[Pacemaker] Problems with configuring colocation and start order

Thu Nov 18 14:15:59 UTC 2010

On 18 November 2010 14:53, Senftleben, Stefan, ITSC
<Stefan.Senftleben at itsc.de> wrote:
> hi!
>
> I made these changes to the cib:
>        colocation syslogng2mysql_on_nagios-group inf: nagios-group  syslogng2mysql
>        order nagios-group_then_syslogng2mysql inf: nagios-group:start syslogng2mysql:start
>
> Then went one by one the resource of resource group "nagios-group" down, only "fs_drbd1" and "fs_drbd" were started.
> Maybe the whole resource group was going  down, but the stop timeout values (360s) could lengthened the resource-group stop process.

I think if you had colocation syslogng2mysql_on_nagios-group inf:
syslogng2mysq nagios-group, you would see the stop of the group.
Please take a look at
http://www.clusterlabs.org/doc/en-US/Pacemaker/1.0/html/Pacemaker_Explained/s-resource-colocation.html
and specially at the following text
"The colocation target. The cluster will decide where to put this
resource first and then decide where to put the resource in the rsc
field"

hope it helps

>
> After that the resource group and the new service (in correct order) started. -> very good!
>
> The test "crm resource stop syslogng2mysql" stopped the resource, and again, the resource group. why?
because of the colocation constraint, try what I wrote and let me know.

