Re: [CentOS] Multiple Virtual SSL sites with Apache httpd on CentOS 5

[James Hogarth]

> > Serving multiple ssl certificates requires SNI on server and also
> > client support.
> >
> >
http://www.rackspace.com/knowledge_center/article/serving-secure-sites-with-sni-on-apache
>
> this works on Centos 6 without messing up openssl libraries :)

Take a look at this freeIPA how to I wrote a while back...

http://freeipa.org/page/Apache_SNI_With_Kerberos

Windows XP does not support SNI on its internal libraries (so no IE on
that) and I'm not 100% sure when MacOSX enabled it.

It worked for subversion but not with older svnkit bindings so do some
testing and ensure that all the clients you expect will support SNI ...

For me it was an interesting experiment (wiki, svn, etc on same IP) but
they drawbacks out weighed the benefits when I opened it up to the
developers to use so switched back to sperate IPs.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] How to limit upload bandwidth for a KVM guest?

[James Hogarth]

> Hello:
>
> I am using a CentOS server as a KVM host using bridged networking
> for the guest servers.
>
> I would like to limit the bandwidth available to a guest.  I tried using
> the tc command, but it is only limiting the download speed to the
> guest.
>
> Is there a way to limit the upload speed from the guests?

Hmm interesting...

I've not had a need to do this but in theory cgroup configuration might be
able to help.

Is this c5 or c6?
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] CentOS 4 kernel panic error

[rajaranjan]

Hi,
I tried to install CentOS 4.1(x86_64) on my virtual machine.
I am getting kernel panic error. Installation program crashed.
I also tried CentOS 4.0 and CentOS 4.2, got same error.
There is  no  problem  with  installing  CentOS 4.1(32  bits).

Can you help me regarding this?
What is the more stable version of centOS 4.

Thanks & Regards,
Raja
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] CentOS 4 kernel panic error

[Nicolas Thierry-Mieg]

rajaranjan wrote:
> Hi,
> I tried to install CentOS 4.1(x86_64) on my virtual machine.
> I am getting kernel panic error. Installation program crashed.
> I also tried CentOS 4.0 and CentOS 4.2, got same error.
> There is  no  problem  with  installing  CentOS 4.1(32  bits).
>
> Can you help me regarding this?
> What is the more stable version of centOS 4.

read this:
http://lists.centos.org/pipermail/centos-announce/2012-February/018462.html
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] skype not starting anymore, prelink issue?

[wwp]

Hello Yves,


On Fri, 1 Feb 2013 08:08:20 -0500 Yves Bellefeuille 
wrote:

> On Friday 01 February 2013, wwp  wrote:
> 
> > This point has already been answered on this mailing list (and
> > elsewhere). A bit of search in the archives and elsewhere would
> > quickly bring you this:
> > http://wiki.centos.org/HowTos/Skype
> 
> I'm very familiar with that document. :-) And many users, including 
> myself, couldn't get Skype 4.1 to work on CentOS 64 bits using those 
> instructions.
> 
> If Skype 4.1 did work on your 64-bit system, I'm very interested in 
> knowing the details; you can contact me directly if you wish.
> 
> (Skype 4.0 does work; in your initial post, you said that 4.1.0.20 used 
> to work on your system, but no longer does.)

Got skype 4.1 to work again here, just unpacked the binary archive over
the existing files, don't know what went broken or how..

FYI, I've made a list of the packages showing from a `ldd skype`, this
may indicate what you need to install, at least (here: CentOS6.3 up-to-date,
epel repo enabled):

 alsa-lib-1.0.22-3.el6.i686
 dbus-libs-1.2.24-7.el6_3.i686
 expat-2.0.1-11.el6_2.i686
 flac-1.2.1-6.1.el6.i686
 fontconfig-2.8.0-3.el6.i686
 freetype-2.3.11-14.el6_3.1.i686
 glib2-2.22.5-7.el6.i686
 glibc-2.12-1.80.el6_3.7.i686
 keyutils-libs-1.4-4.el6.i686
 krb5-libs-1.9-33.el6_3.3.i686
 libasyncns-0.8-1.1.el6.i686
 libcom_err-1.41.12-12.el6.i686
 libgcc-4.4.6-4.el6.i686
 libICE-1.0.6-1.el6.i686
 libogg-1.1.4-2.1.el6.i686
 libpng-1.2.49-1.el6_2.i686
 libselinux-2.0.94-5.3.el6.i686
 libSM-1.1.0-7.1.el6.i686
 libsndfile-1.0.20-5.el6.i686
 libstdc++-4.4.6-4.el6.i686
 libuuid-2.17.2-12.7.el6_3.i686
 libvorbis-1.2.3-4.el6_2.1.i686
 libX11-1.3-2.el6.i686
 libXau-1.0.5-1.el6.i686
 libxcb-1.5-1.el6.i686
 libXcursor-1.1.10-2.el6.i686
 libXext-1.1-3.el6.i686
 libXfixes-4.0.4-1.el6.i686
 libXi-1.3-3.el6.i686
 libXinerama-1.1-1.el6.i686
 libXrandr-1.3.0-4.el6.i686
 libXrender-0.9.5-1.el6.i686
 libXScrnSaver-1.2.0-1.el6.i686
 libXtst-1.0.99.2-3.el6.i686
 libXv-1.0.5-1.el6.i686
 openssl-1.0.0-25.el6_3.1.i686
 pulseaudio-libs-0.9.21-14.el6_3.i686
 pulseaudio-libs-glib2-0.9.21-14.el6_3.i686
 sqlite-3.6.20-1.el6.i686
 tcp_wrappers-libs-7.6-57.el6.i686
 zlib-1.2.3-27.el6.i686

Of course there may have indirect dependencies, but yum would tell you
that in time.

Important: I also have a full Qt 4.7 32-bit in /usr/lib/qt47, on which
my installed skype version depends on:
 libphonon.so.4
 libQtCore.so.4
 libQtDBus.so.4
 libQtGui.so.4
 libQtNetwork.so.4
 libQtWebKit.so.4
 libQtXml.so.4

I was not sure if I got this specific Qt version installed for
GuitarPro6 (commercial) or skype, but my LD_LIBRARY_PATH points
to that dir and it can't using the Qt 4.6 or 4.8 versions installed
locally (install from one of those repositories: base, epel, rpmforge,
puias, atrpms).

For this Qt47, I downloaded the following RPMs:
 phonon-backend-gstreamer-4.7.2-1_18.el6.i686.rpm
 qt47-4.7.2-1_18.el6.i686.rpm
 qt47-config-4.7.2-1_18.el6.i686.rpm
 qt47-webkit-4.7.2-1_18.el6.i686.rpm
 qt47-x11-4.7.2-1_18.el6.i686.rpm

And installed *manually* their contents to /usr/lib/qt47.

This also requires tweaking a bit in /usr/lib/qt47/bin and /usr/bin
(it's safe towards system Qt versions, it's just about fixing the missing q* 
qt47 apps and making correct symlinks isolated in /usr/lib/qt47/bin):
See my /usr/lib/qt47/bin contents:
 qdbus -> ../../../bin/qdbus-qt47
 qdbusviewer -> ../../../bin/qdbusviewer-qt47
 qmlviewer -> ../../../bin/qmlviewer-qt47
 qtconfig -> ../../../bin/qtconfig-qt47
 qtconfig-qt47 -> ../../../bin/qtconfig-qt47


Let me know if you succeed or still miss something!


Regards,

-- 
wwp


signature.asc
Description: PGP signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] skype not starting anymore, prelink issue?

[wwp]

Hello,


On Mon, 4 Feb 2013 10:23:42 +0100 wwp  wrote:

[snip]
> For this Qt47, I downloaded the following RPMs:
>  phonon-backend-gstreamer-4.7.2-1_18.el6.i686.rpm
>  qt47-4.7.2-1_18.el6.i686.rpm
>  qt47-config-4.7.2-1_18.el6.i686.rpm
>  qt47-webkit-4.7.2-1_18.el6.i686.rpm
>  qt47-x11-4.7.2-1_18.el6.i686.rpm
> 
> And installed *manually* their contents to /usr/lib/qt47.
> 
> This also requires tweaking a bit in /usr/lib/qt47/bin and /usr/bin
> (it's safe towards system Qt versions, it's just about fixing the missing q* 
> qt47 apps and making correct symlinks isolated in /usr/lib/qt47/bin):
> See my /usr/lib/qt47/bin contents:
>  qdbus -> ../../../bin/qdbus-qt47
>  qdbusviewer -> ../../../bin/qdbusviewer-qt47
>  qmlviewer -> ../../../bin/qmlviewer-qt47
>  qtconfig -> ../../../bin/qtconfig-qt47
>  qtconfig-qt47 -> ../../../bin/qtconfig-qt47
> 
> 
> Let me know if you succeed or still miss something!

