,
> >
> > Hugo
>
> YES! YES! YES! and link it to the checkin tests coming up in the
> `simulator` view on jenkins.cloudstack.org.
>
> --
> Prasanna.,
>
> 
> Powered by BigRock.com
>
>

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[David Nalley]

On Fri, Jul 19, 2013 at 11:32 AM, Chip Childers
 wrote:
> As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
> know we didn't define that term correctly as of now, but I think that's
> a different discussion).  We currently have that term defined as:
>
>> Lazy Consensus - Lazy consensus requires 3 binding +1 votes and no
>> binding -1 votes.
>
> I'd like to propose that we change the PMC and committer voting rule to
> use the Lazy 2/3 Majority approach defined as:
>
>> Lazy 2/3 majority votes requires at least 3 binding votes and twice as
>> many binding +1 votes as binding -1 votes.
>
> Are there any objections to me starting a VOTE on this change?


So despite this arising from a discussion I originated, I am not sure
I want to see this go away. As frustrating as it is to, we are a
community built on consensus, and even if we can't get that around
adding PMC and committer votes, I am not sure that it's wise to add
new folks. Requiring no dissent is a powerful enabler for individuals,
even if it occasionally thwarts the will of the majority. Having seen
other projects with a majority rules approach to this, I don't see
them as better off. While  I might disagree with some of my fellow PMC
members, I still respect their decision, and hope that when I am the
person casting a -1, they'll respect mine as well.

--David

RE: reviewer wanted

[Alex Huang]

Done.

I like that you just request this on the dev list too.

--Alex

> -Original Message-
> From: Daan Hoogland [mailto:daan.hoogl...@gmail.com]
> Sent: Friday, July 19, 2013 1:15 AM
> To: dev
> Subject: reviewer wanted
> 
> H,
> 
> Can someone have a quick look at review request 12685 , please?
> 
> thanks,
> Daan

Review Request 12762: CLOUDSTACK-3620: [Automation] API removevpnuser failed, in test case test_project_usage.TestVpnUsage.test_01_vpn_usage.

[Sanjay Tripathi]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12762/
---

Review request for cloudstack and Prasanna Santhanam.


Bugs: CLOUDSTACK-3620


Repository: cloudstack-git


Description
---

CLOUDSTACK-3620: [Automation] API removevpnuser failed, in test case 
test_project_usage.TestVpnUsage.test_01_vpn_usage.


Diffs
-

  test/integration/component/test_project_usage.py bb253e1 
  tools/marvin/marvin/integration/lib/base.py 13507bc 

Diff: https://reviews.apache.org/r/12762/diff/


Testing
---

Verified the test locally on my cloudstack setup.


Thanks,

Sanjay Tripathi

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[Noah Slater]

Specifically, Chip is calling for us to change committer / PMC votes from
"lazy consensus" to "2/3 majority". (That is, the vote type for that
specific decision making process changes, but the vote type definitions are
left alone.)


On 19 July 2013 17:32, Chip Childers  wrote:

> On Fri, Jul 19, 2013 at 04:29:07PM +, Chiradeep Vittal wrote:
> > There's several places in the by laws that call for Lazy Consensus. Are
> we
> > discussing modifying all of them or just new committer votes?
>
> New committer and PMC membership.
>
> sorry, I think the email could be more clear.  This is per the $subject:
> new committer / new PMC member votes only.
>
> >
> > On 7/19/13 9:02 PM, "Chip Childers"  wrote:
> >
> > >As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
> > >know we didn't define that term correctly as of now, but I think that's
> > >a different discussion).  We currently have that term defined as:
> > >
> > >> Lazy Consensus - Lazy consensus requires 3 binding +1 votes and no
> > >> binding -1 votes.
> > >
> > >I'd like to propose that we change the PMC and committer voting rule to
> > >use the Lazy 2/3 Majority approach defined as:
> > >
> > >> Lazy 2/3 majority votes requires at least 3 binding votes and twice as
> > >> many binding +1 votes as binding -1 votes.
> > >
> > >Are there any objections to me starting a VOTE on this change?
> >
> >
>



-- 
NS

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[Joe Brockmeier]

On Fri, Jul 19, 2013, at 10:32 AM, Chip Childers wrote:
> As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
> know we didn't define that term correctly as of now, but I think that's
> a different discussion).  We currently have that term defined as:

I would be in favor of defining it correctly, though. :-) 

Best,

jzb
-- 
Joe Brockmeier
j...@zonker.net
Twitter: @jzb
http://www.dissociatedpress.net/

Re: Demo Lab - To rent or To buy

[Ian Duffy]

@Kelly - Assuming you mis-read there were no HVM/VT-passthrough issues with
my version of vmware fusion.

@Joe and @David - Thanks for the feedback, will definitely look into that.

Re: Trouble uploading ISO file

[Mike Tutkowski]

Thanks a lot! I wasn't sure if I was doing something wrong there or if a
bug should be logged.


On Fri, Jul 19, 2013 at 11:36 AM, Soheil Eizadi wrote:

> I logged jira bug to track this problem:
> https://issues.apache.org/jira/browse/CLOUDSTACK-3671
> -Soheil
> 
> From: Soheil Eizadi [seiz...@infoblox.com]
> Sent: Wednesday, July 17, 2013 11:57 AM
> To: dev@cloudstack.apache.org
> Subject: RE: Trouble uploading ISO file
>
> The 192.168.56.1 is due to the fact that you have multiple interfaces and
> have not setup your host/network information properly in the CSM database.
> You need the following three arguments to make sure they are consistent:
>
> host
> management.network.cidr
> secstorage.allowed.internal.sites
>
> Restart your CSM and also delete the System VMs so they have the proper
> host information.
>
> About your DNS, I would check your MAC firewall, may be turn it off and
> see if the DNS queries can go out.
>
> -Soheil
> 
> From: Mike Tutkowski [mike.tutkow...@solidfire.com]
> Sent: Wednesday, July 17, 2013 11:22 AM
> To: dev@cloudstack.apache.org
> Subject: Re: Trouble uploading ISO file
>
> Here are the OS boot parameters. I'm not sure why the host is referenced as
> 192.168.56.1 as that's the address my host OS gets when I run DevCloud2
> (which I'm not doing here). That VirtualBox network is up and running,
> however, as I just happen to be running a VM through it.
>
> -- quiet
>
> console=hvc0%template=domP%type=secstorage%host=192.168.56.1%port=8250%name=s-1-VM%zone=1%pod=1%guid=s-1-VM%resource=org.apache.cloudstack.storage.resource.NfsSecondaryStorageResource%instance=SecStorage%sslcopy=true%role=templateProcessor%mtu=1500%eth2ip=172.16.140.50%eth2mask=255.255.255.0%gateway=172.16.140.2%eth0ip=169.254.2.91%eth0mask=255.255.0.0%eth1ip=172.16.140.37%eth1mask=255.255.255.0%mgmtcidr=
>
> 192.168.56.0/24%localgw=172.16.140.2%private.network.device=eth1%eth3ip=172.16.140.36%eth3mask=255.255.255.0%storageip=172.16.140.36%storagenetmask=255.255.255.0%storagegateway=172.16.140.2%internaldns1=172.16.140.2%dns1=8.8.8.8
>
>
> On Wed, Jul 17, 2013 at 12:12 PM, Mike Tutkowski <
> mike.tutkow...@solidfire.com> wrote:
>
> > I notice from the Ubuntu host that I can ping machines on the same
> private
> > network, but not access the Internet.
> >
> > It looks like this test failed to resolve download.cloud.com. Perhaps I
> > have something misconfigured in Fusion?
> >
> >
> > On Wed, Jul 17, 2013 at 12:10 PM, Mike Tutkowski <
> > mike.tutkow...@solidfire.com> wrote:
> >
> >> Interesting...same results. Thoughts?
> >>
> >> root@s-1-VM:~# /usr/local/cloud/systemvm/ssvm-check.sh
> >> 
> >> First DNS server is  172.16.140.2
> >> PING 172.16.140.2 (172.16.140.2): 56 data bytes
> >> 64 bytes from 172.16.140.2: icmp_seq=0 ttl=64 time=2.796 ms
> >> 64 bytes from 172.16.140.2: icmp_seq=1 ttl=64 time=0.959 ms
> >> --- 172.16.140.2 ping statistics ---
> >> 2 packets transmitted, 2 packets received, 0% packet loss
> >> round-trip min/avg/max/stddev = 0.959/1.877/2.796/0.919 ms
> >> Good: Can ping DNS server
> >> 
> >> ERROR: DNS not resolving download.cloud.com
> >> resolv.conf follows
> >> nameserver 172.16.140.2
> >> nameserver 8.8.8.8
> >>
> >>
> >> On Wed, Jul 17, 2013 at 11:59 AM, Mike Tutkowski <
> >> mike.tutkow...@solidfire.com> wrote:
> >>
> >>> I am now re-trying with the following configuration:
> >>>
> >>> Basic zone using a VMware Fusion private network
> >>> Host OS at 172.16.140.1
> >>> DNS at 172.16.140.2
> >>> Ubuntu (for NFS secondary storage) at 172.16.140.5
> >>> XenServer at 172.16.140.6
> >>> System VM range at 172.16.140.30 to 172.16.140.40
> >>> User VM range at 172.16.140.50 to 172.16.140.60
> >>>
> >>>
> >>> On Wed, Jul 17, 2013 at 11:45 AM, Mike Tutkowski <
> >>> mike.tutkow...@solidfire.com> wrote:
> >>>
>  Doh! I believe I made a mistake there. Let me fix that and retry.
>  Thanks!
> 
> 
>  On Wed, Jul 17, 2013 at 11:43 AM, Soheil Eizadi  >wrote:
> 
> > You did not document but I assume since you are using Fusion and your
> > host is 172.16.140.1, your gateway is going to be 172.16.140.2.
> >
> > The Gateway is where you want to send your DNS requests to not your
> > host 172.16.140.1.
> >
> > -Soheil
> > 
> > From: Mike Tutkowski [mike.tutkow...@solidfire.com]
> > Sent: Wednesday, July 17, 2013 10:34 AM
> > To: dev@cloudstack.apache.org
> > Subject: Re: Trouble uploading ISO file
> >
> > Thanks :) Somehow that other thread got opened with a similar subject
> > line.
> >
> > This is the results I received from the script:
> >
> > 
> > First DNS server is  172.16.140.1
> > PING 172.16.140.1 (172.16.140.1): 56 data bytes
>

