DISCUSS : Vmware to Cloudstack migration support

[Shreya Nair]

Hello,

My team is working on a Vmware to Cloudstack migration task. The CloudStack
environment is set up on the Xenserver hypervisor.

We are able to export a VM on vsphere to OVF template which also provides
the OVF, VMDK disk image(s) and MF file for the particular VM.
In order to create a CloudStack Template from these files, it can be done
by 2 methods, namely, template creation from VHD(as the underlying
hypervisor is XenServer) or template creation from ISO.

VHD Template creation:
 - The VMDK file (Obtained after export OVF operation of the VM) is
converted to VHD format with VBoxManage.exe tool that is available with
VirtualBox.
 -  The VHD file can be used to create a CloudStack template and
instantiate a VM.


ISO Template creation:
 - The OVF file (Obtained after export OVF operation of the VM) is imported
to Citrix XenCenter.
 - XenCenter runs Operating system Fixup on the imported OVF file and
creates a modified ISO file.
 - The modified ISO file can be used for creating a template and
instantiate a VM

However, the newly created VM has booting issues as it requires "You might
have to change the root from /dev/hd[a-d] to /dev/xvd[a-d]"
Is there an alternate way or support to migrate a vSphere VM to CloudStack?
Or a method to change the partitioning on the hard disk from sd[a-d]  to
xvd[a-d]



Thanks & Regards,

Shreya

Re: DISCUSS : Vmware to Cloudstack migration support

[Sateesh Chodapuneedi]

From: Shreya Nair 
Reply-To: "us...@cloudstack.apache.org" 
Date: Tuesday, 4 July 2017 at 4:52 PM
To: "us...@cloudstack.apache.org" , 
"dev@cloudstack.apache.org" , 
"iss...@cloudstack.apache.org" 
Cc: Vinay Patil , Siddheshwar More 

Subject: DISCUSS : Vmware to Cloudstack migration support

Hello,

My team is working on a Vmware to Cloudstack migration task. The CloudStack
environment is set up on the Xenserver hypervisor.

We are able to export a VM on vsphere to OVF template which also provides
the OVF, VMDK disk image(s) and MF file for the particular VM.
In order to create a CloudStack Template from these files, it can be done
by 2 methods, namely, template creation from VHD(as the underlying
hypervisor is XenServer) or template creation from ISO.

VHD Template creation:
- The VMDK file (Obtained after export OVF operation of the VM) is
converted to VHD format with VBoxManage.exe tool that is available with
VirtualBox.
-  The VHD file can be used to create a CloudStack template and
instantiate a VM.


ISO Template creation:
- The OVF file (Obtained after export OVF operation of the VM) is imported
to Citrix XenCenter.
- XenCenter runs Operating system Fixup on the imported OVF file and
creates a modified ISO file.
- The modified ISO file can be used for creating a template and
instantiate a VM

However, the newly created VM has booting issues as it requires "You might
have to change the root from /dev/hd[a-d] to /dev/xvd[a-d]"
Is there an alternate way or support to migrate a vSphere VM to CloudStack?
Or a method to change the partitioning on the hard disk from sd[a-d]  to
xvd[a-d]

Hi Shreya,
Did you remove VMware tools from the VM before exporting?
Try installing xentools inside the guest, that should automatically name the 
disks to xvd*

Regards,
Sateesh



Thanks & Regards,

Shreya

DISCLAIMER
==
This e-mail may contain privileged and confidential information which is the 
property of Accelerite, a Persistent Systems business. It is intended only for 
the use of the individual or entity to which it is addressed. If you are not 
the intended recipient, you are not authorized to read, retain, copy, print, 
distribute or use this message. If you have received this communication in 
error, please notify the sender and delete all copies of this message. 
Accelerite, a Persistent Systems business does not accept any liability for 
virus infected mails.

Re: CS Collab Conf Brazil page is down

[Rafael Weingärtner]

Most people would not need Visas, and that is why I find it so interesting.

For instance, Latin and central Americans do not need a visa for
visiting/tourism; people from europeans country also do not need.