I missed to say where I got the qt47-el6 RPMs from:
  http://pkgs.org/search/?keyword=qt47


Regards,

-- 
wwp


signature.asc
Description: PGP signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] CentOS 4 kernel panic error

[John R Pierce]

On 2/4/2013 1:17 AM, rajaranjan wrote:
> Hi,
> I tried to install CentOS 4.1(x86_64) on my virtual machine.
> I am getting kernel panic error. Installation program crashed.
> I also tried CentOS 4.0 and CentOS 4.2, got same error.
> There is  no  problem  with  installing  CentOS 4.1(32  bits).
>

4.0, 4.1, and 4.2 were released a few months apart in 2005, built with 
kernel 2.6.9.   why are you installing 8 year old software on a new system ?

if you absolutely have to run a EL4 derivative, at least run 4.9, which 
was released in 2011, but be aware there have been no security fixes for 
it since a year ago.


-- 
john r pierce  37N 122W
somewhere on the middle of the left coast

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Traffic shaping problem

[John Doe]

From: Bent Terp 

> I tried last week to do traffic shaping on a production system,
> object of the exercise was simply to throttle the outgoing traffic.
> tc qdisc add dev eth4 root tbf rate 300mbit burst 300kb latency 50ms
> But the server became rather instable, crashing repeatedly without
> anything in the logs.
> Can anybody spot glaring mistakes in the tc command above,
> or tell me what I should have done instead?

I am no tc expert so last time I tested I had some success with the cbq.init 
script...
There are other (more recent?) ones here:
http://linux-ip.net/articles/Traffic-Control-HOWTO/scripts.html

JD
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] CentOs 6 DHCP Server and virtual interface

[Johnny Hughes]

On 02/03/2013 08:11 AM, Grzegorz Sołtys wrote:
> So, if my WAN IP adress is 83.238.*.* then i need to run DHCP server
> in the same class of if address. But, if got only this address and i
> will setup in eg: 83.238.55.10 to 15 range for eth0:0 and there is a
> machine with WAN address like 83.238.55.13 if gonna to make IP address
> colission? or then it will be avalible only as LAN Address for my server?
>

Yes, if you are going to serve dhcp to a subnet, the server has to have
an address on that subnet.  It will indeed need to take one of the
addresses of that subnet, and you can not have one of your other
machines get that address.

You can use the dhcp machine to do other things on that same address
though ... so it can do dhcp AND replace one of the other machines.


> W dniu 2013-02-03 14:50, Johnny Hughes pisze:
>> On 02/03/2013 07:38 AM, Grzegorz Sołtys wrote:
>>> I have made it, and the DHCP server won't start telling me:
>>>
>>> " Feb  3 13:42:21 vlan19 dhcpd: No subnet declaration for eth0:0 (no
>>> IPv4 addresse s).
>>> Feb  3 13:42:21 vlan19 dhcpd: ** Ignoring requests on eth0:0.  If this
>>> is not what
>>> Feb  3 13:42:21 vlan19 dhcpd:you want, please write a subnet
>>> declaration
>>> Feb  3 13:42:21 vlan19 dhcpd:in your dhcpd.conf file for the
>>> network segment
>>> Feb  3 13:42:21 vlan19 dhcpd:to which interface eth0:0 is
>>> attached. **
>>> "
>>>
>>> I have conifgured all files to use DHCPD on this interface.  Do you
>>> need all conifguraiton?
>> Inside your dhcpd.conf file you need a subnet declaration and the IP
>> Address of ETH0:0 has to be in that subnet.
>>
>> See:
>>
>> https://access.redhat.com/knowledge/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Deployment_Guide/sect-Configuring_a_Multihomed_DHCP_Server.html
>>
>>
>>
>> So, for example, if I have:
>>
>> subnet 192.168.0.16 mask 255.255.255.240 {
>> option 
>> option 
>> }
>>
>> Then eth0:0 would have to be inside the IP Range of the subnet ... in
>> this case, 192.168.0.16/255.255.255.240 is the IP addresses from
>> 192.168.0.16 through 192.168.0.31 (with 16 and 31 being not usable as
>> they are the network name and the broadcast address).  This would mean
>> that eth0:0's IP address would need to be somewhere in the range of
>> 192.168.0.17 and 192.168.0.30 to serve that subnet.
>>
>>> W dniu 2013-02-03 14:28, Johnny Hughes pisze:
 On 02/03/2013 06:38 AM, Grzegorz Sołtys wrote:
> Hello All
>
> I have looking for any specific answer for one thing.
>
> I have a virtualized Server with only one physical interface eth0
> (WAN).
> To run OpenVPN i need to use DHCP server. And here is the
> question: is
> there a chance to run DHCP server on eth0:0 interface? Or it is
> impossible ?
>
> Thanks in advance.
>
   From the EL6 Deployemtn Guide:
 https://access.redhat.com/knowledge/docs/en-US/Red_Hat_Enterprise_Linux/6/html/Deployment_Guide/sect-Configuring_a_Multihomed_DHCP_Server.html




 "If a system has three network interfaces cards -- eth0, eth1, and
 eth2
 -- and it is only desired that the DHCP daemon listens on eth0, then
 only specify eth0 in /etc/sysconfig/dhcpd"
 DHCPDARGS="eth0"

 So in your case, edit the file /etc/sysconfig/dhcpd and set:

 DHCPDARGS="eth0:0";





 ___
 CentOS mailing list
 CentOS@centos.org
 http://lists.centos.org/mailman/listinfo/centos
>>>
>>
>>
>>
>> ___
>> CentOS mailing list
>> CentOS@centos.org
>> http://lists.centos.org/mailman/listinfo/centos
>
>




signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] CentOs 6 DHCP Server and virtual interface

[Eliezer Croitoru]

On 2/3/2013 2:38 PM, Grzegorz Sołtys wrote:
>
> Hello All
>
> I have looking for any specific answer for one thing.
>
> I have a virtualized Server with only one physical interface eth0 (WAN).
> To run OpenVPN i need to use DHCP server. And here is the question: is
> there a chance to run DHCP server on eth0:0 interface? Or it is impossible ?
And why is that?
run dhcp server on any of the addresses block dhcp traffic on the eht0 
interface and use a tun\tap interface for openVPN...


>
> Thanks in advance.
>

-- 
Eliezer Croitoru
http://www1.ngtech.co.il
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Multiple Virtual SSL sites with Apache httpd on CentOS 5

[Harold Pritchett]

On 2/3/2013 11:47 PM, Bent Terp wrote:
> So my advice is to use a wildcard domain covering *.mycompany.com - but 
> remember that mycompany.com does not match *.mycompany.com, so you'll need a 
> redirect to 
> www.mycompany.com BR Bent PS Just realised that we did this on nginx not 
> apache, but I really don't expect that to be a problem. 
I should have mentioned that wild cards are not an option.  I need to support 
multiple names with absolutely nothing in common.  It's sorta like

www.mynethost.com
www.gotch.net
www.somewhere.else.org

you get the idea.

Hal

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Multiple Virtual SSL sites with Apache httpd on CentOS 5

[James Hogarth]

On 4 February 2013 14:20, Harold Pritchett  wrote:

> On 2/3/2013 11:47 PM, Bent Terp wrote:
> > So my advice is to use a wildcard domain covering *.mycompany.com - but
> remember that mycompany.com does not match *.mycompany.com, so you'll
> need a redirect to
> > www.mycompany.com BR Bent PS Just realised that we did this on nginx
> not apache, but I really don't expect that to be a problem.
> I should have mentioned that wild cards are not an option.  I need to
> support multiple names with absolutely nothing in common.  It's sorta like
>
> www.mynethost.com
> www.gotch.net
> www.somewhere.else.org
>
> you get the idea.
>

If you can use a single cert for these (ie owned by the same entity) you
could in principle use SubjectAltNames in a single cert to cover these and
use that same cert on each site - no wildcard is then required...