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[Joe Brockmeier]

On Fri, Jul 19, 2013, at 10:32 AM, Chip Childers wrote:
> Are there any objections to me starting a VOTE on this change?

Even though I occasionally disagree with the outcome of the current
bylaws, I think there's value in having a "veto" over adding committers
or PMC members. I'd be -1 to this.

Best,

jzb
-- 
Joe Brockmeier
j...@zonker.net
Twitter: @jzb
http://www.dissociatedpress.net/

Physical Network / Guest Network / Service Offering Interactions for Basic Networking

[Soheil Eizadi]

I am trying to figure out if the behavior I am seeing for Basic Networking:

For my use case I have a Network Service Offering that has replaced Virtual 
Router service, e.g. DHCP/DNS with my plugin service. Then I created a zone 
with Basic Networking configuration with that Network Service Offering.

What I see is that the defaultGuestNetwork that is created attached to the 
Physical Network does not have my Service listed in its Network Service Map.

-Soheil

More detail below:

mysql> select * from physical_network;
+-+--+--++--+---+---++-+-+-+
| id  | uuid | name | data_center_id | 
vnet | speed | domain_id | broadcast_domain_range | state   | created   
  | removed |
+-+--+--++--+---+---++-+-+-+
| 200 | 3a3b7a2e-e771-4847-bd1b-4d9a4ad27baf | cloud-public |  1 | 
NULL | NULL  |  NULL | POD| Enabled | 2013-07-02 
17:33:56 | NULL|
+-+--+--++--+---+---++-+-+-+
1 row in set (0.00 sec)

mysql> select * from physical_network_service_providers;
++--+-+---+--+-+--+---+--+--+---+-+---+-+--++-+-+-+
| id | uuid | physical_network_id | 
provider_name | state| destination_physical_network_id | 
vpn_service_provided | dhcp_service_provided | dns_service_provided | 
gateway_service_provided | firewall_service_provided | 
source_nat_service_provided | load_balance_service_provided | 
static_nat_service_provided | port_forwarding_service_provided | 
user_data_service_provided | security_group_service_provided | 
networkacl_service_provided | removed |
++--+-+---+--+-+--+---+--+--+---+-+---+-+--++-+-+-+
|  1 | 0a60e2e7-8b57-493f-b3f0-2dccf0b79705 | 200 | 
VirtualRouter | Enabled  |   0 |
1 | 1 |1 |  
  1 | 1 |   1 | 
1 |   1 |   
 1 |  1 |   0 | 
  0 | NULL|
|  2 | eeaa91c4-e97f-4a5e-9f82-b82db6959e3a | 200 | 
SecurityGroupProvider | Enabled  |   0 |
0 | 0 |0 |  
  0 | 0 |   0 | 
0 |   0 |   
 0 |  0 |   1 | 
  0 | NULL|
|  3 | adb626c9-7274-4afa-b4da-dfaa479836fa | 200 | 
VpcVirtualRouter  | Disabled |   0 |
1 | 1 |1 |  
  1 | 0 |   1 | 
1 |   1 |   
 1 |  1 |   0 | 
  1 | NULL|
|  4 | da2fa58c-2e0f-4189-a0cf-c32c19028dae | 200 | 
BaremetalDhcpProvider | Disabled |   0 |
0 | 1 |0 |  
  0 | 0 |   0 | 
0 |   0 |   
 0 |  0 |   0 | 
  0 | NULL|
|  5 | 8832b8e4-152

RE: Query String Request Authentication(QSRA) support by S3 providers

[Jessica Wang]

Sanjeev,

Thanks for the confirmation.

Jessica

-Original Message-
From: Sanjeev Neelarapu 
Sent: Thursday, July 18, 2013 11:05 PM
To: dev@cloudstack.apache.org; Jessica Wang
Cc: Edison Su; Thomas O'Dowd
Subject: RE: Query String Request Authentication(QSRA) support by S3 providers

Hi,

I have not seen the issue in extract template/iso after the changes. Works fine.

Thanks,
Sanjeev

-Original Message-
From: Min Chen [mailto:min.c...@citrix.com] 
Sent: Friday, July 19, 2013 5:08 AM
To: Jessica Wang
Cc: dev@cloudstack.apache.org; Edison Su; Thomas O'Dowd
Subject: Re: Query String Request Authentication(QSRA) support by S3 providers

Thanks Jessica. Tom, did you still see the issue?

-min

On 7/8/13 1:20 PM, "Jessica Wang"  wrote:

>Min,
>
>> would you please take a look at this to see if UI can disable 
>>decoding in displaying this download template url  returned from API?
>
>I just changed UI to not decode the URL returned in extractTemplate, 
>extractIso API.
>
>Jessica
>
>
>-Original Message-
>From: Min Chen
>Sent: Wednesday, July 03, 2013 5:53 PM
>To: dev@cloudstack.apache.org; Thomas O'Dowd
>Cc: Jessica Wang
>Subject: Re: Query String Request Authentication(QSRA) support by S3 
>providers
>
>Jessica, would you please take a look at this to see if UI can disable 
>decoding in displaying this download template url returned from API?
>
>Thanks
>-min
>
>On 7/3/13 5:38 PM, "Min Chen"  wrote:
>
>>By examining further what returned from extractTemplateCmd api, I 
>>realized that the URL returned from API is different from what is 
>>displayed from pop-up dialog from UI. Directly using the link returned 
>>from API (with / encoded as %2F can successfully download the 
>>template. So the issue may not be that bad to upgrade Amazon SDK, but 
>>a simple UI issue. That is, is it possible for UI not to decode 
>>URLEncoded string in this case?
>>
>>Thanks
>>-min
>>
>>On 7/3/13 5:22 PM, "Min Chen"  wrote:
>>
>>>Hi Tom,
>>>
>>> I can reproduce this issue using Cloudian, after investigation, I 
>>>realized that this is a bug in Amazon SDK we have used, based on this
>>>thread: 
>>>http://stackoverflow.com/questions/15473582/amazon-s3-presigned-urls-
>>>esc
>>>a
>>>p
>>>e
>>>-the-slashes-in-the-key. When generatePresignedUrl is called it takes 
>>>the entire key and escapes it, and then creates a signature using the 
>>>escaped key. You cannot use the signature from the escaped key and 
>>>combine it with the unescaped key in the URL. See the bug code here:
>>>
>>> String resourcePath = "/" +
>>> ((bucketName != null) ? bucketName + "/" : "") +
>>> ((key != null) ? ServiceUtils.urlEncode(key) : "") +
>>> ((subResource != null) ? "?" + subResource : "");
>>>
>>>We have two options to fix this:
>>> 1. Either upgrade Amazon SDK to use 1.4.3 version, someone in that 
>>>thread claimed that it is fixed in that version, but I haven't 
>>>checked that.
>>>Currently CloudStack is using 1.3.21. Not sure if this will break 
>>>CloudStack cloud_bridge.
>>> 2. Workaround by creating customized AmazonS3Client to change the 
>>>internal implementation on this.
>>>
>>> Thanks
>>> -min
>>>
>>>
>>>
>>>On 7/2/13 11:31 PM, "Thomas O'Dowd"  wrote:
>>>
Excellent. The link is there now. Thank you Min. I verified that bug 
and closed it.

However - now that I can finally click the download link... I ran 
into the issue that the link doesn't work on AWS or Cloudian. Please 
see this bug for details (latest 4.2 updates included in my test).

https://issues.apache.org/jira/browse/CLOUDSTACK-3341

Tom.