Here you have the complete list:
http://www.portalconsular.itamaraty.gov.br/images/qgrv/QGRV-simples-ing-08-12-16.pdf



Find your country in the table that starts on page 2 and check the
corresponding code in column “VITUR (Tourism)”. The caption for the codes
is listed on page 1. The most common codes are 1 (visa is required) and 8
(visa exemption for staying as a tourist for a period of at most 90 days)



I am going to an event in Brazil this month, and for the guys that needed a
Visa, it took between 24-72 HOURS to finish the process and receive the
visa stamp in the passport (this happened in Montreal, I do not know how
other consulates would work, but they all should follow the same protocol).



I also think that organizing something for 2018 is safer. Then, we only
need to decide the time frame and check if we have a quorum to start
organizing everything. The university has a big body of technological
courses from computer science to Engineerings (electrical, mechanical,
automation and so forth) that would probably attend, but I do believe that
contact with you guys (active PMCs and Committers) would make all the
difference.

On Tue, Jul 4, 2017 at 1:22 AM, Gabriel Beims Bräscher  wrote:

> Hi Mike, that is great that you are interested in the CloudStack Collab
> Conf in Brazil.
>
> Unfortunately, I don't know how long it takes to get a Brazilian VISA.
>
> I agree that August-September of 2017 might be too close. I think that 2018
> is safer, so we all can organize and make sure that most can attend. We
> think that March-April or August-September are interesting periods, but we
> are open for dates that suit better with the community schedule.
>
> Thanks for the feedback!
>
> 2017-07-04 1:54 GMT-03:00 Tutkowski, Mike :
>
> > Hi Gabriel,
> >
> > I, for one, would be quite interested in a CloudStack Collab Conf in
> > Brazil.
> >
> > I like your idea of a three-day conference going from a Tuesday through a
> > Thursday.
> >
> > I know many of us would require a VISA to enter Brazil (as well as a
> > passport, of course). I’m not sure how long getting a VISA typically
> takes,
> > so having the conference in August/September might be too soon at this
> > point. Correct me if I’m wrong – perhaps it doesn’t take long to get a
> > VISA. If it is a time-consuming process, perhaps we should instead
> consider
> > your March/April timeframe?
> >
> > Thanks for all the info on this!
> > Mike
> >
> > On 7/3/17, 9:15 PM, "Gabriel Beims Bräscher" 
> wrote:
> >
> > Hello all (users, dev and marketing lists),
> >
> >
> > After exercising the idea and studying some alternatives we (me,
> > Rafael and
> > Lucas) have something in mind to make the Apache CloudStack Collab
> > Conference possible in South America. However, before going forward
> > arranging an infrastructure to host the event, and investing our time
> > and
> > resources, we need the feedback of the community to see if there is
> > enough
> > interest in such event.
> >
> >
> > Sorry for the big e-mail. I hope that at least this proposal is clear
> > and
> > that you all embrace the discussion.
> >
> >
> >
> > What we are willing to do to help the event:
> >
> > (i) Hosting the conference. We can make an effort to use public
> > infrastructure from the Federal University of Santa Catarina [1] as
> > Rafael
> > previously said. However, to do so, this event CANNOT have any
> > commercial
> > appeal; otherwise, it is not allowed to use infrastructure and
> > resources
> > from the Federal University of Santa Catarina (Universidade Federal
> de
> > Santa Catarina – UFSC);
> >
> > (ii) promote the conference for the local public (researchers,
> > students, IT
> > professionals);
> >
> > (iii) propose topics and sessions aligned with South America/Brazil
> > context;
> >
> > (iv) we can suggest restaurants, hotels, and provide general
> > information
> > regarding the city and country;
> >
> >
> > Why host in the Federal University of Santa Catarina?
> >
> > (i) We want to evangelize CloudStack to new generations and we see
> > this as
> > an opportunity to show the Apache way to students, professors, and
> > academic
> > researchers;
> >
> > (ii) additionally, we do not have money to host such event. Using
> this
> > infrastructure would cut a great amount of costs and make this
> > conference
> > possible in South America;
> >
> > (iii) It is one of the best universities in South America, reaching
> > position number 9 among Brazilian universities and position number 25
> > in
> > the overall ranking of the QS University Rankings: Latin America
> 2016;
> >
> > (iv) Florianópolis economy is based on tourism and tech compa

Re: [DISCUSS] - Releases, Project Management & Funding Thereof

[Syed Ahmed]

As Paul said, in theory, running KVM as your base hypervisor in Trillain
shoud be possible. We have done nested KVM in the past with XenServer and
KVM as the nested hypervisors and with KVM being the base hypervisor. I am
not completely sure about how VMWare handles being in a nested environment.
Having said that, I believe if we get KVM and XenServer support with KVM as
being the base hypervisor, we will have a lot more adaptability for
Trillian. I will work with Paul and team on this.