If you need to use different certs (ownership reasons and so on) then SNI
will work on a single IP - but just be aware of the limitations (all the
WinXP clients for example) ... especially if these are public sites.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] weird issue with qemu-kvm network...

[Natxo Asenjo]

On Sat, Feb 2, 2013 at 2:12 AM, James Hogarth  wrote:

> If you need bonding or vlans it becomes more complicated ;-)

LOL. We have been testing this (bonding *and* vlans).

Once you understand how it works it makes sense, but you need to
understand it first and most people researching this are probably used
to the fantastic vmware tooling that keeps it all hidden.

-- 
natxo
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] recent ruby packages?

[m . roth]

Does anyone know of a repository that's *trustworthy* (gotta worry 'bout
malware) with newer ruby rpm's than RHEL has?

OT: the more I deal with ruby, the less I like it. Someone here was ready
to move to a newer version, and from the ruby.org website, they're
apparently actively hostile to all RH-related distros, even though we're
the most common in North America. They've got a how to do it from debian
and arch, how to use their own installer, and, oh, yes, they say a lot of
their community feels you should build from source.

Sorry, that's not my idea of a stable language that I'd ever recommend to
someone

mark

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] How to limit upload bandwidth for a KVM guest?

[Neil Aggarwal]

James:

> Is this c5 or c6?

This is CentOS 6.

Neil

--
Neil Aggarwal, (972)834-1565, http://UnmeteredVPS.net/centos
Virtual private server with CentOS 6 preinstalled
Unmetered bandwidth = no overage charges


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Phil Dobbin]

On 02/04/2013 03:21 PM, m.r...@5-cent.us wrote:

> Does anyone know of a repository that's *trustworthy* (gotta worry 'bout
> malware) with newer ruby rpm's than RHEL has?
> 
> OT: the more I deal with ruby, the less I like it. Someone here was ready
> to move to a newer version, and from the ruby.org website, they're
> apparently actively hostile to all RH-related distros, even though we're
> the most common in North America. They've got a how to do it from debian
> and arch, how to use their own installer, and, oh, yes, they say a lot of
> their community feels you should build from source.
> 
> Sorry, that's not my idea of a stable language that I'd ever recommend to
> someone

I've used the Ruby Version Manager  for all things Ruby
for a few years now & can highly recommend it.

Stable, under constant development, very active community & Michal
Papis, who monitors & co-authors the project, is very quick to reply to
any problems/questions.

Several different ways to install & run Ruby are on offer & it's works
very well on CentOS.

Hope this helps,

Cheers,

  Phil...

-- 
currently (ab)using
CentOS 5.8 & 6.3, Debian Squeeze & Wheezy, Fedora Beefy & Spherical,
Lubuntu 12.10, OS X Snow Leopard & Ubuntu Precise & Quantal
GnuPG Key : http://www.horse-latitudes.co.uk/publickey.asc


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Alias for sending email messages from root

[James B. Byrne]

On Tue, January 29, 2013 17:37, lists-centos wrote:

>
> James -- Look at the MAILFROM bit in crontab(5).
>
>
> - Richard
>

That worked on CentOS-6 but does not on CentOS-5.  I have rearranged
things to run the tests solely from CentOS-6 boxes and using MAILFROM
has circumvented the problem.  Bell Canada is still 'working on a
solution'.

Thanks,

-- 
***  E-Mail is NOT a SECURE channel  ***
James B. Byrnemailto:byrn...@harte-lyne.ca
Harte & Lyne Limited  http://www.harte-lyne.ca
9 Brockley Drive  vox: +1 905 561 1241
Hamilton, Ontario fax: +1 905 561 0757
Canada  L8E 3C3

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] weird issue with qemu-kvm network...

[James Hogarth]

>
> LOL. We have been testing this (bonding *and* vlans).
>
> Once you understand how it works it makes sense, but you need to
> understand it first and most people researching this are probably used
> to the fantastic vmware tooling that keeps it all hidden.
>
>
>
One of the environments I built a year or so back needed this - 8
interfaces in LACP bonds per system with around 10-12 VLANs (( forget
specifically) on top of this and then bridged through to VMs ...

As you say - take it one layer at a time and it's simple enough... but
demoing the setup to people initially left them a little lost following the
various config files.
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] intel_iommu=on => No root device found

[Nils Caspar]

> I did some experimenting with and without the RAID card installed. There are 
> no DMAR errors, when the RAID card isn't present... Too bad!

Just got this confirmed by Adaptec Support:

> VT-D is not supported with our Adaptec RAID Controllers. You will need to 
> disable this in the BIOS of the mainboard in order for the system to run in a 
> stable way. This applies mainly to VMWARE but also to other linux 
> configuration as you have seen. 
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] CentOs 6 DHCP Server and virtual interface

[Les Mikesell]

On Mon, Feb 4, 2013 at 5:42 AM, Johnny Hughes  wrote:
> On 02/03/2013 08:11 AM, Grzegorz Sołtys wrote:
>> So, if my WAN IP adress is 83.238.*.* then i need to run DHCP server
>> in the same class of if address. But, if got only this address and i
>> will setup in eg: 83.238.55.10 to 15 range for eth0:0 and there is a
>> machine with WAN address like 83.238.55.13 if gonna to make IP address
>> colission? or then it will be avalible only as LAN Address for my server?
>>
>
> Yes, if you are going to serve dhcp to a subnet, the server has to have
> an address on that subnet.  It will indeed need to take one of the
> addresses of that subnet, and you can not have one of your other
> machines get that address.
>
> You can use the dhcp machine to do other things on that same address
> though ... so it can do dhcp AND replace one of the other machines.

Or slightly more convoluted, you could run dhcrelay on a machine that
has interfaces on both the subnet serving dhcp and the one where you
want it to be forwarded.   Not sure it would help in this scenario but
it works where the dhcp server is on some other machine.  You still
need the subnet declaration in the dhcp server for the forwarded
subnet.

-- 
  Les Mikesell
lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[John R Pierce]

On 2/4/2013 7:21 AM, m.r...@5-cent.us wrote:
> Does anyone know of a repository that's*trustworthy*  (gotta worry 'bout
> malware) with newer ruby rpm's than RHEL has?
>
> OT: the more I deal with ruby, the less I like it. Someone here was ready
> to move to a newer version, and from the ruby.org website, they're
> apparently actively hostile to all RH-related distros, even though we're
> the most common in North America. They've got a how to do it from debian
> and arch, how to use their own installer, and, oh, yes, they say a lot of
> their community feels you should build from source.
>
> Sorry, that's not my idea of a stable language that I'd ever recommend to
> someone

IMNSHO, Ruby is only suitable for prototyping and low volume uses. it 
doesn't scale well, and  ruby/rails websites perform abysmally under 
heavy workloads.

that said, the 'correct' way of dealing with something like this in the 
RHEL world is to build whatever version you need for your purposes, test 
it, and package it as your OWN rpm's for production deployment.



-- 
john r pierce  37N 122W
somewhere on the middle of the left coast

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[m . roth]

John R Pierce wrote:
> On 2/4/2013 7:21 AM, m.r...@5-cent.us wrote:
>> Does anyone know of a repository that's*trustworthy*  (gotta worry 'bout
>> malware) with newer ruby rpm's than RHEL has?
>>
>> OT: the more I deal with ruby, the less I like it. Someone here was
>> ready to move to a newer version, and from the ruby.org website, they're
>> apparently actively hostile to all RH-related distros, even though we're
>> the most common in North America. They've got a how to do it from debian
>> and arch, how to use their own installer, and, oh, yes, they say a lot
>> of their community feels you should build from source.
>>
>> Sorry, that's not my idea of a stable language that I'd ever recommend
>> to someone
>
> IMNSHO, Ruby is only suitable for prototyping and low volume uses. it
> doesn't scale well, and  ruby/rails websites perform abysmally under
> heavy workloads.
>
ROTFLMAO! A few years ago, a friend who's a professor (was that math, or
CS; think it was the latter) up in Minnesota was real hot on ruby, and
commented on the growning number of books on it in the school bookstore.