On Tue, 2013-07-02 at 22:54 +, Min Chen wrote:
> Hi Tom,
>   I investigated this issue through the db dump you provided in the 
>bug,  this is an issue with our db view template_view creation 
>script, and it  has been fixed in resolving  another bug 
>(https://issues.apache.org/jira/browse/CLOUDSTACK-3314).
>I
> have verified the fix using your db dump on my local setup. Please 
>check  out latest 4.2 or master code to try again.
> 
>   Thanks
>   -min
> 
> On 7/2/13 2:18 PM, "Min Chen"  wrote:
> 
> >Tom, this seems like an issue with entry stored in our DB. I will
>take
>a
> >look at this bug and update you. Just to clarify, this symptom 
> >only happens when you register these templates to Amazon S3, not 
> >for
>Cloudian
> >or RiakCS S3, right?
> >
> >Thanks
> >-min
> >
> >On 7/1/13 7:27 PM, "Thomas O'Dowd"  wrote:
> >
> >>Yes thanks Jessica. I re-opened the bug again. I know its not a 
> >>gui problem per-say in that the template is not ready to show 
> >>the
>download
> >>link. However, it never becomes ready is the actual problem. 
> >>What
>sets
> >>the "isready" property to true? As far as I can see, the objects 
> >>in
>the
> >>S3 stores (AWS or Cloudian) are complet

Re: Review Request 12752: Async commands can inject the job id

[Ryan Dietrich]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12752/
---

(Updated July 19, 2013, 5:54 p.m.)


Review request for cloudstack and Marcus Sorensen.


Changes
---

Thanks for the code change mod Marcus!  (changed the length call to the regex, 
I verified in the database this was the exact format being used)


Repository: cloudstack-git


Description
---

I added "injectedjobid" to the BaseAsyncCmd class as a parameter.
If set, it will allow you to tell Cloudstack what the job id instead of it 
choosing one.
A basic string length test is done to verify the variable passed in is actually 
a UUID.
If it is not valid, it is ignored and the job generates it's own.


Diffs (updated)
-

  api/src/org/apache/cloudstack/api/BaseAsyncCmd.java 0e6f95d 
  server/src/com/cloud/api/ApiServer.java 95f17af 
  server/src/com/cloud/async/AsyncJobVO.java 41eccb4 
  test/integration/smoke/test_deploy_vm.py 425aeb7 
  tools/marvin/marvin/codegenerator.py 632b8c6 
  tools/marvin/marvin/integration/lib/base.py 161d03c 

Diff: https://reviews.apache.org/r/12752/diff/


Testing
---

Updated marvin, updated the deploy vm test.  Ran multiple async commands 
manually, with and without injectedjobid present, no issues detected.


Thanks,

Ryan Dietrich

Re: Review Request 12762: CLOUDSTACK-3620: [Automation] API removevpnuser failed, in test case test_project_usage.TestVpnUsage.test_01_vpn_usage.

[Jenkins Cloudstack.org]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12762/#review23522
---


Review 12762 PASSED the build test
The url of build cloudstack-master-maven-build-with-patch #316 is : 
http://31.22.81.154:8080/job/cloudstack-master-maven-build-with-patch/316/

- Jenkins Cloudstack.org


On July 19, 2013, 3:44 p.m., Sanjay Tripathi wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12762/
> ---
> 
> (Updated July 19, 2013, 3:44 p.m.)
> 
> 
> Review request for cloudstack and Prasanna Santhanam.
> 
> 
> Bugs: CLOUDSTACK-3620
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-3620: [Automation] API removevpnuser failed, in test case 
> test_project_usage.TestVpnUsage.test_01_vpn_usage.
> 
> 
> Diffs
> -
> 
>   test/integration/component/test_project_usage.py bb253e1 
>   tools/marvin/marvin/integration/lib/base.py 13507bc 
> 
> Diff: https://reviews.apache.org/r/12762/diff/
> 
> 
> Testing
> ---
> 
> Verified the test locally on my cloudstack setup.
> 
> 
> Thanks,
> 
> Sanjay Tripathi
> 
>

Re: [DISCUSS] Bylaw changes for new committer / new PMC member votes

[Noah Slater]

Yes, ignoring our egregious mis-use of that term, I am +1 on your idea.
Though, perhaps 3/4 is safer.

Consensus is important, yes. But the bigger the PMC, the harder it is to
achieve. And more often than not, I see no reason to block an action when a
supermajority are clearly in favour of it. Let's not trade hard-nosed
consensus for sclerosis.


On 19 July 2013 16:32, Chip Childers  wrote:

> As it stands now, we currently use a "Lazy Consensus" model (yes Noah, I
> know we didn't define that term correctly as of now, but I think that's
> a different discussion).  We currently have that term defined as:
>
> > Lazy Consensus - Lazy consensus requires 3 binding +1 votes and no
> > binding -1 votes.
>
> I'd like to propose that we change the PMC and committer voting rule to
> use the Lazy 2/3 Majority approach defined as:
>
> > Lazy 2/3 majority votes requires at least 3 binding votes and twice as
> > many binding +1 votes as binding -1 votes.
>
> Are there any objections to me starting a VOTE on this change?
>



-- 
NS

Reviewboard and jenkins

[Hugo Trippaers]

Hey all,

I've been working on some scripting to get patches submitted on review board 
automagically tested with Jenkins.

The way it works is that a script will check if the jenkins user on review 
board (catofong) has commented on the review. If it hasn't it will submit a job 
to jenkins.cloudstack.org with a patch parameter. When the job it ready it will 
add a comment to the review with the result of the build and the url of that 
particular build. If the build succeeds cato will set the ship it flag to 
indicate that it's ok.

I'm finishing up the tests with this script now, so before i move it to 
jenkins.cloudstack.org i would like to feedback if we think this would use 
useful.

An result of the test run can be seen here: https://reviews.apache.org/r/12762/

Cheers,

Hugo

RE: Trouble uploading ISO file

[Soheil Eizadi]

I logged jira bug to track this problem:
https://issues.apache.org/jira/browse/CLOUDSTACK-3671
-Soheil

From: Soheil Eizadi [seiz...@infoblox.com]
Sent: Wednesday, July 17, 2013 11:57 AM
To: dev@cloudstack.apache.org
Subject: RE: Trouble uploading ISO file

The 192.168.56.1 is due to the fact that you have multiple interfaces and have 
not setup your host/network information properly in the CSM database. You need 
the following three arguments to make sure they are consistent:

host
management.network.cidr
secstorage.allowed.internal.sites

Restart your CSM and also delete the System VMs so they have the proper host 
information.

About your DNS, I would check your MAC firewall, may be turn it off and see if 
the DNS queries can go out.

-Soheil

From: Mike Tutkowski [mike.tutkow...@solidfire.com]
Sent: Wednesday, July 17, 2013 11:22 AM
To: dev@cloudstack.apache.org
Subject: Re: Trouble uploading ISO file

Here are the OS boot parameters. I'm not sure why the host is referenced as
192.168.56.1 as that's the address my host OS gets when I run DevCloud2
(which I'm not doing here). That VirtualBox network is up and running,
however, as I just happen to be running a VM through it.

-- quiet
console=hvc0%template=domP%type=secstorage%host=192.168.56.1%port=8250%name=s-1-VM%zone=1%pod=1%guid=s-1-VM%resource=org.apache.cloudstack.storage.resource.NfsSecondaryStorageResource%instance=SecStorage%sslcopy=true%role=templateProcessor%mtu=1500%eth2ip=172.16.140.50%eth2mask=255.255.255.0%gateway=172.16.140.2%eth0ip=169.254.2.91%eth0mask=255.255.0.0%eth1ip=172.16.140.37%eth1mask=255.255.255.0%mgmtcidr=
192.168.56.0/24%localgw=172.16.140.2%private.network.device=eth1%eth3ip=172.16.140.36%eth3mask=255.255.255.0%storageip=172.16.140.36%storagenetmask=255.255.255.0%storagegateway=172.16.140.2%internaldns1=172.16.140.2%dns1=8.8.8.8


On Wed, Jul 17, 2013 at 12:12 PM, Mike Tutkowski <
mike.tutkow...@solidfire.com> wrote:

> I notice from the Ubuntu host that I can ping machines on the same private
> network, but not access the Internet.
>
> It looks like this test failed to resolve download.cloud.com. Perhaps I
> have something misconfigured in Fusion?
>
>
> On Wed, Jul 17, 2013 at 12:10 PM, Mike Tutkowski <
> mike.tutkow...@solidfire.com> wrote:
>
>> Interesting...same results. Thoughts?
>>
>> root@s-1-VM:~# /usr/local/cloud/systemvm/ssvm-check.sh
>> 
>> First DNS server is  172.16.140.2
>> PING 172.16.140.2 (172.16.140.2): 56 data bytes
>> 64 bytes from 172.16.140.2: icmp_seq=0 ttl=64 time=2.796 ms
>> 64 bytes from 172.16.140.2: icmp_seq=1 ttl=64 time=0.959 ms
>> --- 172.16.140.2 ping statistics ---
>> 2 packets transmitted, 2 packets received, 0% packet loss
>> round-trip min/avg/max/stddev = 0.959/1.877/2.796/0.919 ms
>> Good: Can ping DNS server
>> 
>> ERROR: DNS not resolving download.cloud.com
>> resolv.conf follows
>> nameserver 172.16.140.2
>> nameserver 8.8.8.8
>>
>>
>> On Wed, Jul 17, 2013 at 11:59 AM, Mike Tutkowski <
>> mike.tutkow...@solidfire.com> wrote:
>>
>>> I am now re-trying with the following configuration:
>>>
>>> Basic zone using a VMware Fusion private network
>>> Host OS at 172.16.140.1
>>> DNS at 172.16.140.2
>>> Ubuntu (for NFS secondary storage) at 172.16.140.5
>>> XenServer at 172.16.140.6
>>> System VM range at 172.16.140.30 to 172.16.140.40
>>> User VM range at 172.16.140.50 to 172.16.140.60
>>>
>>>
>>> On Wed, Jul 17, 2013 at 11:45 AM, Mike Tutkowski <
>>> mike.tutkow...@solidfire.com> wrote:
>>>
 Doh! I believe I made a mistake there. Let me fix that and retry.
 Thanks!


