[Pacemaker] Best Practice Flush Pacemaker Config / Troubvleshooting unknown error

[Koch, Sebastian]

Hi,

 

currently i am trying to deploy my already running 2 node active/passive
LAMP Cluster to physical machines. I got several problems while
importing the config and therefore i often need to fully flush the
pacemaker config and top aste my edited version. Pacemaker always
complains that the resource is running, which is a safe and correct
behaviour. Would like to know if there is a fast way to flush config and
import new one without stopping all resource and so on.

 

And my second question is how i can troubleshoot unknown erros. When i
try to start my drbd resource it just says unknown error. I am able to
mount it by hand ist a XFS Filesystem with underlying LVM.

 

Normally i troubleshoot like this:

 

1.       Crm_mon

2.       Tail -fn 100 /var/log/syslog

3.       Crm_verify -LV

 

Does somebody have a advise how to troubleshoot this tricky errors?

 

 

Crm configure show

 

node pilot01-node1 \

        attributes standby="off"

node pilot01-node2 \

        attributes standby="off"

primitive drbd_pilot0 ocf:linbit:drbd \

        params drbd_resource="pilot0" \

        operations $id="drbd_pilot0-operations" \

        op monitor interval="60"

primitive pinggw ocf:pacemaker:pingd \

        params host_list="10.1.1.162" multiplier="200" \

        op monitor interval="10"

primitive res_Apache lsb:apache2 \

        operations $id="res_Apache-operations" \

        op monitor interval="15" timeout="20" start-delay="15" \

        meta target-role="Started" is-managed="true"

primitive res_ClusterIP ocf:heartbeat:IPaddr2 \

        params iflabel="ClusterIP" ip="10.1.1.12" \

        operations $id="res_ClusterIP_1-operations" \

        op monitor interval="10" start-delay="0"

primitive res_Filesystem ocf:heartbeat:Filesystem \

        params fstype="xfs" directory="/mnt/cluster" device="/dev/drbd0"
\

        operations $id="res_Filesystem-operations" \

        op monitor interval="20" start-delay="0"

primitive res_MySQL lsb:mysql \

        operations $id="res_MySQL-operations" \

        op monitor interval="15" timeout="60" start-delay="15"

group grp_MySQL res_Filesystem res_ClusterIP res_MySQL res_Apache \

        meta target-role="Started"

ms ms_drbd_mysql0 drbd_pilot0 \

        meta master-max="1" master-node-max="1" clone-max="2"
clone-node-max="1" notify="true"

clone cl-pinggw pinggw \

        meta globally-unique="false"

location cli-prefer-grp_MySQL grp_MySQL \

        rule $id="cli-prefer-rule-grp_MySQL" inf: #uname eq
pilot01-node2

location drbd-fence-by-handler-ms_drbd_mysql0 ms_drbd_mysql0 \

        rule $id="drbd-fence-by-handler-rule-ms_drbd_mysql0"
$role="Master" -inf: #uname ne pilot01-node1

location grp_MySQL-with-pinggw grp_MySQL \

        rule $id="grp_MySQL-with-pinggw-rule-1" -inf: not_defined pingd
or pingd lte 0

colocation col_drbd_on_mysql inf: grp_MySQL ms_drbd_mysql0:Master

order mysql_after_drbd inf: ms_drbd_mysql0:promote grp_MySQL:start

property $id="cib-bootstrap-options" \

        expected-quorum-votes="3" \

        stonith-enabled="false" \

        no-quorum-policy="ignore" \

        dc-version="1.0.8-2c98138c2f070fcb6ddeab1084154cffbf44ba75" \

        cluster-infrastructure="openais" \

        last-lrm-refresh="1274352045" \

        symmetric-cluster="true" \

        default-action-timeout="120s"

rsc_defaults $id="rsc-options" \

        resource-stickiness="100"

 

 

cat /proc/drbd

 

cat /proc/drbd

version: 8.3.7 (api:88/proto:86-91)

GIT-hash: ea9e28dbff98e331a62bcbcc63a6135808fe2917 build by
root at prolog01-pilot1, 2010-06-07 17:34:47

 0: cs:Connected ro:Primary/Secondary ds:UpToDate/UpToDate C r----

    ns:0 nr:0 dw:0 dr:72 al:0 bm:0 lo:0 pe:0 ua:0 ap:0 ep:1 wo:b oos:0

 

cat /etc/drbd.d/pilot.res

 

resource pilot0 {

 

 

syncer {

        rate 125M;

}

        device    /dev/drbd0;

        disk      /dev/cluster/block0;

        meta-disk internal;

 

        on pilot01-node1 {

                address   192.168.100.1:7789;

        }

 

        on pilot01-node2 {

                address   192.168.100.2:7789;

        }

 

        disk {

                fencing resource-only;

        }

 

        handlers {

                fence-peer "/usr/lib/drbd/crm-fence-peer.sh";

                after-resync-target "/usr/lib/drbd/crm-unfence-peer.sh";

        }

}

 

 

 

Thanks in advance.

Best Regards

 

Sebastian Koch
                                                         

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clusterlabs.org/pipermail/pacemaker/attachments/20100611/028bc861/attachment.html>