A year or two ago, I'd seen an article or two about it not scaling, and
sent it to him, which he thanked me for, and hadn't known about. This
isn't a heavily used website, AFAIK, even if it is from the US gov.
Certainly, they were building it in ruby before I started; dunno as I'd
have had any influence on the guy who was good, but enTHUsed about ruby...
and rails. And passenger.


   mark "or was that passenger pigeons, which are extinct?"


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Craig White]

On Feb 4, 2013, at 12:36 PM, John R Pierce wrote:

> On 2/4/2013 7:21 AM, m.r...@5-cent.us wrote:
>> Does anyone know of a repository that's*trustworthy*  (gotta worry 'bout
>> malware) with newer ruby rpm's than RHEL has?
>> 
>> OT: the more I deal with ruby, the less I like it. Someone here was ready
>> to move to a newer version, and from the ruby.org website, they're
>> apparently actively hostile to all RH-related distros, even though we're
>> the most common in North America. They've got a how to do it from debian
>> and arch, how to use their own installer, and, oh, yes, they say a lot of
>> their community feels you should build from source.
>> 
>> Sorry, that's not my idea of a stable language that I'd ever recommend to
>> someone
> 
> IMNSHO, Ruby is only suitable for prototyping and low volume uses. it 
> doesn't scale well, and  ruby/rails websites perform abysmally under 
> heavy workloads.
> 
> that said, the 'correct' way of dealing with something like this in the 
> RHEL world is to build whatever version you need for your purposes, test 
> it, and package it as your OWN rpm's for production deployment.

I don't normally quibble with your opinions but clearly you are dealing with 
anecdotal evidence (rather than first hand experience) of older versions.

Yes, twitter has moved some of the server engine to another platform but it's 
code base is still almost entirely RoR.

There are millions of popular, high trafficked web sites running RoR.

By the time you reach a point where scalability is a bottleneck (and all 
successful web sites do), you are already the next biggest thing and the 
options likely involve people whose technical skills that far exceed the 
original developer.

To the OP - If we are talking about CentOS 5.x and you are determined to use 
RPM packages, Google 'enterprise ruby' and install it (it's Ruby 1.8.7) It's 
not likely to get any more updates though. If you get off the need to have RPM 
packages, both rbenv & rvm install an alternate that downloads ruby source and 
compiles it for you and gives you sufficient shell modifications to make it 
appear somewhat seamless (I'm not promising the world here but it's not that 
difficult and my work has some CentOS 5.x still running enterprise-ruby-1.8.7 
and everything newer has been Ubuntu 10.04 and either uses enterprise-ruby for 
1.8.7 (becoming rare these days) and all new setups are rbenv and ruby 
1.9.3-pXXX

Craig
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Phil Dobbin]

On 02/04/2013 07:36 PM, John R Pierce wrote:

> On 2/4/2013 7:21 AM, m.r...@5-cent.us wrote:
>> Does anyone know of a repository that's*trustworthy*  (gotta worry 'bout
>> malware) with newer ruby rpm's than RHEL has?
>>
>> OT: the more I deal with ruby, the less I like it. Someone here was ready
>> to move to a newer version, and from the ruby.org website, they're
>> apparently actively hostile to all RH-related distros, even though we're
>> the most common in North America. They've got a how to do it from debian
>> and arch, how to use their own installer, and, oh, yes, they say a lot of
>> their community feels you should build from source.
>>
>> Sorry, that's not my idea of a stable language that I'd ever recommend to
>> someone
> 
> IMNSHO, Ruby is only suitable for prototyping and low volume uses. it 
> doesn't scale well, and  ruby/rails websites perform abysmally under 
> heavy workloads.
> 
> that said, the 'correct' way of dealing with something like this in the 
> RHEL world is to build whatever version you need for your purposes, test 
> it, and package it as your OWN rpm's for production deployment.

The doesn't scale well argument hasn't been the case for at least a few
years now. Twitter is just one example. Some of the busiest sites on teh
Interwebs are still using it.

There are also projects, for example, like Puppet that are written in
Ruby that are used by a lot of fairly large organisations.

It may be worth your while reappraising Ruby.

Cheers,

  Phil...

-- 
currently (ab)using
CentOS 5.8 & 6.3, Debian Squeeze & Wheezy, Fedora Beefy & Spherical,
Lubuntu 12.10, OS X Snow Leopard & Ubuntu Precise & Quantal
GnuPG Key : http://www.horse-latitudes.co.uk/publickey.asc


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[m . roth]

Craig White wrote:
> On Feb 4, 2013, at 12:36 PM, John R Pierce wrote:
>> On 2/4/2013 7:21 AM, m.r...@5-cent.us wrote:
>>> Does anyone know of a repository that's*trustworthy*  (gotta worry
>>> 'bout malware) with newer ruby rpm's than RHEL has?
>>>
>>> OT: the more I deal with ruby, the less I like it. Someone here was
>>> ready to move to a newer version, and from the ruby.org website, they're
>>> apparently actively hostile to all RH-related distros, even though
>>> we're the most common in North America. They've got a how to do it from
>>> debian and arch, how to use their own installer, and, oh, yes, they
say a lot
>>> of their community feels you should build from source.

>> IMNSHO, Ruby is only suitable for prototyping and low volume uses. it
>> doesn't scale well, and  ruby/rails websites perform abysmally under
>> heavy workloads.

> There are millions of popular, high trafficked web sites running RoR.

Yeah, and there are even more running Java, and tomcat Btw, the
wikipedia website only mentions a quarter of a million or so.

> To the OP - If we are talking about CentOS 5.x and you are determined to

No, 6.3.

> use RPM packages, Google 'enterprise ruby' and install it (it's Ruby
> 1.8.7) It's not likely to get any more updates though. If you get off the

Sorry, can't do that. As I believe I mentioned, they formerly required the
1.8.7 enterprise version, not the packaged version.

> need to have RPM packages, both rbenv & rvm install an alternate that
> downloads ruby source and compiles it for you and gives you sufficient
> shell modifications to make it appear somewhat seamless (I'm not promising
> the world here but it's not that difficult and my work has some CentOS 5.x
> still running enterprise-ruby-1.8.7 and everything newer has been Ubuntu
> 10.04 and either uses enterprise-ruby for 1.8.7 (becoming rare these days)
> and all new setups are rbenv and ruby 1.9.3-pXXX

Could you tell me what other, widely-used languages that don't have their
most recent stable versions in packages for the most-used distros? I'm not
aware of any.
Why is it that they don't package it?

I see, with a little googling, that it seems to be mostly ruby promoters
arguing it can scale, and a lot of everyone else being aware of issues.
And *I* have issues with it - it reminds me of python, 10-12 years ago,
when each subrelease would break code that was working fine. IIRC, when I
went to get a newer python required by one package I wanted to use, it
broke yum on RH 7.3 or 9, something like that, and ruby seems to be like
that.

AND I can't just rsync our internal repo with the latest volume, it looks
like I'll have to build it separately on each machine - I mean, if it
needs compiling

mark

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[m . roth]

Phil Dobbin wrote:
> On 02/04/2013 07:36 PM, John R Pierce wrote:
>> On 2/4/2013 7:21 AM, m.r...@5-cent.us wrote:
>>> Does anyone know of a repository that's*trustworthy*  (gotta worry
>>> 'bout malware) with newer ruby rpm's than RHEL has?
>>>
>>> OT: the more I deal with ruby, the less I like it. Someone here was
>>> ready to move to a newer version, and from the ruby.org website, they're
>>> apparently actively hostile to all RH-related distros, even though
>>> we're the most common in North America. They've got a how to do it from
>>> debian and arch, how to use their own installer, and, oh, yes, they
say a lot
>>> of their community feels you should build from source.
>>>
>>> Sorry, that's not my idea of a stable language that I'd ever recommend
>>> to someone
>>
>> IMNSHO, Ruby is only suitable for prototyping and low volume uses. it
>> doesn't scale well, and  ruby/rails websites perform abysmally under
>> heavy workloads.