 On Wed, Jul 17, 2013 at 11:43 AM, Soheil Eizadi 
 wrote:

> You did not document but I assume since you are using Fusion and your
> host is 172.16.140.1, your gateway is going to be 172.16.140.2.
>
> The Gateway is where you want to send your DNS requests to not your
> host 172.16.140.1.
>
> -Soheil
> 
> From: Mike Tutkowski [mike.tutkow...@solidfire.com]
> Sent: Wednesday, July 17, 2013 10:34 AM
> To: dev@cloudstack.apache.org
> Subject: Re: Trouble uploading ISO file
>
> Thanks :) Somehow that other thread got opened with a similar subject
> line.
>
> This is the results I received from the script:
>
> 
> First DNS server is  172.16.140.1
> PING 172.16.140.1 (172.16.140.1): 56 data bytes
> 64 bytes from 172.16.140.1: icmp_seq=0 ttl=64 time=1.977 ms
> 64 bytes from 172.16.140.1: icmp_seq=1 ttl=64 time=1.508 ms
> --- 172.16.140.1 ping statistics ---
> 2 packets transmitted, 2 packets received, 0% packet loss
> round-trip min/avg/max/stddev = 1.508/1.742/1.977/0.235 ms
> Good: Can ping DNS server
> 

Re: [discuss] add code style checker rules to master?

[Mike Tutkowski]

I think this is a really good idea, Donal.

Nice post, too, by the way.


On Fri, Jul 19, 2013 at 6:08 AM, Donal Lafferty
wrote:

> For automated code style checking, I created a rules file for CheckStyle
> based on the ACS coding conventions, and blogged on how to use it with
> Eclipse (
> http://dlafferty.blogspot.co.uk/2013/07/apache-cloudstack-java-coding.html).
>
> However, the rules are not 100% correct, and it is difficult for the
> community to correct them.
>
> Would people be interested in adding it to master under
> ./tools/eclipse/acs_checkstyle.xml ?
>



-- 
*Mike Tutkowski*
*Senior CloudStack Developer, SolidFire Inc.*
e: mike.tutkow...@solidfire.com
o: 303.746.7302
Advancing the way the world uses the
cloud
*™*

RE: code formatting for enums

[Frank Zhang]

Alex, John, I do think we should think carefully about changing the enum names.
First to me it's not a surprise that changing enum names may break API and 
database,
as in java enum name natively represent value. If we want to overcome this 
restriction
I would suggest directly using String so variable name won't effect final value 
in DB/API
anymore. 

Introducing some translation layer for enum would be a nightmare for 
maintenance.
Developers have to add lots of annotation on members of API class just for 
converting passed
in parameters to internal enum because java's enum convention is all 
capitalized. 

> -Original Message-
> From: Alex Huang [mailto:alex.hu...@citrix.com]
> Sent: Friday, July 19, 2013 8:38 AM
> To: dev@cloudstack.apache.org
> Subject: RE: code formatting for enums
> 
> John,
> 
> Just to be clear.  Not doing enum change correctly can cause problems with
> existing databases.  That's a known problem with enum persistence in Java
> because enums by default using the name of the constant to represent the
> string of the constant.  I was just cautioning anyone who wants to take that 
> up
> to make sure they're doing the right thing.  It doesn't require some huge (or
> even minor) framework/architectural change in CloudStack in order for enum
> changes to work with db.
> 
> --Alex
> 
> > -Original Message-
> > From: John Burwell [mailto:jburw...@basho.com]
> > Sent: Friday, July 19, 2013 6:33 AM
> > To: dev@cloudstack.apache.org
> > Subject: Re: code formatting for enums
> >
> > Frank,
> >
> > I expect that renaming an enum value to be a minor change (i.e. once
> > the codebase compiles, the change is good).  However, i can say I am
> > surprised to learn that such an operation could break the API and/or
> > corrupt that database.  To my mind, this issue has nothing to do with
> developer choice.
> > Instead, our architecture needs to recognize that an enumerated value
> > may have multiple representations dependent on the context (e.g.
> > CloudStack HTTP API, AWS API, UI, code, database, etc), and provide
> > mechanisms to perform transformations between them when they need to
> > differ.  From a persistence perspective, we need to should be keying
> > off of a value that is unlikely to change (e.g. an artificial ID).  I
> > would submit that Java enumeration value names, as code symbols, are
> > subject to an unacceptable level of change to be used for persistence.
> >
> > Thanks,
> > -John
> >
> > On Jul 18, 2013, at 8:49 PM, Frank Zhang  wrote:
> >
> > > Yes it's brittle.  I assume "law of least astonishment" here means
> > > reducing learning curve of developers and the changes they made
> > > should not surprise user. To achieve this the code must provide very
> > > flexibility to
> > developer. However, I am thinking of such kind of flexibility is
> > really needed, isn't it?
> > > Currently "Convention over configuration" which sacrifices
> > > flexibility for simplicity is getting more and more popular. If we
> > > can state clearly to developers on constructing data structure from
> > > database tables
> > to API responses, we are achieving great simplicity and overall coding
> > convention.
> > >
> > > Sometimes, it's better off offering only choice to developer and
> > > don't make
> > them think. Anyway, a [DISCUSS] thread is necessary.
> > >
> > >> -Original Message-
> > >> From: John Burwell [mailto:jburw...@basho.com]
> > >> Sent: Thursday, July 18, 2013 4:39 PM
> > >> To: dev@cloudstack.apache.org
> > >> Subject: Re: code formatting for enums
> > >>
> > >> Alex and Frank,
> > >>
> > >> In terms of conventions, the APIs exposed Java SDK and many other
> > >> common APIs following this convention.  Hence, the reason for my
> > suggestion.
> > >>
> > >> The notion that changing an enum key can break the API and/or
> > >> database persistence feels a bit brittle.  In particular, it
> > >> doesn't conform to the Law of Least Surprise.  It feels like we
> > >> should have a transformation mechanism from the API endpoint to an
> > >> enum value, and employ foreign, artificial keys to code tables in
> > >> schema.  This conversation has expanded a bit, and it seems
> > >> appropriate to open a new [DISCUSS] thread to delve into it further.  Do
> y'all agree?
> > >>
> > >> Thanks,
> > >> -John
> > >>
> > >> On Jul 18, 2013, at 7:19 PM, Frank Zhang 
> wrote:
> > >>
> > >>> Those enums cannot be simply considered as internal data
> > >>> structures where
> > >> code convention applies to, they should be considered in API level.
> > >>>
> > >>> Most CloudStack API responses retrieve its fields from xxxVO
> > >>> classes which represent database table. In our example, enum State
> > >>> directly maps to UserVmResponse. state. Then the most important
> > >>> factor of name
> > >> convention is user experience where Running is more user friendly
> > >> than IS_RUNNING or whatever all capitalized sentence splitting by
> > underscore.
> > >>>
> > >>> And any ch

RE: code formatting for enums

[Alex Huang]

John, 

That's actually not necessary.  Sometimes code speak clearer than words.

Take Donal's example:

Enum HypervisorType {
  Hyperv;
}

Let's say we have a VO with the following declared.

@Column(name="type", updatable = true, nullable=false)
@Enumerated(value=EnumType.STRING)
private HypervisorType type;

This would persist the String literal hyperv into the database.

Now Donal doesn't like that and he's correct, it should be HyperV.  Normally he 
just makes this change.  
Enum HypervisorType {
  HyperV;
}

But because the String literal is stored in db, he instead should make this 
change

Enum HypervisorType {
  HyperV("Hyperv");
  
Private final String _dbValue;
Private HypervisorType(String dbValue) {
  _dbValue = dbValue;
}

Public String name() {return _dbValue; }
}

That will maintain db backwards compatibility because the DAO layer knows to 
use the name to map to the string literal stored in the db.  You don't need any 
helper classes to store it and everything is clearly specified in the 
declaration of the enum itself.  It's straightforward to read and understand.  
Something similar can be done if the enum was stored as an ordinal instead of a 
String.  I generally prefer storing enums as Strings because it's much easier 
to read when looking at the DB but I have noticed others who have used ordinals 
to store.  

On API side is different, mainly because API side didn't have a framework like 
the DAO layer that already defined how to deal with enums.  And also, API also 
needs i18n etc.  That side probably require a discussion.

