[Pacemaker] Two-node cluster, able to function with only a single-node?

[Andrew Beekhof]

On Tue, Dec 13, 2011 at 9:40 AM, Reid, Mike <MBReid at thepei.com> wrote:
> I've been doing a lot of testing while I've been learning
> Pacemaker/Heartbeat, DRBD, OCFS2, etc (Nginx, PHP-FPM, Apache…)  Below I've
> attached my working CIB, but I worry I am missing something obvious (or
> perhaps going about it incorrectly altogether): It seems I cannot start the
> cluster as a single node, and sometimes if the other node is having issues,
> the functioning node does not mount the file system, etc….…until I've set
> the other node to "Standby", restart, etc. Shouldn't I be able to run on a
> single node with HA? I presume I've not configured something correctly,
> possibly I'm using "clone" incorrectly based on my needs?

This looks suspicious:
   primitive resST-NULL stonith:null

We're not going to do anything until we've fenced the second node.
So depending on what this does, it might explain what you're seeing.
>
> Basically, for this two-node web server cluster, I'm running a OCFS2
> Primary/Primary DRBD, with Nginx + PHP-FPM / Apache. I need a little help
> confirming if the CIB configuration below is the appropriate way to handle
> this: As long as one of the node's DRBD is UpToDate (Primary), then ideally
> I'd be able to mount the OCFS2 Filesystem (resFS), start PHP-FPM (resPHP),
> start NGINX (resPROXY), and then Apache (resAPACHE) and handle web traffic.
> Don't worry, not storing any access logs or sessions in DRBD, I'm using
> local disk + memcached for that.
>
> Is my usage of clone / order correct here, or is that perhaps what's
> blocking me from running on a single node more reliably? …I've been able to
> do it by going from a working cluster to setting one node the "standby" or
> simulating a power fault, however I haven't been able to simply start a
> single node and have work as described above.
>
> Software:
> OS: Ubuntu 10.10 (Maverick) / Kernel: 2.6.35
> Corosync 1.2.1
> DRBD 8.3.10
> OCFS2: v1.5.0
>
>
> Pacemakr config / CIB:
>
> node MACHINE1 \
>         attributes standby="off"
> node MACHINE2 \
>         attributes standby="off"
> primitive resAPACHE ocf:heartbeat:apache \
>         params configfile="/usr/local/apache/conf/httpd.conf" \
>         op monitor interval="1min" \
>         op start interval="0" timeout="40" \
>         op stop interval="0" timeout="60"
> primitive resDLM ocf:pacemaker:controld \
>         op monitor interval="120s"
> primitive resDRBD ocf:linbit:drbd \
>         params drbd_resource="repdata" \
>         operations $id="resDRBD-operations" \
>         op monitor interval="20s" role="Master" timeout="120s" \
>         op monitor interval="30s" role="Slave" timeout="120s"
> primitive resFS ocf:heartbeat:Filesystem \
>         params device="/dev/drbd/by-res/repdata" directory="/data"
> fstype="ocfs2" \
>         op monitor interval="120s"
> primitive resO2CB ocf:pacemaker:o2cb \
>         op monitor interval="120s"
> primitive resPHP ocf:heartbeat:anything \
>         params binfile="/usr/local/sbin/php-fpm"
> cmdline_options="--fpm-config /usr/local/etc/php-fpm.conf"
> pidfile="/var/run/php-fpm.pid" \
>         op start interval="0" timeout="20" \
>         op stop interval="0" timeout="30" \
>         op monitor interval="20" \
>         meta target-role="Started"
> primitive resPROXY ocf:heartbeat:nginx \
>         params conffile="/etc/nginx/nginx.conf" \
>         op monitor interval="60s" \
>         op start interval="0" timeout="40" \
>         op stop interval="0" timeout="60"
> primitive resST-NULL stonith:null \
>         params hostlist="MACHINE1 MACHINE2"
> ms msDRBD resDRBD \
>         meta resource-stickines="100" notify="true" master-max="2"
> interleave="true"
> clone cloneDLM resDLM \
>         meta globally-unique="false" interleave="true"
> clone cloneFS resFS \
>         meta interleave="true" ordered="true" target-role="Started"
> clone cloneHTTPD resAPACHE \
>         meta globally-unique="false" interleave="true" ordered="true"
> target-role="Started"
> clone cloneO2CB resO2CB \
>         meta globally-unique="false" interleave="true"
> clone clonePHP resPHP \
>         meta globally-unique="false" interleave="true" ordered="true"
> target-role="Started"
> clone clonePROXY resPROXY \
>         meta globally-unique="false" interleave="true" ordered="true"
> target-role="Started"
> clone fencing resST-NULL
> colocation colDLMDRBD inf: cloneDLM msDRBD:Master
> colocation colFSO2CB inf: cloneFS cloneO2CB
> colocation colO2CBDLM inf: cloneO2CB cloneDLM
> order ordDLMO2CB inf: cloneDLM cloneO2CB
> order ordDRBDDLM inf: msDRBD:promote cloneDLM
> order ordO2CBFS inf: cloneO2CB cloneFS
> order ordPHP inf: cloneFS clonePHP
> order ordPROXY inf: clonePHP clonePROXY
> order ordWEB inf: cloneFS cloneHTTPD
> property $id="cib-bootstrap-options" \
>         dc-version="1.0.9-unknown" \
>         cluster-infrastructure="openais" \
>         stonith-enabled="true" \
>         no-quorum-policy="ignore" \
>         expected-quorum-votes="2"
>
>
>
>
> NOTE: Please disregard stonith:null usage in this example…
>
>
> Any feedback or suggestions would be highly appreciated!
>
>
> _______________________________________________
> 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://bugs.clusterlabs.org
>