> The doesn't scale well argument hasn't been the case for at least a few
> years now. Twitter is just one example. Some of the busiest sites on teh
> Interwebs are still using it.

Um, according to wikipedia, twitter went to scala, and uses ror for the
user interface.
>
> There are also projects, for example, like Puppet that are written in
> Ruby that are used by a lot of fairly large organisations.
>
> It may be worth your while reappraising Ruby.

I'm an admin these days, and don't get to argue this. However, when it's
packageable, and pushed out that way, so that someone can update a ton of
machines, and not hand-craft it, *AND* subreleases don't break working
code, I'll reconsider my attitude.

And as I think I said, I find the RoR website rather obnoxious in its
refusal to pay any attention to the biggest market in North America, RH
and RH-derived repos.

   mark

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Craig White]

On Feb 4, 2013, at 1:40 PM, m.r...@5-cent.us wrote:
> 
>> use RPM packages, Google 'enterprise ruby' and install it (it's Ruby
>> 1.8.7) It's not likely to get any more updates though. If you get off the
> 
> Sorry, can't do that. As I believe I mentioned, they formerly required the
> 1.8.7 enterprise version, not the packaged version.

enterprise ruby - clearly the best version of ruby 1.8.7 available anywhere and 
is available in RPM form.

http://www.rubyenterpriseedition.com/

>> need to have RPM packages, both rbenv & rvm install an alternate that
>> downloads ruby source and compiles it for you and gives you sufficient
>> shell modifications to make it appear somewhat seamless (I'm not promising
>> the world here but it's not that difficult and my work has some CentOS 5.x
>> still running enterprise-ruby-1.8.7 and everything newer has been Ubuntu
>> 10.04 and either uses enterprise-ruby for 1.8.7 (becoming rare these days)
>> and all new setups are rbenv and ruby 1.9.3-pXXX
> 
> Could you tell me what other, widely-used languages that don't have their
> most recent stable versions in packages for the most-used distros? I'm not
> aware of any.
> Why is it that they don't package it?

shouldn't you be asking this of upstream? They're the ones who choose which 
versions to include.

> 
> I see, with a little googling, that it seems to be mostly ruby promoters
> arguing it can scale, and a lot of everyone else being aware of issues.
> And *I* have issues with it - it reminds me of python, 10-12 years ago,
> when each subrelease would break code that was working fine. IIRC, when I
> went to get a newer python required by one package I wanted to use, it
> broke yum on RH 7.3 or 9, something like that, and ruby seems to be like
> that.
> 
> AND I can't just rsync our internal repo with the latest volume, it looks
> like I'll have to build it separately on each machine - I mean, if it
> needs compiling

rbenv and rvm have wonderful mechanisms for downloading & building ruby and 
even allow you multiple versions on the same computer running at the same time. 
The simplification of the process is quite complete.

Of course you wouldn't understand these things because you made up your mind a 
long time ago.

Craig
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Craig White]

On Feb 4, 2013, at 1:46 PM, m.r...@5-cent.us wrote:

> Phil Dobbin wrote:
>> 
>> The doesn't scale well argument hasn't been the case for at least a few
>> years now. Twitter is just one example. Some of the busiest sites on teh
>> Interwebs are still using it.
> 
> Um, according to wikipedia, twitter went to scala, and uses ror for the
> user interface.

What's wrong with that. They became the next biggest thing - so big that they 
had to make scaling adjustments. Successful sites do that.

>> 
>> There are also projects, for example, like Puppet that are written in
>> Ruby that are used by a lot of fairly large organisations.
>> 
>> It may be worth your while reappraising Ruby.
> 
> I'm an admin these days, and don't get to argue this. However, when it's
> packageable, and pushed out that way, so that someone can update a ton of
> machines, and not hand-craft it, *AND* subreleases don't break working
> code, I'll reconsider my attitude.
> 
> And as I think I said, I find the RoR website rather obnoxious in its
> refusal to pay any attention to the biggest market in North America, RH
> and RH-derived repos.

It's packaged and pushed out in a way that someone can update a ton of machines.

Trust me, I'm a DevOPS person… that's my job.

Even if Red Hat actually tried to keep up with ruby releases, I wouldn't use 
them and haven't used them for quite some time. The Enterprise Ruby versions 
were far superior to any version ever packaged by RH (garbage collection, 
performance, etc.). The reality is that if you are supporting Ruby/Ruby on 
Rails apps in any meaningful way, RH's ruby packaging is meaningless.

Craig

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[m . roth]

Craig White wrote:
> On Feb 4, 2013, at 1:40 PM, m.r...@5-cent.us wrote:
>>
>>> use RPM packages, Google 'enterprise ruby' and install it (it's Ruby
>>> 1.8.7) It's not likely to get any more updates though. If you get off
>>> the
>>
>> Sorry, can't do that. As I believe I mentioned, they formerly required
>> the 1.8.7 enterprise version, not the packaged version.
> 
> enterprise ruby - clearly the best version of ruby 1.8.7 available
> anywhere and is available in RPM form.
>
> http://www.rubyenterpriseedition.com/

Except we had it installed not from rpm.
> 
>>> need to have RPM packages, both rbenv & rvm install an alternate that
>>> downloads ruby source and compiles it for you and gives you sufficient
>>> shell modifications to make it appear somewhat seamless (I'm not
>>> promising the world here but it's not that difficult and my work has
>>> some CentOS 5.x still running enterprise-ruby-1.8.7 and everything
>>> newer has been Ubuntu 10.04 and either uses enterprise-ruby for 1.8.7
>>> (becoming rare these days) and all new setups are rbenv and ruby
1.9.3-pXXX
>>
>> Could you tell me what other, widely-used languages that don't have
>> their most recent stable versions in packages for the most-used
distros? I'm
>> not aware of any.

>> Why is it that they don't package it?
> 
> shouldn't you be asking this of upstream? They're the ones who choose
> which versions to include.

No. If I cared enough, I'd ask on the RUBY list. It's ruby.org that
appears to ignore CentOS and all other RH-derived distros.

Btw, you might notice we're on the CentOS, not ubuntu, or some other
distro list.
>>
>> I see, with a little googling, that it seems to be mostly ruby promoters
>> arguing it can scale, and a lot of everyone else being aware of issues.
>> And *I* have issues with it - it reminds me of python, 10-12 years ago,
>> when each subrelease would break code that was working fine. IIRC, when
>> I went to get a newer python required by one package I wanted to use, it
>> broke yum on RH 7.3 or 9, something like that, and ruby seems to be like
>> that.
>>
>> AND I can't just rsync our internal repo with the latest volume, it
>> looks like I'll have to build it separately on each machine - I mean,
if it
>> needs compiling
> 
> rbenv and rvm have wonderful mechanisms for downloading & building ruby
> and even allow you multiple versions on the same computer running at the
> same time. The simplification of the process is quite complete.

I notice that you are ignoring my issues, and go on about how wonderful
the unique ruby package manager is, and say nothing of installing on a
number of machines at once.
>
> Of course you wouldn't understand these things because you made up your
> mind a long time ago.

I've only slowly made up my mind, but the more I have to deal with ruby,
as I said, the less I like it.

You, on the other hand, have already come here with the attitude of "my
way or the highway"; this is *NOT* the way to encourage folks to change
their minds.* Nor is it helpful to me.

mark

* "You can catch more flies with honey than vinegar"

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Craig White]

On Feb 4, 2013, at 3:12 PM, m.r...@5-cent.us wrote:

> Craig White wrote:
>>> Why is it that they don't package it?
>> 
>> shouldn't you be asking this of upstream? They're the ones who choose
>> which versions to include.
> 
> No. If I cared enough, I'd ask on the RUBY list. It's ruby.org that
> appears to ignore CentOS and all other RH-derived distros.
> 
> Btw, you might notice we're on the CentOS, not ubuntu, or some other
> distro list.

and you'd get no answer from the ruby list. The decisions about which versions 
are packaged are made by Red Hat and obviously CentOS gets from upstream. Just 
the same as perl, python, etc.

