[Pacemaker] Help needed crm did not wirte the resources permanent
Winkler Norbert
norbert.winkler at karmel.at
Mon Mar 15 21:07:38 UTC 2010
Thank you for answering
I tried a lot but i din't suceed
In this mail are less write errors (i hope so)
Here are different config files
vmstore1:/etc # crm_mon -1f
============
Last updated: Mon Mar 15 21:22:57 2010
Stack: openais
Current DC: vmstore2 - partition with quorum
Version: 1.0.7-d3fa20fc76c7947d6de66db7e52526dc6bd7d782
2 Nodes configured, 2 expected votes
0 Resources configured.
============
Online: [ vmstore1 vmstore2 ]
Migration summary:
* Node vmstore1:
* Node vmstore2:
----------------------------------------------------------------
crm(live)# configure show
ERROR: /root/.crm_help_index open: [Errno 2] No such file or directory:
'/root/.crm_help_index'
ERROR: extensive help system is not available
node vmstore1 \
attributes standby="false"
node vmstore2 \
attributes standby="false"
property $id="cib-bootstrap-options" \
dc-version="1.0.7-d3fa20fc76c7947d6de66db7e52526dc6bd7d782" \
stonith-action="poweroff" \
expected-quorum-votes="2" \
cluster-infrastructure="openais" \
node-health-red="0" \
stonith-enabled="false"
rsc_defaults $id="rsc_defaults-options" \
is-managed="true"
------------------------------------------------------
In the first section you can see the resource in the cbl and when i make
as next a cibadmin -Q the resource is gone again
-------------------------------------------------
vmstore1:/etc # cibadmin -C -o resources -x resourceIP.xml
Call cib_create failed (-47): Update does not conform to the configured
schema/DTD
<cib validate-with="pacemaker-1.0" crm_feature_set="3.0.1"
have-quorum="1" dc-uuid="vmstore2" admin_epoch="0" epoch="101"
num_updates="1">
<configuration>
<crm_config>
<cluster_property_set id="cib-bootstrap-options">
<nvpair id="cib-bootstrap-options-dc-version" name="dc-version"
value="1.0.7-d3fa20fc76c7947d6de66db7e52526dc6bd7d782"/>
<nvpair id="nvpair-638aff86-19c8-4d62-8f30-ef419416fa5a"
name="stonith-action" value="poweroff"/>
<nvpair id="cib-bootstrap-options-expected-quorum-votes"
name="expected-quorum-votes" value="2"/>
<nvpair id="cib-bootstrap-options-cluster-infrastructure"
name="cluster-infrastructure" value="openais"/>
<nvpair id="nvpair-16ff93f5-e0e7-48a5-a7d8-a0778d791af2"
name="node-health-red" value="0"/>
<nvpair id="nvpair-c729123b-3fa4-4f78-88e1-2e2644e47310"
name="stonith-enabled" value="false"/>
</cluster_property_set>
</crm_config>
<nodes>
<node id="vmstore1" type="normal" uname="vmstore1">
<instance_attributes id="nodes-vmstore1">
<nvpair id="standby-vmstore1" name="standby" value="false"/>
</instance_attributes>
</node>
<node id="vmstore2" type="normal" uname="vmstore2">
<instance_attributes id="nodes-vmstore2">
<nvpair id="standby-vmstore2" name="standby" value="false"/>
</instance_attributes>
</node>
</nodes>
<resources>
<primitive id="resIP" class="ocf" type="IPaddr2" provider="heartbeat">
<instances_attributes id="resIP_instance_attrs">
<nvpair id="resIP_IP" name="ip" value="192.168.1.23"/>
<nvpair id="resIP_NIC" name="nic" value="eth1"/>
</instances_attributes>
</primitive>
</resources>
<constraints/>
<op_defaults>
<meta_attributes id="op_defaults-options"/>
</op_defaults>
<rsc_defaults>
<meta_attributes id="rsc_defaults-options">
<nvpair id="nvpair-68b16f0e-1e5c-4dd5-85a4-28a14ea385fc"
name="is-managed" value="true"/>
</meta_attributes>
</rsc_defaults>
</configuration>
<status>
<node_state uname="vmstore1" ha="active" in_ccm="true" crmd="online"
join="member" expected="member" shutdown="0" id="vmstore1"
crm-debug-origin="do_state_transition">
<lrm id="vmstore1">
<lrm_resources/>
</lrm>
<transient_attributes id="vmstore1">
<instance_attributes id="status-vmstore1">
<nvpair id="status-vmstore1-probe_complete" name="probe_complete"
value="true"/>
</instance_attributes>
</transient_attributes>
</node_state>
<node_state uname="vmstore2" ha="active" in_ccm="true" crmd="online"
join="member" expected="member" shutdown="0" id="vmstore2"
crm-debug-origin="do_state_transition">
<transient_attributes id="vmstore2">
<instance_attributes id="status-vmstore2">
<nvpair id="status-vmstore2-probe_complete" name="probe_complete"
value="true"/>
</instance_attributes>
</transient_attributes>