As a side note, I have been working on getting to run integration testing
from a docker container. We need this because we require our tests to be
done on real hardware for cloud.ca. I really like the container approach as
it bundles all dependencies required by Marvin into a single container
which can be launched from any machine which has docker. This would hugely
benefit running it via Jenkins for example. I will open-source it as soon
as I am happy with it.



On Mon, Jul 3, 2017 at 2:34 PM, Will Stevens  wrote:

> I have added Syed to this.  He has done some initial review to get a port
> to KVM working, but I am not sure how far he got yet.
>
> *Will Stevens*
> CTO
>
> 
>
> On Mon, Jul 3, 2017 at 2:26 PM, Paul Angus 
> wrote:
>
>> I will take an action to look at trillian on KVM.  There's nothing
>> explicit or implicit in trillian itself that it requires vmware as long as
>> we can trunk the guest VLANs and virtualise the hypervisors.
>>
>> 
>> From: Will Stevens 
>> Sent: 3 Jul 2017 2:06 am
>> To: dev@cloudstack.apache.org
>> Subject: Re: [DISCUSS] - Releases, Project Management & Funding Thereof
>>
>> Sorry, I have been keeping up with these threads while on vacation at a
>> campsite.  :)  Finally back to a computer.
>>
>> Yes, ideally we would have more people actually committing code and
>> validating the PRs are ready for merge.  Right now, we have VERY limited
>> CI
>> setups, so we are bottlenecked on the ability to actually test the PRs in
>> a
>> timely fashion.  This leads to PRs sitting for a week at a time in some
>> 'phase' of the process.  This means that the developers get pushed off to
>> other items to pay the bills and it then causes a lag everywhere.  Because
>> of this, the RM has basically had to fill the role of making sure
>> everything is moving and understanding and unblocking the different PRs as
>> they move through the review/test/commit phases.
>>
>> Yes, we need more reviewers.  That is very true.
>>
>> Also true, is that we need more CI environments.  I would love to see
>> Trillian be able to be run on KVM on a developers laptop to at least test
>> the core components.  We could then start to standardize the dev/test
>> cycle
>> for developers so we can start focusing on a 'minimum support feature
>> set'.  We could also hopefully leverage the developers setups to run CI
>> overnight if they choose to participate (or at least their own PRs).  If
>> we
>> could standardize well enough to push the workload to the edge, I think we
>> would end up with more active rigs in the game.
>>
>> I personally feel that if we can put the CI on rails and standardize our
>> dev environment, a lot of our 'we need an full time RM' problems go away.
>> However, I do think we will need a full time RM for at least a year to be
>> able to shepherd that into existence though.
>>
>> *Will Stevens*
>> CTO
>>
>> 
>>
>>
>> paul.an...@shapeblue.com
>> www.shapeblue.com
>> 53 Chandos Place, Covent Garden, London  WC2N 4HSUK
>> @shapeblue
>>
>>
>>
>> On Sun, Jul 2, 2017 at 8:06 PM, Syed Ahmed  wrote:
>>
>> > Agree with Ron about this being a role of the commiter but in what I
>> have
>> > seen, it is mostly the RM who has to run around and ask for updates/make
>> > sure fixes are done.
>> >
>> > Part of the problem also is that there is a lack of reviewers. We've had
>> > some issues recently [1] which were code which was committed was not
>> > properly reviewed and later lead to problems. Having a RM and some core
>> > reviewers is essential to maintain quality and sanity of the release.
>> >
>> > I also agree with testing on a known setup with known parameters. The
>> > community can pool resources for hardware and Trillian can be used as
>> the
>> > CI framework. There was supposedly hadware donated by Citrix to the ASF.
>> > Anyone know what happened to that?
>> >
>> > [1] https://github.com/apache/cloudstack/pull/1834
>> >
>> >
>> > On Sun, Jul 2, 2017 at 9:55 AM, Ron Wheeler > > com>
>> > wrote:
>> >
>> > > I think you are describing the roles of all of the committers
>> > >
>> > > Is it normal at Apache for the RM to be doing all of this stuff?
>> > >
>> > > I would expect that the RM has a QC role in these activities but
>> others
>> > > are doing the work.
>> > >
>> > > Ron
>> > >
>> > >
>> > > On 01/07/2017 7:18 PM, Will Stevens wrote:
>> > >
>> > >> Oh, and if a system VM is touched, then you have to add in a new
>> system
>> 