--Alex


> -Original Message-
> From: John Burwell [mailto:jburw...@basho.com]
> Sent: Friday, July 19, 2013 12:06 PM
> To: dev@cloudstack.apache.org
> Subject: Re: code formatting for enums
> 
> Alex/Frank/Daan,
> 
> In terms of naming convention, it is not the highest priority to me.  However,
> I would encourage us to follow the conventions of the larger Java eco system
> unless we have a good reason to diverge.  Closely tracking the conventions
> and norms of the larger Java eco-system reduces the learning curve for new
> contributors and observers of our efforts.  In this case, the larger community
> has selected the standard I enumerated previously.
> 
> The greater concern to me is the brittleness we have been discussing, and its
> impact on the long term maintainability and extensibility of the code base.
> To the best of my knowledge, Java does not require that an enumerated
> value be persisted by its named.  For the reasons stated below, I have never
> personally persisted that representation because Java only guarantees the
> consistency of the name for the lifetime the enum's class file.  I typically
> create an Identifiable interface on all persistent objects that provides a
> durable identifier with a stronger consistency guarantee - the persistent
> lifetime of the object.  The persistence mechanism need to be modified to
> understand the semantics of such a mechanism, the existing enums would
> need to be fitted. Coupled with a couple of helper classes used by every
> enum, this approach can be extremely compact and easy to maintain.  To my
> way of thinking, such a change is not complicated, but the wide impact to the
> code base and persisted data would make it a large endeavor none the less.
> 
> I will open a [DISCUSS] early next week to discuss more deeply.
> 
> Thanks,
> -John
> 
> On Jul 19, 2013, at 2:58 PM, Frank Zhang  wrote:
> 
> > Alex, John, I do think we should think carefully about changing the enum
> names.
> > First to me it's not a surprise that changing enum names may break API
> > and database, as in java enum name natively represent value. If we
> > want to overcome this restriction I would suggest directly using
> > String so variable name won't effect final value in DB/API anymore.
> >
> > Introducing some translation layer for enum would be a nightmare for
> maintenance.
> > Developers have to add lots of annotation on members of API class just
> > for converting passed in parameters to internal enum because java's enum
> convention is all capitalized.
> >
> >> -Original Message-
> >> From: Alex Huang [mailto:alex.hu...@citrix.com]
> >> Sent: Friday, July 19, 2013 8:38 AM
> >> To: dev@cloudstack.apache.org
> >> Subject: RE: code formatting for enums
> >>
> >> John,
> >>
> >> Just to be clear.  Not doing enum change correctly can cause problems
> >> with existing databases.  That's a known problem with enum
> >> persistence in Java because enums by default using the name of the
> >> constant to represent the string of the constant.  I was just
> >> cautioning anyone who wants to take that up to make sure they're
> >> doing the right thing.  It doesn't require some huge (or even minor)
> >> framework/architectural change in CloudStack in order for enum changes
> to work with db.
> >>
> >> --Alex
> >>
> >>> -Original Message-
> >>> From: John Burwell [mailto:jburw...@basho.com]
> >>> Sent: 

Re: code formatting for enums

[John Burwell]

Alex/Frank/Daan,

In terms of naming convention, it is not the highest priority to me.  However, 
I would encourage us to follow the conventions of the larger Java eco system 
unless we have a good reason to diverge.  Closely tracking the conventions and 
norms of the larger Java eco-system reduces the learning curve for new 
contributors and observers of our efforts.  In this case, the larger community 
has selected the standard I enumerated previously.

The greater concern to me is the brittleness we have been discussing, and its 
impact on the long term maintainability and extensibility of the code base.  To 
the best of my knowledge, Java does not require that an enumerated value be 
persisted by its named.  For the reasons stated below, I have never personally 
persisted that representation because Java only guarantees the consistency of 
the name for the lifetime the enum's class file.  I typically create an 
Identifiable interface on all persistent objects that provides a durable 
identifier with a stronger consistency guarantee - the persistent lifetime of 
the object.  The persistence mechanism need to be modified to understand the 
semantics of such a mechanism, the existing enums would need to be fitted. 
Coupled with a couple of helper classes used by every enum, this approach can 
be extremely compact and easy to maintain.  To my way of thinking, such a 
change is not complicated, but the wide impact to the code base and persisted 
data would make it a large endeavor none the less.

I will open a [DISCUSS] early next week to discuss more deeply.

Thanks,
-John

On Jul 19, 2013, at 2:58 PM, Frank Zhang  wrote:

> Alex, John, I do think we should think carefully about changing the enum 
> names.
> First to me it's not a surprise that changing enum names may break API and 
> database,
> as in java enum name natively represent value. If we want to overcome this 
> restriction
> I would suggest directly using String so variable name won't effect final 
> value in DB/API
> anymore. 
> 
> Introducing some translation layer for enum would be a nightmare for 
> maintenance.
> Developers have to add lots of annotation on members of API class just for 
> converting passed
> in parameters to internal enum because java's enum convention is all 
> capitalized. 
> 
>> -Original Message-
>> From: Alex Huang [mailto:alex.hu...@citrix.com]
>> Sent: Friday, July 19, 2013 8:38 AM
>> To: dev@cloudstack.apache.org
>> Subject: RE: code formatting for enums
>> 
>> John,
>> 
>> Just to be clear.  Not doing enum change correctly can cause problems with
>> existing databases.  That's a known problem with enum persistence in Java
>> because enums by default using the name of the constant to represent the
>> string of the constant.  I was just cautioning anyone who wants to take that 
>> up
>> to make sure they're doing the right thing.  It doesn't require some huge (or
>> even minor) framework/architectural change in CloudStack in order for enum
>> changes to work with db.
>> 
>> --Alex
>> 
>>> -Original Message-
>>> From: John Burwell [mailto:jburw...@basho.com]
>>> Sent: Friday, July 19, 2013 6:33 AM
>>> To: dev@cloudstack.apache.org
>>> Subject: Re: code formatting for enums
>>> 
>>> Frank,
>>> 
>>> I expect that renaming an enum value to be a minor change (i.e. once
>>> the codebase compiles, the change is good).  However, i can say I am
>>> surprised to learn that such an operation could break the API and/or
>>> corrupt that database.  To my mind, this issue has nothing to do with
>> developer choice.
>>> Instead, our architecture needs to recognize that an enumerated value
>>> may have multiple representations dependent on the context (e.g.
>>> CloudStack HTTP API, AWS API, UI, code, database, etc), and provide
>>> mechanisms to perform transformations between them when they need to
>>> differ.  From a persistence perspective, we need to should be keying
>>> off of a value that is unlikely to change (e.g. an artificial ID).  I
>>> would submit that Java enumeration value names, as code symbols, are
>>> subject to an unacceptable level of change to be used for persistence.
>>> 
>>> Thanks,
>>> -John
>>> 
>>> On Jul 18, 2013, at 8:49 PM, Frank Zhang  wrote:
>>> 
 Yes it's brittle.  I assume "law of least astonishment" here means
 reducing learning curve of developers and the changes they made
 should not surprise user. To achieve this the code must provide very
 flexibility to
>>> developer. However, I am thinking of such kind of flexibility is
>>> really needed, isn't it?
 Currently "Convention over configuration" which sacrifices
 flexibility for simplicity is getting more and more popular. If we
 can state clearly to developers on constructing data structure from
 database tables
>>> to API responses, we are achieving great simplicity and overall coding
>>> convention.
 
 Sometimes, it's better off offering only choice to developer and
 don't make
>>> the

Re: Query String Request Authentication(QSRA) support by S3 providers

[Min Chen]

Thanks, Tom and Sanjeev for verification.

On 7/19/13 1:26 AM, "Thomas O'Dowd"  wrote:

>Hi Min,
>
>I will have time next week to check through the bugs and verify
>everything. Thanks for checking on me.
>
>Tom.
>
>On Thu, 2013-07-18 at 23:37 +, Min Chen wrote:
>> Thanks Jessica. Tom, did you still see the issue?
>> 
>> -min
>> 
>> On 7/8/13 1:20 PM, "Jessica Wang"  wrote:
>> 
>> >Min, 
>> >
>> >> would you please take a look at this to see if UI can disable
>>decoding
>> >>in displaying this download template url
>> >> returned from API?
>> >
>> >I just changed UI to not decode the URL returned in extractTemplate,
>> >extractIso API.
>> >
>> >Jessica
>> >
>> >
>> >-Original Message-
>> >From: Min Chen 
>> >Sent: Wednesday, July 03, 2013 5:53 PM
>> >To: dev@cloudstack.apache.org; Thomas O'Dowd
>> >Cc: Jessica Wang
>> >Subject: Re: Query String Request Authentication(QSRA) support by S3
>> >providers
>> >
>> >Jessica, would you please take a look at this to see if UI can disable
>> >decoding in displaying this download template url returned from API?
>> >
>> >Thanks
>> >-min
>> >
>> >On 7/3/13 5:38 PM, "Min Chen"  wrote:
>> >
>> >>By examining further what returned from extractTemplateCmd api, I
>> >>realized
>> >>that the URL returned from API is different from what is displayed
>>from
>> >>pop-up dialog from UI. Directly using the link returned from API
>>(with /
>> >>encoded as %2F can successfully download the template. So the issue
>>may
>> >>not be that bad to upgrade Amazon SDK, but a simple UI issue. That
>>is, is
>> >>it possible for UI not to decode URLEncoded string in this case?
>> >>
>> >>Thanks
>> >>-min
>> >>
>> >>On 7/3/13 5:22 PM, "Min Chen"  wrote:
>> >>
>> >>>Hi Tom,
>> >>>
>> >>>  I can reproduce this issue using Cloudian, after investigation, I
>> >>>realized that this is a bug in Amazon SDK we have used, based on this
>> >>>thread: 
>> 
>http://stackoverflow.com/questions/15473582/amazon-s3-presigned-urls-e
>sc
>> >>>a
>> >>>p
>> >>>e
>> >>>-the-slashes-in-the-key. When generatePresignedUrl is called it takes
>> >>>the
>> >>>entire key and escapes it, and then creates a signature using the
>> >>>escaped
>> >>>key. You cannot use the signature from the escaped key and combine it
>> >>>with
>> >>>the unescaped key in the URL. See the bug code here:
>> >>>
>> >>>  String resourcePath = "/" +
>> >>>  ((bucketName != null) ? bucketName + "/" : "") +
>> >>>  ((key != null) ? ServiceUtils.urlEncode(key) : "") +
>> >>>  ((subResource != null) ? "?" + subResource : "");
>> >>>
>> >>>We have two options to fix this:
>> >>>  1. Either upgrade Amazon SDK to use 1.4.3 version, someone in that
>> >>>thread
>> >>>claimed that it is fixed in that version, but I haven't checked that.
>> >>>Currently CloudStack is using 1.3.21. Not sure if this will break
>> >>>CloudStack cloud_bridge.
>> >>>  2. Workaround by creating customized AmazonS3Client to change the
>> >>>internal implementation on this.
>> >>>
>> >>>  Thanks
>> >>>  -min
>> >>>
>> >>>
>> >>>
>> >>>On 7/2/13 11:31 PM, "Thomas O'Dowd"  wrote:
>> >>>
>> Excellent. The link is there now. Thank you Min. I verified that bug
>> and
>> closed it.
>> 
>> However - now that I can finally click the download link... I ran
>>into
>> the issue that the link doesn't work on AWS or Cloudian. Please see
>> this
>> bug for details (latest 4.2 updates included in my test).
>> 
>> https://issues.apache.org/jira/browse/CLOUDSTACK-3341
>> 
>> Tom.
>> 
>> On Tue, 2013-07-02 at 22:54 +, Min Chen wrote:
>> > Hi Tom,
>> >I investigated this issue through the db dump you provided in 
>> > the
>> >bug,
>> > this is an issue with our db view template_view creation script,
>>and
>> >it
>> > has been fixed in resolving
>> > another bug
>>(https://issues.apache.org/jira/browse/CLOUDSTACK-3314).
>> >I
>> > have verified the fix using your db dump on my local setup. Please
>> >check
>> > out latest 4.2 or master code to try again.
>> > 
>> >Thanks
>> >-min
>> > 
>> > On 7/2/13 2:18 PM, "Min Chen"  wrote:
>> > 
>> > >Tom, this seems like an issue with entry stored in our DB. I will
>> >take
>> >a
>> > >look at this bug and update you. Just to clarify, this symptom
>>only
>> > >happens when you register these templates to Amazon S3, not for
>> >Cloudian
>> > >or RiakCS S3, right?
>> > >
>> > >Thanks
>> > >-min
>> > >
>> > >On 7/1/13 7:27 PM, "Thomas O'Dowd"  wrote:
>> > >
>> > >>Yes thanks Jessica. I re-opened the bug again. I know its not a
>>gui
>> > >>problem per-say in that the template is not ready to show the
>> >download
>> > >>link. However, it never becomes ready is the actual problem.
>>What
>> >sets
>> > >>the "isready" property to true? As far as I can see, the
>>objects in
>> >the
>> > >>S3 stores (AWS or C

Re: Review Request 12712: Add a new snapshot flag in cloudstack for KVM 6.3 and verify it during the upgrade

[edison su]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12712/#review23548
---

Ship it!


Ship It!

- edison su


On July 17, 2013, 10:34 p.m., Fang Wang wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12712/
> ---
> 
> (Updated July 17, 2013, 10:34 p.m.)
> 
> 
> Review request for cloudstack and edison su.
> 
> 
> Bugs: CLOUDSTACK-732
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> This is the CS code change part to add back the snapshot support for KVM 6.3. 
> The qemu patch is a separate rpm patch.
> This part is for CS to check if the support flag is on or not. 
> 
> 
> Diffs
> -
> 
>   engine/schema/src/com/cloud/upgrade/dao/Upgrade410to420.java f89c862 
>   server/src/com/cloud/configuration/Config.java e77f6ce 
>   server/src/com/cloud/storage/snapshot/SnapshotManagerImpl.java 2c9ef82 
> 
> Diff: https://reviews.apache.org/r/12712/diff/
> 
> 
> Testing
> ---
> 
> 
> Thanks,
> 
> Fang Wang
> 
>

Re: New user - Introduction

[Ahmad Emneina]

Awesome, the more fixes the better. Welcome!!


On Fri, Jul 19, 2013 at 1:19 AM, René Diepstraten
wrote:

> Hi everyone.
>
> Since I have joined the mailing list, I'd like to say hello and introduce
> myself.
> My name is Rene Diepstraten, am a system administrator and the company I
> work for ( PCextreme ) uses Apache CloudStack.
>
> I'd like to contribute here and there with my skills in python and bash,
> and perhaps learn some java.
> A few small patches that I wrote have been accepted into Master, and I'd
> like to contribute more to this great project in the future.
>
> Best regards,
>
> Rene Diepstraten
>

Re: Object Store Region Assignment

[John Burwell]

Chip,

So two regions won't share the same database?

Thanks,
-John

On Jul 19, 2013, at 4:06 PM, Chip Childers  wrote:

> On Fri, Jul 19, 2013 at 03:46:36PM -0400, John Burwell wrote:
>> All,
>> 
>> In reviewing the 4.2 object store enhancements, object stores are defined 
>> with region scope.  However, there is no way to assign an object store 
>> definition to a region.  To my way of thinking, we must be able to assign an 
>> object store to a region to avoid transfer of large assets across WAN links. 
>>  Furthermore, it greatly reduces availability of the object store cluster 
>> because the management server/ssvm may attempt to connect to an object store 
>> over a WAN link when co-located instance is available.  Why isn't assignment 
>> of an object store to a region supported? 
>> 
>> Thanks,
>> -John
> 
> AFAIK, a management server is only a control point for a single region.
> The Region entity is used to provide users with a reference to other
> regions that may be part of the same service.  If I'm right, then the
> locally configured object-store would be for that region only.
> 
> Other feel free to correct me.

Re: [DISCUSS] coding convention for method - and class length

[John Burwell]

All,

I agree with Chip's sentiments.  Personally, I strive for the "one screen full" 
rule of thumb.  However, I think a hard and fast rule is a bit restrictive.

Thanks,
-John

On Jul 18, 2013, at 5:21 PM, Mike Tutkowski  
wrote:

> Like eight or so years ago, I sent a method I had to modify to the printer
> (so I could study it on regular paper) and it came out on 14 pages.
> 
> 
> On Thu, Jul 18, 2013 at 3:04 PM, Donal Lafferty
> wrote:
> 
>>> -Original Message-
>>> From: Chip Childers [mailto:chip.child...@sungard.com]
>>> Sent: 18 July 2013 9:43 PM
>>> To: dev@cloudstack.apache.org
>>> Subject: Re: [DISCUSS] coding convention for method - and class length
>>> 
>>> On Thu, Jul 18, 2013 at 02:34:36PM -0600, Mike Tutkowski wrote:
 I'm not sure how I feel about an arbitrary number of lines per method
 (although 200 is obviously quite high and I would recommend
 modularizing such a method), but I'm not in favor of limiting the
 number of methods per class (especially not to just 10). Some types of
 objects simply need many discrete operations and 10 is too limiting.
>>> 
>>> +1 to both thoughts.
>>> 
>>> Smaller methods is good, but a specific number of lines as policy vs. a
>> rule of
>>> thumb are two different things.
>>> 
>>> I completely agree with Daan's underlying concern though...  some of
>> these
>>> class files are horrible to try and comprehend.
>>> 
>>> My favorite example:
>>> 
>>> https://git-wip-
>>> 
>> us.apache.org/repos/asf/cloudstack/?p=cloudstack.git;a=blob_plain;f=plugin
>>> s/hypervisors/xen/src/com/cloud/hypervisor/xen/resource/CitrixResourceB
>>> ase.java;hb=master
>>> 
>> [Donal Lafferty]
>> StyleCheck sets a default limit of 150 lines.
>> 
>> Meanwhile, have you heard of the function 'best kick' in a certain
>> company's soccer video game?  IIRC, it's several 1000 lines.
>> 
>> 
> 
> 
> -- 
> *Mike Tutkowski*
> *Senior CloudStack Developer, SolidFire Inc.*
> e: mike.tutkow...@solidfire.com
> o: 303.746.7302
> Advancing the way the world uses the
> cloud
> *™*

Re: Object Store Region Assignment

[Chip Childers]

On Fri, Jul 19, 2013 at 4:31 PM, John Burwell  wrote:

> Chip,
>
> So two regions won't share the same database?
>

Nope, that's why you have to use something (not provided yet) to sync the
acct / domain metadata between them.