>> rbenv and rvm have wonderful mechanisms for downloading & building ruby
>> and even allow you multiple versions on the same computer running at the
>> same time. The simplification of the process is quite complete.
> 
> I notice that you are ignoring my issues, and go on about how wonderful
> the unique ruby package manager is, and say nothing of installing on a
> number of machines at once.

ignoring your issue because it doesn't apply to my workflow. All new deploys 
are done via ruby so it's the first and almost only thing installed on a fresh 
'base' installation. At the point that ruby is installed, I install the puppet 
gem and then invoke the first puppet run. At that point, all software 
installations are done via puppet.

As for which version of ruby, we decide that prior to installation of a server 
since we have virtual servers and some hardware servers that we deploy 
applications for specific versions of ruby.

On say my development machine, I can merely type 'sudo rbenv install 
1.9.3-p194' and it will be downloaded and installed automatically.

>> 
>> Of course you wouldn't understand these things because you made up your
>> mind a long time ago.
> 
> I've only slowly made up my mind, but the more I have to deal with ruby,
> as I said, the less I like it.

sorry but whether Mark likes it or not is not meaningful to me. I've been 
developing on RoR since like 0.0.7 version. It's become much of my livelihood.

> You, on the other hand, have already come here with the attitude of "my
> way or the highway"; this is *NOT* the way to encourage folks to change
> their minds.* Nor is it helpful to me.

not really - it's that you don't have much first hand experience so since you 
can't merely type 'yum install rails' and be done with it, it confuses you.

Craig
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Les Mikesell]

On Mon, Feb 4, 2013 at 4:41 PM, Craig White  wrote:
>
> 
>> You, on the other hand, have already come here with the attitude of "my
>> way or the highway"; this is *NOT* the way to encourage folks to change
>> their minds.* Nor is it helpful to me.
> 
> not really - it's that you don't have much first hand experience so since you 
> can't merely type 'yum install rails' and be done with it, it confuses you.

The confusing thing is that for almost everything else that is useful,
stable, and publicly available, someone will maintain packages.   If
they aren't accepted in the distribution or in EPEL, the projects
generally have their own repo for them.   So why did you have to roll
your own installer, and why do you thing that is a good thing?

-- 
  Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Craig White]

On Feb 4, 2013, at 3:54 PM, Les Mikesell wrote:

> On Mon, Feb 4, 2013 at 4:41 PM, Craig White  wrote:
>> 
>> 
>>> You, on the other hand, have already come here with the attitude of "my
>>> way or the highway"; this is *NOT* the way to encourage folks to change
>>> their minds.* Nor is it helpful to me.
>> 
>> not really - it's that you don't have much first hand experience so since 
>> you can't merely type 'yum install rails' and be done with it, it confuses 
>> you.
> 
> The confusing thing is that for almost everything else that is useful,
> stable, and publicly available, someone will maintain packages.   If
> they aren't accepted in the distribution or in EPEL, the projects
> generally have their own repo for them.   So why did you have to roll
> your own installer, and why do you thing that is a good thing?

there may very well be ruby versions in EPEL, I don't know and never looked.

I also never 'rolled my own installer' - the 2 'ruby managers' (rbenv and rvm) 
have that functionality.

When you are developing, it became necessary to maintain applications 
undoubtedly on ruby 1.8.7 and work on new applications (1.9.3x) and so have a 
version manager for ruby was essential - that's why they were created.

FTR… rbenv & rvm are cross platform and recommended not only for Linux but also 
for Macintosh (as opposed to using Apple's supplied ruby). Windows is less than 
optimal for ruby/ruby on rails.

What we are discussing here is more likely deployment servers and CentOS-5.x is 
stuck on ruby-1.8.5 which is pretty much useless and CentOS-6.x is as I 
understand it, stuck on ruby-1.8.7. Even still, the enterprise ruby package 
(1.8.7) is vastly superior to RH's build and if you are running an application 
that you care about, you would want to use that or better yet, ruby 1.9.3x has 
vastly improved performance of all 1.8.7 builds.

By the way, I am pretty certain that PuppetLabs (puppet) maintains ruby 
packages for CentOS, Ubuntu, Debian, Windows too.

And lastly, Ruby is an ecosystem far beyond the base language. It has a 'gem' 
package management system which again is cross platform and even when you try 
to package ruby in rpm's, there's no way an RH or EPEL will keep up with 
updates.

Craig

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

[CentOS] Questions about software RAID, LVM.

[Robert Heller]

I am planning to increase the disk space on my desktop system.  It is
running CentOS 5.9 w/XEN.  I have two 160Gig 2.5" laptop (2.5") SATA drives
in two slots of a 4-slot hot swap bay configured like this:

Disk /dev/sda: 160.0 GB, 160041885696 bytes
255 heads, 63 sectors/track, 19457 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot  Start End  Blocks   Id  System
/dev/sda1   *   1 125 1004031   fd  Linux raid autodetect
/dev/sda2 126   19457   155284290   fd  Linux raid autodetect

Disk /dev/sdb: 160.0 GB, 160041885696 bytes
255 heads, 63 sectors/track, 19457 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot  Start End  Blocks   Id  System
/dev/sdb1   *   1 125 1004031   fd  Linux raid autodetect
/dev/sdb2 126   19457   155284290   fd  Linux raid autodetect

sauron.deepsoft.com% cat /proc/mdstat 
Personalities : [raid1] 
md0 : active raid1 sdb1[1] sda1[0]
  1003904 blocks [2/2] [UU]
  
md1 : active raid1 sdb2[1] sda2[0]
  155284224 blocks [2/2] [UU]
  
unused devices: 

That is I have two RAID1 arrays: a small (1Gig) one mounted as /boot
and a larger 148Gig one that is a LVM Volume Group (which contains a
pile of file systems, some for DOM0 and some that are for other VMs). 
What I plan on doing is getting a pair of 320Gig 2.5" (laptop) SATA
disks and fail over the existing disks to this new pair.  I believe I
can then 'grow' the second RAID array to be like ~300Gig.  My question
is: what happens to the LVM Volume Group?  Will it grow when the RAID
array grows?  Or should I leave /dev/md1 its current size and create a
new RAID array and add this as a second PV and grow the Volume Group
that way?  The documentation is not clear as to what happens -- the VG
is marked 'resisable'.

sauron.deepsoft.com% sudo pvdisplay
  --- Physical volume ---
  PV Name   /dev/md1
  VG Name   sauron
  PV Size   148.09 GB / not usable 768.00 KB
  Allocatable   yes 
  PE Size (KByte)   4096
  Total PE  37911
  Free PE   204
  Allocated PE  37707
  PV UUID   ttB15B-3eWx-4ioj-TUvm-lAPM-z9rD-Prumee
   
sauron.deepsoft.com% sudo vgdisplay
  --- Volume group ---
  VG Name   sauron
  System ID 
  Formatlvm2
  Metadata Areas1
  Metadata Sequence No  65
  VG Access read/write
  VG Status resizable
  MAX LV0
  Cur LV17
  Open LV   12
  Max PV0
  Cur PV1
  Act PV1
  VG Size   148.09 GB
  PE Size   4.00 MB
  Total PE  37911
  Alloc PE / Size   37707 / 147.29 GB
  Free  PE / Size   204 / 816.00 MB
  VG UUID   qG8gCf-3vou-7dp2-Ar0B-p8jz-eXZF-3vOONr
   


-- 
Robert Heller -- 978-544-6933 / hel...@deepsoft.com
Deepwoods Software-- http://www.deepsoft.com/
()  ascii ribbon campaign -- against html e-mail
/\  www.asciiribbon.org   -- against proprietary attachments



___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Les Mikesell]

On Mon, Feb 4, 2013 at 5:36 PM, Craig White  wrote:
>
>
> And lastly, Ruby is an ecosystem far beyond the base language. It has a 'gem' 
> package management system which again is cross platform and even when you try 
> to package ruby in rpm's, there's no way an RH or EPEL will keep up with 
> updates.
>

I guess I still don't understand why you think that is a good thing.
If the developers didn't get it right the first dozen times, why do
you think the next update will be better?  That is, if EPEL can't keep
up, why would anyone want to?  If you don't have the QA that a
packager does it means you have to do it yourself.

-- 
   Les Mikesell
 lesmikes...@gmail.com
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] Questions about software RAID, LVM.

