Re: [Pacemaker] A function demand of the new environment.

2009-10-29 Thread renayama19661014
Hi Andrew, > Or are you looking for the ability to do more than simply prevent > resources from running? Yes. For the realization of the function to assure the practice of the resources of the cluster, we use respawn. We found the function that it might be replaced of respawn in a road map of

Re: [Pacemaker] trying to import a constraint

2009-10-29 Thread Shravan Mishra
I just found my problem, so here it is in case someone spends a couple of hours like me to figure out why it's not working The culprit is -X, don't use that. The correct command is : cibadmin -o constraints -C -p << END END Shravan On Thu, Oct 29, 2009 at 1:57 PM, Shravan Mishra

[Pacemaker] trying to import a constraint

2009-10-29 Thread Shravan Mishra
Hi guys, Below is a partial snippet of my config: = === When I try doing : cibadmin -o constraints -C -X -p << END

Re: [Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
Yes, files are there. See my first post. Ran ocf-tester on mysql RA and get this: /usr/lib/ocf/resource.d/heartbeat/mysql passed all tests But, after modifying the crm Python code to uncomment one of the error messages, I see this: crm(live)configure# primitive mysql ocf:heartbeat:mysql ERROR

Re: [Pacemaker] CRM help

2009-10-29 Thread Johan Verrept
have you checked the /usr/lib/ocf directory? J. On Thu, 2009-10-29 at 12:06 -0400, Eliot Gable wrote: > Actually, that was a bad example because it is called 'pgsql', but I get it > for all of them: > > ERROR: ocf:heartbeat:apache: no such resource agent > ERROR: ocf:heartbeat:mysql: no

Re: [Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
Actually, that was a bad example because it is called 'pgsql', but I get it for all of them: ERROR: ocf:heartbeat:apache: no such resource agent ERROR: ocf:heartbeat:mysql: no such resource agent ERROR: ocf:heartbeat:pgsql: no such resource agent ERROR: ocf:heartbeat:pingd: no such resource agent

Re: [Pacemaker] CRM help

2009-10-29 Thread Johan Verrept
On Thu, 2009-10-29 at 12:02 -0400, Eliot Gable wrote: > Ok, I figured out I need to change the clone to primitive to define the thing > first and then reference the primitive using the clone command. I now no > longer get the object does not exist message, but it now says it cannot find > the re

Re: [Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
Ok, I figured out I need to change the clone to primitive to define the thing first and then reference the primitive using the clone command. I now no longer get the object does not exist message, but it now says it cannot find the resource agent: ERROR: ocf:heartbeat:postgresql: no such resour

Re: [Pacemaker] CRM help

2009-10-29 Thread Johan Verrept
Ah. you could also try this: primitive apacheserver ocf:heartbeat:apache params configfile=/etc/httpd/httpd.conf clone apache apacheserver meta clone-max="2" No guarantees this is 100% correct (i didn't try) but you should get the idea of what is wrong. Anyway, if you are creating a script for a

Re: [Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
Ok, that still does not work. When I run the script, it just sits there and does nothing (does not even return me to the prompt). When I hit CTRL-C to try to terminate it, it just gives me the same errors I was getting before about ocf:heartbeat:apache and etc not existing. But I have to keep hi

Re: [Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
Nevermind. I had a leading 'configure' statement by itself. Eliot Gable Senior Engineer 1228 Euclid Ave, Suite 390 Cleveland, OH 44115 Direct: 216-373-4808 Fax: 216-373-4657 ega...@broadvox.net CONFIDENTIAL COMMUNICATION. This e-mail and any files transmitted with it are confidential and are

Re: [Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
That just results in syntax errors on every line. Eliot Gable Senior Engineer 1228 Euclid Ave, Suite 390 Cleveland, OH 44115 Direct: 216-373-4808 Fax: 216-373-4657 ega...@broadvox.net CONFIDENTIAL COMMUNICATION. This e-mail and any files transmitted with it are confidential and are intended s

Re: [Pacemaker] Cluster Test Suite: Bad News

2009-10-29 Thread Dejan Muhamedagic
Hi, On Wed, 28 Oct 2009 20:33 +0100, "Lars Marowsky-Bree" wrote: > On 2009-10-28T21:25:25, Andrew Beekhof wrote: > > > > ./CTSlab.py --at-boot 1 --nodes 'mysqlha1 mysqlha2' --stack ais > > > --syslog-facility local6 --schema pacemaker-1.0 --logfile > > > /var/log/messages > > > --stonith 1 --s

Re: [Pacemaker] CRM help

2009-10-29 Thread Johan Verrept
If you use scripted input to crm, use full command lines: CRM < In the past, I have always done things by manually creating a CIB XML > file and then importing it. But, to save time, I thought I would try > CRM. So, I made this script: > > > > #!/bin/bash > > > > crm< > configure > > e

[Pacemaker] CRM help

2009-10-29 Thread Eliot Gable
In the past, I have always done things by manually creating a CIB XML file and then importing it. But, to save time, I thought I would try CRM. So, I made this script: #!/bin/bash crm

Re: [Pacemaker] RFC: Compacting constraints

2009-10-29 Thread Dejan Muhamedagic
Hi, On Thu, 29 Oct 2009 14:39 +0100, "Lars Marowsky-Bree" wrote: > Hi all, > > I have a pretty common use case - 4-16 nodes with OCFS2 etc, hosting a > ton of Xen/KVM guests. > > Compacting the OCFS2 setup was pretty easy - > http://www.advogato.org/person/lmb/diary.html?start=104 - and that pa