>
> Thanks,
> -John
>
> On Jul 19, 2013, at 4:06 PM, Chip Childers 
> wrote:
>
> > On Fri, Jul 19, 2013 at 03:46:36PM -0400, John Burwell wrote:
> >> All,
> >>
> >> In reviewing the 4.2 object store enhancements, object stores are
> defined with region scope.  However, there is no way to assign an object
> store definition to a region.  To my way of thinking, we must be able to
> assign an object store to a region to avoid transfer of large assets across
> WAN links.  Furthermore, it greatly reduces availability of the object
> store cluster because the management server/ssvm may attempt to connect to
> an object store over a WAN link when co-located instance is available.  Why
> isn't assignment of an object store to a region supported?
> >>
> >> Thanks,
> >> -John
> >
> > AFAIK, a management server is only a control point for a single region.
> > The Region entity is used to provide users with a reference to other
> > regions that may be part of the same service.  If I'm right, then the
> > locally configured object-store would be for that region only.
> >
> > Other feel free to correct me.
>
>
>

Re: Object Store Region Assignment

[John Burwell]

Chip,

Why do we have a region entity in the database if everything is assumed to be 
one and only region?

Thanks,
-John

On Jul 19, 2013, at 4:34 PM, Chip Childers  wrote:

> On Fri, Jul 19, 2013 at 4:31 PM, John Burwell  wrote:
> 
>> Chip,
>> 
>> So two regions won't share the same database?
>> 
> 
> Nope, that's why you have to use something (not provided yet) to sync the
> acct / domain metadata between them.
> 
> 
>> 
>> Thanks,
>> -John
>> 
>> On Jul 19, 2013, at 4:06 PM, Chip Childers 
>> wrote:
>> 
>>> On Fri, Jul 19, 2013 at 03:46:36PM -0400, John Burwell wrote:
 All,
 
 In reviewing the 4.2 object store enhancements, object stores are
>> defined with region scope.  However, there is no way to assign an object
>> store definition to a region.  To my way of thinking, we must be able to
>> assign an object store to a region to avoid transfer of large assets across
>> WAN links.  Furthermore, it greatly reduces availability of the object
>> store cluster because the management server/ssvm may attempt to connect to
>> an object store over a WAN link when co-located instance is available.  Why
>> isn't assignment of an object store to a region supported?
 
 Thanks,
 -John
>>> 
>>> AFAIK, a management server is only a control point for a single region.
>>> The Region entity is used to provide users with a reference to other
>>> regions that may be part of the same service.  If I'm right, then the
>>> locally configured object-store would be for that region only.
>>> 
>>> Other feel free to correct me.
>> 
>> 
>> 

RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP

[Musayev, Ilya]

Animesh,

Would this have any effect on non-CCP installations?

I see this move as a positive thing and it would be ideal to have in 4.2, but 
we don't want to risk stability, hence I'm curious on the safety and possible 
issues for non-CCP users.

Thanks
ilya
 

> -Original Message-
> From: Animesh Chaturvedi [mailto:animesh.chaturv...@citrix.com]
> Sent: Friday, July 19, 2013 4:43 PM
> To: dev@cloudstack.apache.org
> Subject: RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> 
> 
> 
> > -Original Message-
> > From: Chip Childers [mailto:chip.child...@sungard.com]
> > Sent: Friday, July 19, 2013 9:39 AM
> > To: dev@cloudstack.apache.org
> > Subject: Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> >
> > On Thu, Jul 18, 2013 at 10:01:02PM +, Animesh Chaturvedi wrote:
> > > [Animesh>] Yes this seems late for 4.2, but it makes it easier to
> > > add
> > support now. If there is objection we can track it for 4.3.
> >
> > Are you sure it's easier to add support now?  That basically puts us
> > in a position where it's *possible* to attempt it.  That means we have
> > to test it, because if things go wrong then we are in a mess for 4.3.
> >
> > Am I getting it wrong?
> [Animesh>] Yes it is one upgrade path from 3.0.7 to 4.1 and follow through
> same upgrade path from 4.1 to 4.2 with couple of SQL files only. . Upgrade
> from 3.0,4-> 3.0.7 has already been tested in CCP releases and becomes the
> checkpoint. It was planned to be tested for CCP anyway. Sudha do you see it
> as a concern. Again if community thinks it is late, we can track for 4.3

Review Request 12775: Migrate the dozens of ssh calls for KVM VmDataCommand to a single router command

[Marcus Sorensen]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12775/
---

Review request for cloudstack, ilya musayev and Wido den Hollander.


Bugs: cloudstack-3163


Repository: cloudstack-git


Description
---

Migrate the dozens of ssh calls for KVM VmDataCommand to a single router 
command.


Diffs
-

  core/src/com/cloud/agent/resource/virtualnetwork/VirtualRoutingResource.java 
0b26220 
  patches/systemvm/debian/config/opt/cloud/bin/vmdata_kvm.py PRE-CREATION 
  scripts/network/domr/vm_data.sh c861723 

Diff: https://reviews.apache.org/r/12775/diff/


Testing
---

Tested on 4.2 advanced zone, as well as a version against 4.1 (patch applies 
with fuzz) since that's what we happen to be using at the moment. Started VM, 
then ran 'curl http://10.1.1.1/latest/instance-id', then ran 
updateVirtualMachine, setting userdata, and pulled it on the guest via 'curl 
http://10.1.1.1/latest/user-data'.


Thanks,

Marcus Sorensen

Re: Trigger System VM Upgrade - without destroy and deploy

[Ian Service]

Yes! Shut them down and then *very quickly* (before they're automatically
restarted) you can issue the changeServiceForSystemVm API call to change
their Service Offering.

- Ian


On Fri, Jul 19, 2013 at 4:48 PM, Musayev, Ilya  wrote:

> Is there a way to trigger system VM upgrade without destroying and
> recreating?
>
> Reason behind this question, i don't want to mess up current system VMs as
> they have firewall rules bound to them and recreating them would mean new
> IP set and new firewall requests.
>
> Thanks
> ilya
>

Re: Reviewboard and jenkins

[Hugo Trippaers]

On Jul 19, 2013, at 7:15 PM, Daan Hoogland  wrote:

> good stuff, does it run the risk of running while another instance is busy?

Probably not, it's a single threaded script. I intend to use jenkins to 
schedule it and i can tell jenkins not to start another instance of a job once 
one is running.

There are probably some bug still in the script, but hey we got to start 
somewhere. ;-)

Cheers,

Hugo

> 
> 
> On Fri, Jul 19, 2013 at 6:57 PM, Prasanna Santhanam  wrote:
> 
>> On Fri, Jul 19, 2013 at 06:51:29PM +0200, Hugo Trippaers wrote:
>>> Hey all,
>>> 
>>> I've been working on some scripting to get patches submitted on review
>> board automagically tested with Jenkins.
>>> 
>>> The way it works is that a script will check if the jenkins user on
>>> review board (catofong) has commented on the review. If it hasn't it
>>> will submit a job to jenkins.cloudstack.org with a patch parameter.
>>> When the job it ready it will add a comment to the review with the
>>> result of the build and the url of that particular build. If the
>>> build succeeds cato will set the ship it flag to indicate that it's
>>> ok.
>>> 
>>> I'm finishing up the tests with this script now, so before i move it
>>> to jenkins.cloudstack.org i would like to feedback if we think this
>>> would use useful.
>>> 
>>> An result of the test run can be seen here:
>>> https://reviews.apache.org/r/12762/
>>> 
>>> Cheers,
>>> 
>>> Hugo
>> 
>> YES! YES! YES! and link it to the checkin tests coming up in the
>> `simulator` view on jenkins.cloudstack.org.
>> 
>> --
>> Prasanna.,
>> 
>> 
>> Powered by BigRock.com
>> 
>> 

Re: Review Request 12623: CLOUDSTACK-2328: Linux native VXLAN support on KVM hypervisor

[Chiradeep Vittal]

---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/12623/#review23523
---



scripts/vm/network/vnet/modifyvxlan.sh


I think there is a need to prevent the guest vm from spoofing the multicast?



scripts/vm/network/vnet/modifyvxlan.sh


What about vxlan with OVS, will this work?


I just wanted to make sure that you have tested your patch with regular VLANs 
as well. 
And, what the behavior will be when VxLAN is enabled in the zone, but only Xen 
/ VMW hypervisors are there
Also, some documentation on how the cloud operator can get this feature (KVM 
version/Which version of Centos/Ubuntu/etc), configuration of switches, 
bridges, etc would be useful.