[SilverTip257]

On Mon, Feb 4, 2013 at 6:40 PM, Robert Heller  wrote:

> I am planning to increase the disk space on my desktop system.  It is
> running CentOS 5.9 w/XEN.  I have two 160Gig 2.5" laptop (2.5") SATA drives
> in two slots of a 4-slot hot swap bay configured like this:
>
>
I would certainly suggest testing these steps out in some form or another
as a dry run.  And verify your backups and/or create new ones ahead of
time. :)

Create your new softraid mirror with the larger disks, add that device
(we'll say md3 - since the new /boot will be md2 temporarily) to your
existing LVM volume group, then pvmove the physical extents from the
existing disk /dev/md1 to /dev/md3.  Then verify that the extents have been
moved off md1 and are on md3.  Finally remove md1 from the VG with pvremove.

I don't recall if I specifically used these notes, but the process matches
the one I used. [0]
A few more bits of info I recognize from my readings a while back. [1] [2]
[3]

I expect you could easily tweak your mdadm config and rebuild your initial
ramdisk so that the next time you reboot there isn't an issue.  This all
depends on how much of this process you plan on doing online.

[0]
http://www.rhcedan.com/2010/10/20/migrating-physical-volumes-in-a-lvm2-volume-group/
[1]
http://www.centos.org/docs/5/html/Cluster_Logical_Volume_Manager/move_new_ex4.html
[2] http://tldp.org/HOWTO/LVM-HOWTO/removeadisk.html
[3]
http://www.whmcr.com/2011/06/21/moving-part-of-an-lvm-vg-from-one-pv-to-another/

sauron.deepsoft.com% cat /proc/mdstat
> Personalities : [raid1]
> md0 : active raid1 sdb1[1] sda1[0]
>   1003904 blocks [2/2] [UU]
>
> md1 : active raid1 sdb2[1] sda2[0]
>   155284224 blocks [2/2] [UU]
>
> unused devices: 
>
> That is I have two RAID1 arrays: a small (1Gig) one mounted as /boot
> and a larger 148Gig one that is a LVM Volume Group (which contains a
> pile of file systems, some for DOM0 and some that are for other VMs).
> What I plan on doing is getting a pair of 320Gig 2.5" (laptop) SATA
> disks and fail over the existing disks to this new pair.  I believe I
> can then 'grow' the second RAID array to be like ~300Gig.  My question
> is: what happens to the LVM Volume Group?  Will it grow when the RAID
>

In my experience the VG will grow by whatever extents the new PV has.
I found it quite helpful to use loopback devices to test possible softraid
and/or LVM scenarios.


> array grows?  Or should I leave /dev/md1 its current size and create a
> new RAID array and add this as a second PV and grow the Volume Group
>

Just add md3 to the VG and move the extents as noted above.


> that way?  The documentation is not clear as to what happens -- the VG
> is marked 'resisable'.
>

The biggest gotcha is with PVs.
Example even though this doesn't apply:
In that if you plan on enlarging a disk (ex: LVM backing for a VM that uses
LVM), the PV in the VM has to be offline to resize (the OS doesn't
recognize the larger disk until after a reboot, but cannot be resized since
the PV is online again!).  You avoid that pitfall since you will have a
separate disk (md3) that can be hot added.


>
> sauron.deepsoft.com% sudo pvdisplay
>   --- Physical volume ---
>   PV Name   /dev/md1
>   VG Name   sauron
>   PV Size   148.09 GB / not usable 768.00 KB
>   Allocatable   yes
>   PE Size (KByte)   4096
>   Total PE  37911
>   Free PE   204
>   Allocated PE  37707
>   PV UUID   ttB15B-3eWx-4ioj-TUvm-lAPM-z9rD-Prumee
>
> sauron.deepsoft.com% sudo vgdisplay
>   --- Volume group ---
>   VG Name   sauron
>   System ID
>   Formatlvm2
>   Metadata Areas1
>   Metadata Sequence No  65
>   VG Access read/write
>   VG Status resizable
>   MAX LV0
>   Cur LV17
>   Open LV   12
>   Max PV0
>   Cur PV1
>   Act PV1
>   VG Size   148.09 GB
>   PE Size   4.00 MB
>   Total PE  37911
>   Alloc PE / Size   37707 / 147.29 GB
>   Free  PE / Size   204 / 816.00 MB
>   VG UUID   qG8gCf-3vou-7dp2-Ar0B-p8jz-eXZF-3vOONr
>
>
>
> --
> Robert Heller -- 978-544-6933 / hel...@deepsoft.com
> Deepwoods Software-- http://www.deepsoft.com/
> ()  ascii ribbon campaign -- against html e-mail
> /\  www.asciiribbon.org   -- against proprietary attachments
>
>
>
> ___
> CentOS mailing list
> CentOS@centos.org
> http://lists.centos.org/mailman/listinfo/centos
>


Hope this helps.

Best Regards,
-- 
---~~.~~---
Mike
//  SilverTip257  //
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Craig White]

On Mon, 2013-02-04 at 18:01 -0600, Les Mikesell wrote:
> On Mon, Feb 4, 2013 at 5:36 PM, Craig White  wrote:
> >
> >
> > And lastly, Ruby is an ecosystem far beyond the base language. It has a 
> > 'gem' package management system which again is cross platform and even when 
> > you try to package ruby in rpm's, there's no way an RH or EPEL will keep up 
> > with updates.
> >
> 
> I guess I still don't understand why you think that is a good thing.
> If the developers didn't get it right the first dozen times, why do
> you think the next update will be better?  That is, if EPEL can't keep
> up, why would anyone want to?  If you don't have the QA that a
> packager does it means you have to do it yourself.

Once installed, I rarely have to update ruby on a server.

Gems however are another matter altogether and a typical ruby
application (Rails or otherwise) is likely to require many ruby gems
(think Perl CPAN).

When I first started deploying RoR applications on CentOS or RHEL, all
of the gem RPM's lagged way behind and it was a pain. So you discover
that even if the base ruby install was RPM, you pretty much abandoned
RPM's for gems. The gem package management system is self contained and
excellent, even compiling gems that require 'native extensions' on the
fly. There are thousands of gems (again, think CPAN). No packager is
going to take on the commitment of building rpm's for all of them so
they might build RPM's for the most popular gems and they will fall
behind quickly.

So the history is - ruby RPM's from RH and Debian tended to be generic,
featureless and updated only when security issues arose (hardly ever).
Enterprise Ruby developers (the same that write/maintain passenger) came
up with a far superior ruby build, required far less memory to run,
didn't leak and was substantially faster. Why look back? Even so, the
ruby packages on say CentOS are minimal (ruby, ruby-doc, ruby-ri,
ruby-irb, ruby-dev). The rest is all gems and RPM's are not useful
there.

Craig


-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Bry8 Star]

Thanks for this discussion. I also had (and was about to
ask) similar question(s).
I've already tried to get/use Latest+Stable ruby compiled
& used on other RHEL based repo, but something conflicted,
as i'm new to these, so beyond my understanding what was
it at this point.
But need to look-into/try-out what you've
discussed/suggested here, may be useful for my case,
(solutions which use very low memory footprint, and large
vswap or swap or database use is/are ok).
-- Bright Star.



