[Pacemaker] SLES11+HAE: Resources on a single node with two configured?

Aleksey Zholdak aleksey at zholdak.com
Thu May 6 08:27:35 UTC 2010


Andrew,

> These logs:
> 
> Apr 28 10:08:11 sles2 crmd: [6107]: info: do_state_transition: State
> transition S_INTEGRATION -> S_ELECTION [ input=I_ELECTION
> cause=C_FSA_INTERNAL origin=crm_timer_popped ]
> Apr 28 10:08:11 sles2 crmd: [6107]: info: do_state_transition: State
> transition S_ELECTION -> S_INTEGRATION [ input=I_ELECTION_DC
> cause=C_FSA_INTERNAL origin=do_election_check ]
I'm sorry, but these lines I do not say anything ...

> almost only even happen when there is a firewall in the way.
> Yes, even on a single node firewalls can be a problem.
After turning off filrewall run only sbd_fence, the remaining resources 
will not run.
At the same time, see the following:

sles2:/ # crm_verify -L -V
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource 
sbd_fense:1 cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource dlm:1 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource clvm:1 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource 
eCluster_vg0:1 cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource o2cb:1 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource fs:1 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource pingd:1 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource int_ip 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource ext_ip 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource mysql 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: native_color: Resource apache 
cannot run anywhere
crm_verify[15054]: 2010/05/06_11:19:50 WARN: stage6: Scheduling Node sles1 
for STONITH
Warnings found during check: config may not be valid


STONITH trying to "kill" the first node, but does not understand that it is 
not and resources do not start.
And in this state, we are infinite.


> What version of openais is this btw?
openais-0.80.3-26.8.1

> 2010/4/28 Aleksey Zholdak <aleksey at zholdak.com>:
>> Andrew Beekhof пишет:
>>> 2010/4/26 Aleksey Zholdak <aleksey at zholdak.com>:
>>>> Andrew Beekhof:
>>>> What do you mean here? Need logs? I am pleased to show you! Only their
>>>> size
>>>> will occupy much space here, tell me what to choose and what to look ...
>>> Better to just send everything and compress them first.
>> Log and openais configuration I attached to this letter. The situation is
>> this: run only one of two nodes configured. Resources are not running, even
>> DC is not selected. Help, please, understand what the problem ...
>>
>> node sles1
>> node sles2
>> primitive apache ocf:heartbeat:apache \
>>        meta target-role="Started" \
>>        params configfile="/eCluster/configures/apache2/httpd.conf"
>> httpd="/usr/sbin/httpd2" statusurl="http://127.0.0.1:8088/server-status" \
>>        op monitor interval="10s" timeout="60s"
>> primitive clvm ocf:lvm2:clvmd \
>>        params daemon_timeout="30"
>> primitive dlm ocf:pacemaker:controld \
>>        op monitor interval="120s"
>> primitive eCluster_vg0 ocf:heartbeat:LVM \
>>        params volgrpname="eCluster_vg0"
>> primitive ext_ip ocf:heartbeat:IPaddr2 \
>>        meta target-role="Started" \
>>        params ip="z.z.z.z" nic="vlan101" cidr_netmask="24" \
>>        op monitor interval="10s" timeout="20s"
>> primitive fs ocf:heartbeat:Filesystem \
>>        params device="/dev/eCluster_vg0/eCluster_lv0" directory="/eCluster"
>> fstype="ocfs2" \
>>        op monitor interval="120s"
>> primitive int_ip ocf:heartbeat:IPaddr2 \
>>        meta target-role="Started" \
>>        params ip="172.16.0.20" nic="vlan100" cidr_netmask="22" \
>>        op monitor interval="10s" timeout="20s"
>> primitive mysql ocf:heartbeat:mysql \
>>        meta target-role="Started" migration-threshold="100" \
>>        params binary="/usr/bin/mysqld_safe" config="/etc/my.cnf"
>> datadir="/eCluster/mysql" user="mysql" pid="/eCluster/mysql/mysql.pid"
>> socket="/var/lib/mysql/mysql.sock" enable_creation="0"
>> test_table="heartbeattest.connectioncheck" test_user="heartbeattest"
>> test_passwd="sdjh287js2" \
>>        op monitor interval="10s" timeout="60s"
>> primitive o2cb ocf:ocfs2:o2cb \
>>        op monitor interval="120s"
>> primitive pingd ocf:pacemaker:pingd \
>>        params host_list="172.16.3.250" multiplier="100" \
>>        op monitor interval="5s" timeout="20s"
>> primitive sbd_fense stonith:external/sbd \
>>        params sbd_device="/dev/mapper/SBD"
>> group web_cluster int_ip ext_ip mysql apache
>> clone clvm-clone clvm \
>>        meta target-role="Started" interleave="true" ordered="true"
>> clone dlm-clone dlm \
>>        meta target-role="Started" globally-unique="false" interleave="true"
>> ordered="true"
>> clone eCluster_vg0-clone eCluster_vg0 \
>>        meta interleave="true" ordered="true"
>> clone fs-clone fs \
>>        meta interleave="true" ordered="true"
>> clone o2cb-clone o2cb \
>>        meta globally-unique="false" interleave="true"
>> clone pingd-clone pingd \
>>        meta globally-unique="false" ordered="true"
>> clone sbd-clone sbd_fense \
>>        meta target-role="Started" ordered="true"
>> location location-web_cluster_on_connected_node web_cluster \
>>        rule $id="location-web_cluster_on_connected_node-rule" -inf:
>> not_defined pingd or pingd lte 0
>> location location-web_cluster_on_preffered_sles2 web_cluster \
>>        rule $id="location-web_cluster_on_preffered_sles2-rule" 50: #uname eq
>> sles2
>> colocation colo-clvm inf: clvm-clone dlm-clone
>> colocation colo-eCluster_vg0 inf: eCluster_vg0-clone clvm-clone
>> colocation fs-with-o2cb inf: fs-clone o2cb-clone
>> colocation o2cb-with-eCluster_vg0 inf: o2cb-clone eCluster_vg0-clone
>> colocation pingd-with-fs inf: pingd-clone fs-clone
>> colocation web_cluster-with-fs inf: web_cluster fs-clone
>> order order-clvm inf: dlm-clone clvm-clone
>> order order-eCluster_vg0 inf: clvm-clone eCluster_vg0-clone
>> order order-fs inf: o2cb-clone fs-clone
>> order order-o2cb inf: eCluster_vg0-clone o2cb-clone
>> order order-pingd inf: fs-clone pingd-clone
>> order order-web_cluster inf: fs-clone web_cluster
>> property $id="cib-bootstrap-options" \
>>        dc-version="1.0.6-c48e3360eb18c53fd68bb7e7dbe39279ccbc0354" \
>>        cluster-infrastructure="openais" \
>>        expected-quorum-votes="2" \
>>        stonith-enabled="true" \
>>        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
>>


-- 

С уважением,
ЖОЛДАК Алексей

ICQ   150074
MSN   aleksey at zholdak.com
Skype aleksey.zholdak
Voice +380442388043




More information about the Pacemaker mailing list