Re: [DISCUSS] - Releases, Project Management & Funding Thereof

[Ivan Kudryavtsev]

Hello, guys.

What kind of hardware do you need actually to improve?
I host own DC and probably could give you our spare hardware connected to
dedicated managed switch. I can not guarantee the resources will be granted
and when the customer need it I'll take it, but we usually have 5-10 spare
servers with Xeon E5620, E3-1230.

Best wishes, Ivan.

2017-07-04 19:57 GMT+07:00 Syed Ahmed :

> As Paul said, in theory, running KVM as your base hypervisor in Trillain
> shoud be possible. We have done nested KVM in the past with XenServer and
> KVM as the nested hypervisors and with KVM being the base hypervisor. I am
> not completely sure about how VMWare handles being in a nested environment.
> Having said that, I believe if we get KVM and XenServer support with KVM as
> being the base hypervisor, we will have a lot more adaptability for
> Trillian. I will work with Paul and team on this.
>
> As a side note, I have been working on getting to run integration testing
> from a docker container. We need this because we require our tests to be
> done on real hardware for cloud.ca. I really like the container approach
> as
> it bundles all dependencies required by Marvin into a single container
> which can be launched from any machine which has docker. This would hugely
> benefit running it via Jenkins for example. I will open-source it as soon
> as I am happy with it.
>
>
>
> On Mon, Jul 3, 2017 at 2:34 PM, Will Stevens 
> wrote:
>
> > I have added Syed to this.  He has done some initial review to get a port
> > to KVM working, but I am not sure how far he got yet.
> >
> > *Will Stevens*
> > CTO
> >
> > 
> >
> > On Mon, Jul 3, 2017 at 2:26 PM, Paul Angus 
> > wrote:
> >
> >> I will take an action to look at trillian on KVM.  There's nothing
> >> explicit or implicit in trillian itself that it requires vmware as long
> as
> >> we can trunk the guest VLANs and virtualise the hypervisors.
> >>
> >> 
> >> From: Will Stevens 
> >> Sent: 3 Jul 2017 2:06 am
> >> To: dev@cloudstack.apache.org
> >> Subject: Re: [DISCUSS] - Releases, Project Management & Funding Thereof
> >>
> >> Sorry, I have been keeping up with these threads while on vacation at a
> >> campsite.  :)  Finally back to a computer.
> >>
> >> Yes, ideally we would have more people actually committing code and
> >> validating the PRs are ready for merge.  Right now, we have VERY limited
> >> CI
> >> setups, so we are bottlenecked on the ability to actually test the PRs
> in
> >> a
> >> timely fashion.  This leads to PRs sitting for a week at a time in some
> >> 'phase' of the process.  This means that the developers get pushed off
> to
> >> other items to pay the bills and it then causes a lag everywhere.
> Because
> >> of this, the RM has basically had to fill the role of making sure
> >> everything is moving and understanding and unblocking the different PRs
> as
> >> they move through the review/test/commit phases.
> >>
> >> Yes, we need more reviewers.  That is very true.
> >>
> >> Also true, is that we need more CI environments.  I would love to see
> >> Trillian be able to be run on KVM on a developers laptop to at least
> test
> >> the core components.  We could then start to standardize the dev/test
> >> cycle
> >> for developers so we can start focusing on a 'minimum support feature
> >> set'.  We could also hopefully leverage the developers setups to run CI
> >> overnight if they choose to participate (or at least their own PRs).  If
> >> we
> >> could standardize well enough to push the workload to the edge, I think
> we
> >> would end up with more active rigs in the game.
> >>
> >> I personally feel that if we can put the CI on rails and standardize our
> >> dev environment, a lot of our 'we need an full time RM' problems go
> away.
> >> However, I do think we will need a full time RM for at least a year to
> be
> >> able to shepherd that into existence though.
> >>
> >> *Will Stevens*
> >> CTO
> >>
> >> 
> >>
> >>
> >> paul.an...@shapeblue.com
> >> www.shapeblue.com
> >> 53 Chandos Place, Covent Garden, London  WC2N 4HSUK
> >> @shapeblue
> >>
> >>
> >>
> >> On Sun, Jul 2, 2017 at 8:06 PM, Syed Ahmed  wrote:
> >>
> >> > Agree with Ron about this being a role of the commiter but in what I
> >> have
> >> > seen, it is mostly the RM who has to run around and ask for
> updates/make
> >> > sure fixes are done.
> >> >
> >> > Part of the problem also is that there is a lack of reviewers. We've
> had
> >> > some issues recently [1] which were code which was committed was not
> >> > properly reviewed and later lead to problems. Having a RM and some
> core
> >> > reviewers is essential to maintain quality and sanity of the release.
> >> >
> >> > I also agree with testing on a known setup with known parameters. The
> >> > community can pool resources for hardware and Trillian can be used as
> >> the
> >> > CI framework. There was supposedly hadware donated by Citrix to the
> ASF.
> >> 