Re: [Pacemaker] resource's start/stop not getting called

2009-10-29 Thread Lars Marowsky-Bree
On 2009-10-28T21:16:17, Shravan Mishra wrote: > crm_attribute -t crm_config -n stonith-enabled -v false Remember to not use any shared storage in this environment. BTW, if you had configured the primitive resource via the CRM shell, you'd have gotten this info message on the first attempt ;-)

Re: [Pacemaker] Remote managmenet of a cluster

2009-10-29 Thread Lars Marowsky-Bree
On 2009-10-29T15:21:18, Michael Schwartzkopff wrote: > > > I haven't been able to find too much documentation about the managenemt > > > of a cluster from a remote machine by CLI (crm subshell or cibadmin). Any > > > hint for me where I can find more information about the stuff? > > You mean remo

Re: [Pacemaker] Remote managmenet of a cluster

2009-10-29 Thread Michael Schwartzkopff
Am Donnerstag 29 Oktober 2009 15:29 schrieb Andrew Beekhof: > There is > > http://www.clusterlabs.org/doc/pacemaker-explained/ch-advanced-options.html >#s-remote-connection > > But you need to be running code from no older than last week. Thanks. That is was I was looking for. I will try it wit

Re: [Pacemaker] Remote managmenet of a cluster

2009-10-29 Thread Andrew Beekhof
There is http://www.clusterlabs.org/doc/pacemaker-explained/ch-advanced-options.html#s-remote-connection But you need to be running code from no older than last week. On Thu, Oct 29, 2009 at 4:21 PM, Michael Schwartzkopff wrote: > Am Donnerstag 29 Oktober 2009 15:01 schrieb Lars Marowsky-Bre

Re: [Pacemaker] Remote managmenet of a cluster

2009-10-29 Thread Michael Schwartzkopff
Am Donnerstag 29 Oktober 2009 15:01 schrieb Lars Marowsky-Bree: > On 2009-10-29T14:41:10, Michael Schwartzkopff wrote: > > Hi, > > > > I haven't been able to find too much documentation about the managenemt > > of a cluster from a remote machine by CLI (crm subshell or cibadmin). Any > > hint for

Re: [Pacemaker] Pacemaker Build Issues on CentOS 5.3 64-bit

2009-10-29 Thread Johan Verrept
On Thu, 2009-10-29 at 10:00 -0400, Eliot Gable wrote: > Once I modify configure.ac to remove this flag and regenerate > everything, you could have used: ./configure --disable-fatal-warnings > I run into linking errors because the Makefile LIBS= line tells > Pacemaker to use /usr/lib and /lib64

[Pacemaker] Pacemaker Build Issues on CentOS 5.3 64-bit

2009-10-29 Thread Eliot Gable
When building Pacemaker (1-0-78e8ab00359a) on CentOS 5.3 64-bit, I run into all sorts of errors. First, the compiler flag '-Werror' is enabled. On CentOS 5.3, gcc issues a warning when you drop a const qualifier on a pointer, which Pacemaker code seems to do all over the place (why does it do th

Re: [Pacemaker] Remote managmenet of a cluster

2009-10-29 Thread Lars Marowsky-Bree
On 2009-10-29T14:41:10, Michael Schwartzkopff wrote: > Hi, > > I haven't been able to find too much documentation about the managenemt of a > cluster from a remote machine by CLI (crm subshell or cibadmin). Any hint for > me where I can find more information about the stuff? You mean remote c

[Pacemaker] Remote managmenet of a cluster

2009-10-29 Thread Michael Schwartzkopff
Hi, I haven't been able to find too much documentation about the managenemt of a cluster from a remote machine by CLI (crm subshell or cibadmin). Any hint for me where I can find more information about the stuff? I only found a thread on the mailing list indicating that the remote management i

[Pacemaker] RFC: Compacting constraints

2009-10-29 Thread Lars Marowsky-Bree
Hi all, I have a pretty common use case - 4-16 nodes with OCFS2 etc, hosting a ton of Xen/KVM guests. Compacting the OCFS2 setup was pretty easy - http://www.advogato.org/person/lmb/diary.html?start=104 - and that part seems short enough. For each guest, I need an order and collocation constrai

Re: [Pacemaker] A function demand of the new environment.

2009-10-29 Thread Andrew Beekhof
Apologies for the delay On Tue, Sep 22, 2009 at 3:13 AM, wrote: > Hi Andrew, > >> Ok, but I'm trying to understand why. > > Thank you. > >> What function are these daemons performing?  Why do they need to keep >> running even when the node is in standby? > > For example, respawn can detect the a

Re: [Pacemaker] how to run a remote script [SOLVED]

2009-10-29 Thread gilberto migliavacca
Hi I solved the problem using the APACHE for the moment I created the the apache resource and the ip resource qith the crm command line tool than with the crm resource start command I can start (and after that stop) the apache server on the node there was a problem with the APACHE installatio

Re: [Pacemaker] server lockup failures

2009-10-29 Thread Andrew Beekhof
On Thu, Oct 29, 2009 at 12:51 AM, Bernd Schubert wrote: > On Wednesday 28 October 2009, Andrew Beekhof wrote: >> On Wed, Oct 28, 2009 at 2:44 PM, Bernd Schubert >> >> wrote: >> > On Wednesday 28 October 2009, Andrew Beekhof wrote: >> >> On Wed, Oct 28, 2009 at 1:05 PM, Bernd Schubert >> >> >> >>