- Chiradeep Vittal


On July 17, 2013, 11:54 p.m., Toshiaki Hatano wrote:
> 
> ---
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/12623/
> ---
> 
> (Updated July 17, 2013, 11:54 p.m.)
> 
> 
> Review request for cloudstack, Alena Prokharchyk, Chiradeep Vittal, Murali 
> Reddy, Hugo Trippaers, and Sheng Yang.
> 
> 
> Bugs: https://issues.apache.org/jira/browse/CLOUDSTACK-2328
> 
> 
> Repository: cloudstack-git
> 
> 
> Description
> ---
> 
> CLOUDSTACK-2328: Linux native VXLAN support on KVM hypervisor
> 
> Initial patch for VXLAN support.
> Fully functional, hopefully, for GuestNetwork - AdvancedZone.
> 
> Patch Note:
>  in cloudstack-server
> - Add isolation method VXLAN
> - Add VxlanGuestNetworkGuru as plugin for VXLAN isolation
> - Modify NetworkServiceImpl to handle extended vNet range for VXLAN isolation
> - Add VXLAN isolation option in zoneWizard UI
> 
>  in cloudstack-agent (kvm)
> - Add modifyvxlan.sh script that handle bridge/vxlan interface manipulation 
> script
> -- Usage is exactly same to modifyvlan.sh
> - BridgeVifDriver will call modifyvxlan.sh instead of modifyvlan.sh when 
> VXLAN is used for isolation
> 
> Database changes:
> - No change in database structure.
> - VXLAN isolation uses same tables that VLAN uses to store vNet allocation 
> status.
> 
> Known Issue:
> - Some resource still says 'VLAN' in log even if VXLAN is used
> - in UI, "Network - GuestNetworks" dosen't display VNI
> -- VLAN ID field displays "N/A"
> 
> 
> Diffs
> -
> 
>   api/src/com/cloud/network/Networks.java 5aede05 
>   api/src/com/cloud/network/PhysicalNetwork.java f6cb1a6 
>   client/pom.xml 32ab94a 
>   client/tomcatconf/componentContext.xml.in 1fbec61 
>   
> plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/BridgeVifDriver.java
>  195cf40 
>   
> plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/LibvirtComputingResource.java
>  da86612 
>   plugins/network-elements/vxlan/pom.xml PRE-CREATION 
>   
> plugins/network-elements/vxlan/src/com/cloud/network/guru/VxlanGuestNetworkGuru.java
>  PRE-CREATION 
>   
> plugins/network-elements/vxlan/test/com/cloud/network/guru/VxlanGuestNetworkGuruTest.java
>  PRE-CREATION 
>   plugins/pom.xml 261e8e8 
>   scripts/vm/network/vnet/modifyvlan.sh 8ed3905 
>   scripts/vm/network/vnet/modifyvxlan.sh PRE-CREATION 
>   server/src/com/cloud/network/NetworkManagerImpl.java f6e9a0a 
>   server/src/com/cloud/network/NetworkServiceImpl.java ccd23bf 
>   ui/scripts/ui-custom/zoneWizard.js 877dbc0 
> 
> Diff: https://reviews.apache.org/r/12623/diff/
> 
> 
> Testing
> ---
> 
> #) Test set up
> - Components
>   - 1x management server
>   - 1x nfs storage
>   - 3x Linux KVM host
>   -- CentOS 6.4 based
>   -- Replace kernel by version 3.8.13, VXLAN kernel module built as loadable 
> module
>   -- Replace iproute2 by version iproute2-ss130430
>   -- BridgeVifDriver (Default)
> 1. create advanced zone from zone wizard without security group option
> 2. select hypervisor: KVM
> 3. assign Guest network to separated physical network, isolated by VXLAN
>specify bridge name (traffic label) for Guest network, this bridge should 
> have IPv4 address (global/private both are OK).
> 4. assign Guest vNet range 1-2
> 5. other parameter are normal
> 6. add 2 more hosts into same zone/pod/cluster after zone wizard is finished
> 
> #) Test case 1: start/stop VR
> 1. Create network offering, same configuration as 
> DefaultIsolatedNetworkOfferingWithSourceNatService but persistent
> 2. Create network with network offering which is created in step 0
> 3. Confirm VR is started and bridge/vxlan device created on host
> 4. Delete network which is created in step 1
> 5. Confirm VR is deleted and bridge/vxlan device deleted on host
> 
> #) Test case 2: start/stop an instance (VR is on same host)
> 1. Add an instance from UI, create network during wizard.
> 2. Confirm VM and VR are on the same host
> 3.

storage.max.volume.size

[Damien Mannix]

Hi,

I have to present a 3TB Volume, I add it as a primary storage add the 
appropriate tags, add an offering. When customer tries to attach it it gets 
stuck in creating state.

I have set the value in global settings to 5000GB. Is there a limit to the size 
of a volume?

Input appreciated.

Thanks

Damien

RE: [DISCUSS] Upgrade path to ACS 4.2 from CCP

[Animesh Chaturvedi]

> -Original Message-
> From: Chip Childers [mailto:chip.child...@sungard.com]
> Sent: Friday, July 19, 2013 9:39 AM
> To: dev@cloudstack.apache.org
> Subject: Re: [DISCUSS] Upgrade path to ACS 4.2 from CCP
> 
> On Thu, Jul 18, 2013 at 10:01:02PM +, Animesh Chaturvedi wrote:
> > [Animesh>] Yes this seems late for 4.2, but it makes it easier to add
> support now. If there is objection we can track it for 4.3.
> 
> Are you sure it's easier to add support now?  That basically puts us in
> a position where it's *possible* to attempt it.  That means we have to
> test it, because if things go wrong then we are in a mess for 4.3.
> 
> Am I getting it wrong?
[Animesh>] Yes it is one upgrade path from 3.0.7 to 4.1 and follow through same 
upgrade path from 4.1 to 4.2 with couple of SQL files only. . Upgrade from 
3.0,4-> 3.0.7 has already been tested in CCP releases and becomes the 
checkpoint. It was planned to be tested for CCP anyway. Sudha do you see it as 
a concern. Again if community thinks it is late, we can track for 4.3

Re: Reviewboard and jenkins

[Prasanna Santhanam]

On Fri, Jul 19, 2013 at 06:51:29PM +0200, Hugo Trippaers wrote:
> Hey all,
> 
> I've been working on some scripting to get patches submitted on review board 
> automagically tested with Jenkins.
> 
> The way it works is that a script will check if the jenkins user on
> review board (catofong) has commented on the review. If it hasn't it
> will submit a job to jenkins.cloudstack.org with a patch parameter.
> When the job it ready it will add a comment to the review with the
> result of the build and the url of that particular build. If the
> build succeeds cato will set the ship it flag to indicate that it's
> ok.
> 
> I'm finishing up the tests with this script now, so before i move it
> to jenkins.cloudstack.org i would like to feedback if we think this
> would use useful.
> 
> An result of the test run can be seen here:
> https://reviews.apache.org/r/12762/
> 
> Cheers,
> 
> Hugo

YES! YES! YES! and link it to the checkin tests coming up in the
`simulator` view on jenkins.cloudstack.org. 

-- 
Prasanna.,


Powered by BigRock.com

Re: Object Store Region Assignment

[Chip Childers]

Well, you can add / list the other regions.  I'm still trying to get my
head around why, but honestly I think it's more of a future looking
thing.

On Fri, Jul 19, 2013 at 04:39:14PM -0400, John Burwell wrote:
> Chip,
> 
> Why do we have a region entity in the database if everything is assumed to be 
> one and only region?
> 
> Thanks,
> -John
> 
> On Jul 19, 2013, at 4:34 PM, Chip Childers  wrote:
> 
> > On Fri, Jul 19, 2013 at 4:31 PM, John Burwell  wrote:
> > 
> >> Chip,
> >> 
> >> So two regions won't share the same database?
> >> 
> > 
> > Nope, that's why you have to use something (not provided yet) to sync the
> > acct / domain metadata between them.
> > 
> > 
> >> 
> >> Thanks,
> >> -John
> >> 
> >> On Jul 19, 2013, at 4:06 PM, Chip Childers 
> >> wrote:
> >> 
> >>> On Fri, Jul 19, 2013 at 03:46:36PM -0400, John Burwell wrote:
>  All,
>  
>  In reviewing the 4.2 object store enhancements, object stores are
> >> defined with region scope.  However, there is no way to assign an object
> >> store definition to a region.  To my way of thinking, we must be able to
> >> assign an object store to a region to avoid transfer of large assets across
> >> WAN links.  Furthermore, it greatly reduces availability of the object
> >> store cluster because the management server/ssvm may attempt to connect to
> >> an object store over a WAN link when co-located instance is available.  Why
> >> isn't assignment of an object store to a region supported?
>  
>  Thanks,
>  -John
> >>> 
> >>> AFAIK, a management server is only a control point for a single region.
> >>> The Region entity is used to provide users with a reference to other
> >>> regions that may be part of the same service.  If I'm right, then the
> >>> locally configured object-store would be for that region only.
> >>> 
> >>> Other feel free to correct me.
> >> 
> >> 
> >> 
> 
>