Solved: Silent failure of deploydb on developer machine

[Daan Hoogland]

For the interested: This is the behaviour on macos 10.12 with mysql
5.6. Downgrading mysql solves the problem

On Mon, Jul 3, 2017 at 5:14 PM, Daan Hoogland
 wrote:
> Has anybody outthere seen a
> mvn – P developer –pl developer –Ddeploydb
> report success without executing any sql commands?
> No create schema is not even executed.
>
> daan.hoogl...@shapeblue.com
> www.shapeblue.com
> 53 Chandos Place, Covent Garden, London  WC2N 4HSUK
> @shapeblue
>
>
>



-- 
Daan

Re: [VOTE] Apache CloudStack 4.10.0.0 RC7

[Rajani Karuturi]

Thanks Mike.

Is anyone else testing? We need atleast 3 binding votes to create
the Release.

Thanks,

~ Rajani

http://cloudplatform.accelerite.com/

On July 3, 2017 at 11:03 AM, Tutkowski, Mike
(mike.tutkow...@netapp.com) wrote:

+1

I examined the changes for this RC when compared to the previous
one and my positive vote remains.

On Jul 2, 2017, at 10:53 PM, Rajani Karuturi 
wrote:

Hi All,

I've created 4.10.0.0 release with the following artifacts up
for a vote:

Git Branch and Commit SH:
https://github.com/apache/cloudstack/commit/9d2893d44a3c3a4829be0964cc991272c1d13e4d
Commit:9d2893d44a3c3a4829be0964cc991272c1d13e4d
Branch:
https://github.com/apache/cloudstack/tree/4.10.0.0-RC20170703T1006

Source release (checksums and signatures are available at the
same
location):
https://dist.apache.org/repos/dist/dev/cloudstack/4.10.0.0/

SystemVm Templates:
http://download.cloudstack.org/systemvm/4.10/RC7/

PRs merged since RC4: #2150 and #2089
PRs merged since RC5: revert of #2084
PRs merged since RC6: #2162

PGP release keys (signed using CBB44821):
https://dist.apache.org/repos/dist/release/cloudstack/KEYS

Vote will be open for 72 hours.

For sanity in tallying the vote, can PMC members please be sure
to indicate
"(binding)" with their vote?

[ ] +1 approve
[ ] +0 no opinion
[ ] -1 disapprove (and reason why)

Thanks,
~Rajani
http://cloudplatform.accelerite.com/

Re: [VOTE] Apache CloudStack 4.10.0.0 RC7

[Tutkowski, Mike]

For the sake of clarity, I should say +1 (binding).