Received from Craig White, on 2013-02-04 9:52 PM:
> 
> On Feb 4, 2013, at 1:46 PM, m.r...@5-cent.us wrote:
> 
>> Phil Dobbin wrote:
>>> 
>>> The doesn't scale well argument hasn't been the
>>> case for at least a few years now. Twitter is just
>>> one example. Some of the busiest sites on teh 
>>> Interwebs are still using it.
>> 
>> Um, according to wikipedia, twitter went to scala,
>> and uses ror for the user interface.
>  What's wrong with that. They became the next
> biggest thing - so big that they had to make scaling
> adjustments. Successful sites do that. 
>>> 
>>> There are also projects, for example, like Puppet
>>> that are written in Ruby that are used by a lot of
>>> fairly large organisations.
>>> 
>>> It may be worth your while reappraising Ruby.
>> 
>> I'm an admin these days, and don't get to argue this.
>> However, when it's packageable, and pushed out that
>> way, so that someone can update a ton of machines,
>> and not hand-craft it, *AND* subreleases don't break
>> working code, I'll reconsider my attitude.
>> 
>> And as I think I said, I find the RoR website rather
>> obnoxious in its refusal to pay any attention to the
>> biggest market in North America, RH and RH-derived
>> repos.
>  It's packaged and pushed out in a way that someone
> can update a ton of machines.
> 
> Trust me, I'm a DevOPS person… that's my job.
> 
> Even if Red Hat actually tried to keep up with ruby
> releases, I wouldn't use them and haven't used them for
> quite some time. The Enterprise Ruby versions were far
> superior to any version ever packaged by RH (garbage
> collection, performance, etc.). The reality is that if
> you are supporting Ruby/Ruby on Rails apps in any
> meaningful way, RH's ruby packaging is meaningless.
> 
> Craig
> 
> ___
> CentOS
> mailing list CentOS@centos.org 
> http://lists.centos.org/mailman/listinfo/centos
> 



signature.asc
Description: OpenPGP digital signature
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[Phil Dobbin]

On 02/04/2013 11:36 PM, Craig White wrote:
> 
> On Feb 4, 2013, at 3:54 PM, Les Mikesell wrote:
> 
>> On Mon, Feb 4, 2013 at 4:41 PM, Craig White  wrote:
>>>
>>> 
 You, on the other hand, have already come here with the attitude of "my
 way or the highway"; this is *NOT* the way to encourage folks to change
 their minds.* Nor is it helpful to me.
>>> 
>>> not really - it's that you don't have much first hand experience so since 
>>> you can't merely type 'yum install rails' and be done with it, it confuses 
>>> you.
>>
>> The confusing thing is that for almost everything else that is useful,
>> stable, and publicly available, someone will maintain packages.   If
>> they aren't accepted in the distribution or in EPEL, the projects
>> generally have their own repo for them.   So why did you have to roll
>> your own installer, and why do you thing that is a good thing?
> 
> there may very well be ruby versions in EPEL, I don't know and never looked.
> 
> I also never 'rolled my own installer' - the 2 'ruby managers' (rbenv and 
> rvm) have that functionality.
> 
> When you are developing, it became necessary to maintain applications 
> undoubtedly on ruby 1.8.7 and work on new applications (1.9.3x) and so have a 
> version manager for ruby was essential - that's why they were created.
> 
> FTR… rbenv & rvm are cross platform and recommended not only for Linux but 
> also for Macintosh (as opposed to using Apple's supplied ruby). Windows is 
> less than optimal for ruby/ruby on rails.
> 
> What we are discussing here is more likely deployment servers and CentOS-5.x 
> is stuck on ruby-1.8.5 which is pretty much useless and CentOS-6.x is as I 
> understand it, stuck on ruby-1.8.7. Even still, the enterprise ruby package 
> (1.8.7) is vastly superior to RH's build and if you are running an 
> application that you care about, you would want to use that or better yet, 
> ruby 1.9.3x has vastly improved performance of all 1.8.7 builds.
> 
> By the way, I am pretty certain that PuppetLabs (puppet) maintains ruby 
> packages for CentOS, Ubuntu, Debian, Windows too.
> 
> And lastly, Ruby is an ecosystem far beyond the base language. It has a 'gem' 
> package management system which again is cross platform and even when you try 
> to package ruby in rpm's, there's no way an RH or EPEL will keep up with 
> updates.

Puppet is available via EPEL & as a separate rpm for Fedora. It "works"
on basically any distro.

Fedora's Ruby is 'ruby 1.9.3p362 (2012-12-25 revision 38607)
[x86_64-linux]' which is the latest stable version & can be installed
via yum.

As I mentioned before, rvm is invaluable (to me at any rate) for
managing Rubies & their associated gemsets & was written with large
scale deployments in mind.

I find it unfortunate that the myth is still being perpetuated about its
so called shortcomings albeit generally by people who have never used it
to any great extent.

They also tend to imply that Ruby equals Rails & vice versa (which is
kind of like saying Python equals Django).

Cheers,

  Phil...

-- 
currently (ab)using
CentOS 5.8 & 6.3, Debian Squeeze & Wheezy, Fedora Beefy & Spherical,
Lubuntu 12.10, OS X Snow Leopard & Ubuntu Precise & Quantal
GnuPG Key : http://www.horse-latitudes.co.uk/publickey.asc


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos

Re: [CentOS] recent ruby packages?

[pnorton3 . 14]

Maybe we should think about writing the kernel in java/python/ruby/php etc?  
Wonder why this hasn't happened before? 
--

-Original Message-
From: Phil Dobbin 
Sender: centos-boun...@centos.org
Date: Tue, 05 Feb 2013 07:30:59 
To: CentOS mailing list
Reply-To: CentOS mailing list 
Subject: Re: [CentOS] recent ruby packages?

On 02/04/2013 11:36 PM, Craig White wrote:
> 
> On Feb 4, 2013, at 3:54 PM, Les Mikesell wrote:
> 
>> On Mon, Feb 4, 2013 at 4:41 PM, Craig White  wrote:
>>>
>>> 
 You, on the other hand, have already come here with the attitude of "my
 way or the highway"; this is *NOT* the way to encourage folks to change
 their minds.* Nor is it helpful to me.
>>> 
>>> not really - it's that you don't have much first hand experience so since 
>>> you can't merely type 'yum install rails' and be done with it, it confuses 
>>> you.
>>
>> The confusing thing is that for almost everything else that is useful,
>> stable, and publicly available, someone will maintain packages.   If
>> they aren't accepted in the distribution or in EPEL, the projects
>> generally have their own repo for them.   So why did you have to roll
>> your own installer, and why do you thing that is a good thing?
> 
> there may very well be ruby versions in EPEL, I don't know and never looked.
> 
> I also never 'rolled my own installer' - the 2 'ruby managers' (rbenv and 
> rvm) have that functionality.
> 
> When you are developing, it became necessary to maintain applications 
> undoubtedly on ruby 1.8.7 and work on new applications (1.9.3x) and so have a 
> version manager for ruby was essential - that's why they were created.
> 
> FTR… rbenv & rvm are cross platform and recommended not only for Linux but 
> also for Macintosh (as opposed to using Apple's supplied ruby). Windows is 
> less than optimal for ruby/ruby on rails.
> 
> What we are discussing here is more likely deployment servers and CentOS-5.x 
> is stuck on ruby-1.8.5 which is pretty much useless and CentOS-6.x is as I 
> understand it, stuck on ruby-1.8.7. Even still, the enterprise ruby package 
> (1.8.7) is vastly superior to RH's build and if you are running an 
> application that you care about, you would want to use that or better yet, 
> ruby 1.9.3x has vastly improved performance of all 1.8.7 builds.
> 
> By the way, I am pretty certain that PuppetLabs (puppet) maintains ruby 
> packages for CentOS, Ubuntu, Debian, Windows too.
> 
> And lastly, Ruby is an ecosystem far beyond the base language. It has a 'gem' 
> package management system which again is cross platform and even when you try 
> to package ruby in rpm's, there's no way an RH or EPEL will keep up with 
> updates.

Puppet is available via EPEL & as a separate rpm for Fedora. It "works"
on basically any distro.

Fedora's Ruby is 'ruby 1.9.3p362 (2012-12-25 revision 38607)
[x86_64-linux]' which is the latest stable version & can be installed
via yum.

As I mentioned before, rvm is invaluable (to me at any rate) for
managing Rubies & their associated gemsets & was written with large
scale deployments in mind.

I find it unfortunate that the myth is still being perpetuated about its
so called shortcomings albeit generally by people who have never used it
to any great extent.

They also tend to imply that Ruby equals Rails & vice versa (which is
kind of like saying Python equals Django).

Cheers,

  Phil...

-- 
currently (ab)using
CentOS 5.8 & 6.3, Debian Squeeze & Wheezy, Fedora Beefy & Spherical,
Lubuntu 12.10, OS X Snow Leopard & Ubuntu Precise & Quantal
GnuPG Key : http://www.horse-latitudes.co.uk/publickey.asc


___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos
___
CentOS mailing list
CentOS@centos.org
http://lists.centos.org/mailman/listinfo/centos