>
> And in syslog I found that the lrmd tried to start the resource "syslogng2mysql" on the other node, that did not was master of resource-group "nagios-group".
i don't know but I do see few things in your that I don't kind like,
check my previous mail
>
>
>
>
> -----Ursprüngliche Nachricht-----
> Von: Pavlos Parissis [mailto:pavlos.parissis at gmail.com]
> Gesendet: Donnerstag, 18. November 2010 14:14
> An: The Pacemaker cluster resource manager
> Betreff: Re: [Pacemaker] Problems with configuring colocation and start order
>
> On 18 November 2010 13:50, Senftleben, Stefan, ITSC <Stefan.Senftleben at itsc.de> wrote:
>> Hello everybody,
>>
>> I have setup an additional resource „syslogng2mysql“:
>>
>> primitive syslogng2mysql ocf:heartbeat:anything \
>>
>>         params binfile="/etc/init.d/syslog2mysql" cmdline_options="start"
>> pidfile="/var/run/syslog-ng2mysql.pid" \
>>
>>         op start timeout="20s" \
>>
>>         op stop timeout="20s" \
>>
>>         op monitor interval="30s" timeout="60s" depth="0" \
>>
>>         meta target-role="Stopped"
>>
>> That works fine. BUT J I have to setup a colocation to a existing
>> resource group “nagios-group”. (please check the output of “crm
>> configure show” at the end of the mail)
>>
>> I tried to setup up a colocation (syslogng2mysql must run on the same
>> node like resource group “nagios-group”), but after commiting the
>> setup, the resources mailto, nagios, apache2 and ip1 stopped.
>
> only these resources from the group stopped?
>
>> colocation syslogng2mysql_on_nagios-group inf: nagios-group
>> syslogng2mysql
>>
>> Surely I must configure a order to start the resource “syslogng2mysql”
>> after the resource group “nagios group”, but I have no clue how to do it.
>
> the following should work, i don't think you need to specify the start but I do it in order to read the rule easier.
>
> order nagios-group_then_syslogng2mysql inf: nagios-group:start syslogng2mysql:start
>>
>> Could anyone help me, please?
>>
>> Thanks
>>
>> Stefan
>>
>> node lxds05 \
>>
>>         attributes standby="off"
>>
>> node lxds07 \
>>
>>         attributes standby="off"
>>
>> primitive apache2 ocf:heartbeat:apache \
>>
>>         params configfile="/services/etc/apache2/apache2.conf"
>> httpd="/usr/sbin/apache2" \
>>
>>         op monitor interval="15s" \
>>
>>         op start interval="0" timeout="240s" \
>>
>>         op stop interval="0" timeout="240s"
>>
>> primitive drbd_disk ocf:linbit:drbd \
>>
>>         params drbd_resource="nagios" \
>>
>>         op monitor interval="20s" role="Slave" timeout="240s" \
>>
>>         op monitor interval="10s" role="Master" timeout="240s"
>>
>> primitive drbd_disk1 ocf:linbit:drbd \
>>
>>         params drbd_resource="pnp4nagios" \
>>
>>         op monitor interval="20s" role="Slave" timeout="240s" \
>>
>>         op monitor interval="10s" role="Master" timeout="240s"
>>
>> primitive drbd_disk2 ocf:linbit:drbd \
>>
>>         params drbd_resource="services" \
>>
>>         op monitor interval="20s" role="Slave" timeout="240s" \
>>
>>         op monitor interval="10s" role="Master" timeout="240s"
>>
>> primitive fs_drbd ocf:heartbeat:Filesystem \
>>
>>         params device="/dev/drbd0" directory="/usr/local/nagios"
>> fstype="ext3" \
>>
>>         op monitor interval="15s" \
>>
>>         op start interval="0" timeout="240s" \
>>
>>         op stop interval="0" timeout="360s"
>>
>> primitive fs_drbd1 ocf:heartbeat:Filesystem \
>>
>>         params device="/dev/drbd1" directory="/usr/local/pnp4nagios"
>> fstype="ext3" \
>>
>>         op monitor interval="15s" \
>>
>>         op start interval="0" timeout="240s" \
>>
>>         op stop interval="0" timeout="360s"
>>
>> primitive fs_drbd2 ocf:heartbeat:Filesystem \
>>
>>         params device="/dev/drbd2" directory="/services/etc"
>> fstype="ext3" \
>>
>>         op monitor interval="15s" \
>>
>>         op start interval="0" timeout="240s" \
>>
>>         op stop interval="0" timeout="360s"
>>
>> primitive ip1 ocf:heartbeat:IPaddr2 \
>>
>>         params ip="192.168.1.120" nic="eth5" cidr_netmask="24" \
>>
>>         op monitor interval="10s" timeout="20s" \
>>
>>         meta target-role="Started"
>>
>> primitive mailto ocf:heartbeat:MailTo \
>>
>>         params email="username at mail.com" \
>>
>>         op monitor interval="10" timeout="10" depth="0"
>>
>> primitive nagios ocf:naprax:nagios \
>>
>>         params configfile="/usr/local/nagios/etc/nagios.cfg"
>> nagios="/usr/local/nagios/bin/nagios" \
>>
>>         op monitor interval="15s" \
>>
>>         op start interval="0" timeout="240s" \
>>
>>         op stop interval="0" timeout="240s"
>>
>> primitive pingd ocf:pacemaker:pingd \
>>
>>         params host_list="host1 host2 host3 host4" multiplier="100"
>> dampen="5s" \
>>
>>         op monitor interval="15s" timeout="5s"
>>
>> primitive syslogng ocf:heartbeat:syslog-ng \
>>
>>         params configfile="/etc/syslog-ng/syslog-ng.conf"
>> syslog_ng_binary="/usr/sbin/syslog-ng" \
>>
>>         op monitor interval="10s" timeout="60s" depth="0"
>>
>> primitive syslogng2mysql ocf:heartbeat:anything \
>>
>>         params binfile="/usr/local/scripts/system/syslog-ng2mysql"
>> pidfile="/var/run/syslog-ng2mysql.pid" \
>>
>>         op start interval="0" timeout="20s" \
>>
>>         op stop interval="0" timeout="20s" \
>>
>>         op monitor interval="30s" timeout="60s" depth="0" \
>>
>>         meta target-role="Stopped"
>>
>> group nagios-group fs_drbd fs_drbd1 fs_drbd2 ip1 apache2 nagios mailto
>> \
>>
>>         meta target-role="Started"
>>
>> ms ms_drbd drbd_disk \
>>
>>         meta master-max="1" master-node-max="1" clone-max="2"
>> clone-node-max="1" notify="true" target-role="Started"
>>
>> ms ms_drbd1 drbd_disk1 \
>>
>>         meta master-max="1" master-node-max="1" clone-max="2"
>> clone-node-max="1" notify="true"
>>
>> ms ms_drbd2 drbd_disk2 \
>>
>>         meta master-max="1" master-node-max="1" clone-max="2"
>> clone-node-max="1" notify="true"
>>
>> clone pingdclone pingd \
>>
>>         meta globally-unique="false" target-role="Started"
>>
>> clone syslogng-clone syslogng \
>>
>>         meta globally-unique="false" target-role="Started"
>>
>> location cli-prefer-ip1 nagios-group \
>>
>>         rule $id="cli-prefer-rule-ip1" inf: #uname eq lxds07 and
>> #uname eq
>> lxds05
>>
>> location nagios-group_on_connected_node nagios-group \
>>
>>         rule $id="pingd-rule" pingd: defined pingd
>>
>> colocation drbd_on_disks inf: ms_drbd ms_drbd1 ms_drbd2 nagios-group
>>
>> colocation syslogng2mysql_on_nagios-group inf: nagios-group
>> syslogng2mysql
>
> where is the colocation for start fs resource on the Master ms-drbd?
> You need to make sure that the fs_drbd1 and fs_drbd2 run on node with the role master for ms_drbd1 and ms_drbd2.
>
>>
>> order mount_after_drbd inf: ms_drbd:promote nagios-group:start
>>
>> order mount_after_drbd1 inf: ms_drbd1:promote nagios-group:start
>>
>> order mount_after_drbd2 inf: ms_drbd2:promote nagios-group:start
>>
>> property $id="cib-bootstrap-options" \
>>
>>         dc-version="1.0.8-042548a451fce8400660f6031f4da6f0223dd5dd" \
>>
>>         cluster-infrastructure="openais" \
>>
>>         expected-quorum-votes="2" \
>>
>>         stonith-enabled="false" \
>>
>>         no-quorum-policy="ignore"
>>
>> rsc_defaults $id="rsc-options" \
>>
>>         resource-stickiness="100"
>>
>> _______________________________________________
>> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
>> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>>
>> Project Home: http://www.clusterlabs.org Getting started:
>> http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
>> Bugs:
>> http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacema
>> ker
>>
>>
>
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker
>