> On Jul 4, 2017, at 11:07 PM, Rajani Karuturi  wrote:
> 
> Thanks Mike.
> 
> Is anyone else testing? We need atleast 3 binding votes to create
> the Release.
> 
> Thanks,
> 
> ~ Rajani
> 
> http://cloudplatform.accelerite.com/
> 
> On July 3, 2017 at 11:03 AM, Tutkowski, Mike
> (mike.tutkow...@netapp.com) wrote:
> 
> +1
> 
> I examined the changes for this RC when compared to the previous
> one and my positive vote remains.
> 
> On Jul 2, 2017, at 10:53 PM, Rajani Karuturi 
> wrote:
> 
> Hi All,
> 
> I've created 4.10.0.0 release with the following artifacts up
> for a vote:
> 
> Git Branch and Commit SH:
> https://github.com/apache/cloudstack/commit/9d2893d44a3c3a4829be0964cc991272c1d13e4d
> Commit:9d2893d44a3c3a4829be0964cc991272c1d13e4d
> Branch:
> https://github.com/apache/cloudstack/tree/4.10.0.0-RC20170703T1006
> 
> Source release (checksums and signatures are available at the
> same
> location):
> https://dist.apache.org/repos/dist/dev/cloudstack/4.10.0.0/
> 
> SystemVm Templates:
> http://download.cloudstack.org/systemvm/4.10/RC7/
> 
> PRs merged since RC4: #2150 and #2089
> PRs merged since RC5: revert of #2084
> PRs merged since RC6: #2162
> 
> PGP release keys (signed using CBB44821):
> https://dist.apache.org/repos/dist/release/cloudstack/KEYS
> 
> Vote will be open for 72 hours.
> 
> For sanity in tallying the vote, can PMC members please be sure
> to indicate
> "(binding)" with their vote?
> 
> [ ] +1 approve
> [ ] +0 no opinion
> [ ] -1 disapprove (and reason why)
> 
> Thanks,
> ~Rajani
> http://cloudplatform.accelerite.com/

Re: [VOTE] Apache CloudStack 4.10.0.0 RC7

[Wido den Hollander]

I will test today

> Op 5 jul. 2017 om 07:13 heeft Tutkowski, Mike  het 
> volgende geschreven:
> 
> For the sake of clarity, I should say +1 (binding).
> 
>> On Jul 4, 2017, at 11:07 PM, Rajani Karuturi  wrote:
>> 
>> Thanks Mike.
>> 
>> Is anyone else testing? We need atleast 3 binding votes to create
>> the Release.
>> 
>> Thanks,
>> 
>> ~ Rajani
>> 
>> http://cloudplatform.accelerite.com/
>> 
>> On July 3, 2017 at 11:03 AM, Tutkowski, Mike
>> (mike.tutkow...@netapp.com) wrote:
>> 
>> +1
>> 
>> I examined the changes for this RC when compared to the previous
>> one and my positive vote remains.
>> 
>> On Jul 2, 2017, at 10:53 PM, Rajani Karuturi 
>> wrote:
>> 
>> Hi All,
>> 
>> I've created 4.10.0.0 release with the following artifacts up
>> for a vote:
>> 
>> Git Branch and Commit SH:
>> https://github.com/apache/cloudstack/commit/9d2893d44a3c3a4829be0964cc991272c1d13e4d
>> Commit:9d2893d44a3c3a4829be0964cc991272c1d13e4d
>> Branch:
>> https://github.com/apache/cloudstack/tree/4.10.0.0-RC20170703T1006
>> 
>> Source release (checksums and signatures are available at the
>> same
>> location):
>> https://dist.apache.org/repos/dist/dev/cloudstack/4.10.0.0/
>> 
>> SystemVm Templates:
>> http://download.cloudstack.org/systemvm/4.10/RC7/
>> 
>> PRs merged since RC4: #2150 and #2089
>> PRs merged since RC5: revert of #2084
>> PRs merged since RC6: #2162
>> 
>> PGP release keys (signed using CBB44821):
>> https://dist.apache.org/repos/dist/release/cloudstack/KEYS
>> 
>> Vote will be open for 72 hours.
>> 
>> For sanity in tallying the vote, can PMC members please be sure
>> to indicate
>> "(binding)" with their vote?
>> 
>> [ ] +1 approve
>> [ ] +0 no opinion
>> [ ] -1 disapprove (and reason why)
>> 
>> Thanks,
>> ~Rajani
>> http://cloudplatform.accelerite.com/