<lrm id="vmstore2">
<lrm_resources/>
</lrm>
</node_state>
</status>
</cib>
vmstore1:/etc # cibadmin -Q
<cib validate-with="pacemaker-1.0" crm_feature_set="3.0.1"
have-quorum="1" dc-uuid="vmstore2" admin_epoch="0" epoch="100"
num_updates="5">
<configuration>
<crm_config>
<cluster_property_set id="cib-bootstrap-options">
<nvpair id="cib-bootstrap-options-dc-version" name="dc-version"
value="1.0.7-d3fa20fc76c7947d6de66db7e52526dc6bd7d782"/>
<nvpair id="nvpair-638aff86-19c8-4d62-8f30-ef419416fa5a"
name="stonith-action" value="poweroff"/>
<nvpair id="cib-bootstrap-options-expected-quorum-votes"
name="expected-quorum-votes" value="2"/>
<nvpair id="cib-bootstrap-options-cluster-infrastructure"
name="cluster-infrastructure" value="openais"/>
<nvpair id="nvpair-16ff93f5-e0e7-48a5-a7d8-a0778d791af2"
name="node-health-red" value="0"/>
<nvpair id="nvpair-c729123b-3fa4-4f78-88e1-2e2644e47310"
name="stonith-enabled" value="false"/>
</cluster_property_set>
</crm_config>
<nodes>
<node id="vmstore1" type="normal" uname="vmstore1">
<instance_attributes id="nodes-vmstore1">
<nvpair id="standby-vmstore1" name="standby" value="false"/>
</instance_attributes>
</node>
<node id="vmstore2" type="normal" uname="vmstore2">
<instance_attributes id="nodes-vmstore2">
<nvpair id="standby-vmstore2" name="standby" value="false"/>
</instance_attributes>
</node>
</nodes>
<resources/>
<constraints/>
<op_defaults>
<meta_attributes id="op_defaults-options"/>
</op_defaults>
<rsc_defaults>
<meta_attributes id="rsc_defaults-options">
<nvpair id="nvpair-68b16f0e-1e5c-4dd5-85a4-28a14ea385fc"
name="is-managed" value="true"/>
</meta_attributes>
</rsc_defaults>
</configuration>
<status>
<node_state uname="vmstore1" ha="active" in_ccm="true" crmd="online"
join="member" expected="member" shutdown="0" id="vmstore1"
crm-debug-origin="do_state_transition">
<lrm id="vmstore1">
<lrm_resources/>
</lrm>
<transient_attributes id="vmstore1">
<instance_attributes id="status-vmstore1">
<nvpair id="status-vmstore1-probe_complete" name="probe_complete"
value="true"/>
</instance_attributes>
</transient_attributes>
</node_state>
<node_state uname="vmstore2" ha="active" in_ccm="true" crmd="online"
join="member" expected="member" shutdown="0" id="vmstore2"
crm-debug-origin="do_state_transition">
<transient_attributes id="vmstore2">
<instance_attributes id="status-vmstore2">
<nvpair id="status-vmstore2-probe_complete" name="probe_complete"
value="true"/>
</instance_attributes>
</transient_attributes>
<lrm id="vmstore2">
<lrm_resources/>
</lrm>
</node_state>
</status>
</cib>
------------------------------------------------------------------
I dont know what do make maybe using opensuse 11.1 because it is in the
book ?? Maybe a lot of work but we are a monestary
(karmeliten Linz Austria)and we dont have enough money for datacore so
we hoped we can manage this with pacemaker and drbd and openiscsi
but i am a one man team and some times this is really hard
So if possible please have a look again on my config files
Thank you in advance
Schwartzkopff schrieb:
> Am Montag, 15. März 2010 14:11:58 schrieb Norbert Winkler:
>
>> Hello my name is Norbert
>> and i want to make a pacemaker cluster to use openiscsitarget with drbd
>> for a vmware storage ( I have 3 luns a 2TB)
>> i have synced the drbd luns and they seems to work well.
>> My problem is with the pacemaker crm
>>
>> *I am using opensuse 11.2 with default kernel*
>> on both nodes and i am logged in as root
>> I have also the user hacluster and the group haclient which i use to
>> login into gui
>> After i fail to make a pacemaker cluster with opensuse 11.2
>> repositories i try the repository from clusterlab:
>> I am using:
>> cluster-glue 1.0.3-1 (x86_64) from clusterlab-rep
>> corosync 1.2.0-1 (x86_64) from clusterlab-rep
>> heartbeat 3.0.2-2 (x86_64) from clusterlab-rep
>> libcorosync 1.2.1-1 (x86-64) from clustrerlab-rep
>> libglue2 1.0.3-1 (x86_64) from clusterlab-rep
>> libopenais2 1.1.0-1 (x86_64) from clusterlab-rep
>> libpacemaker3 1.0.7-4 (x86_64) from clusterlab-rep
>> openais 1.1.0-1 (x86_64) from clusterlab-rep
>> pacemaker 1.0.7-4 (x86_64) from clusterlab-rep
>> resource-agents 1.0.1-1 (x86_64) from clusterlab-rep
>> pacemaker-pygui 1.4.15-1 (x86_64) opensuse 11.2 repo (no gui available
>> from clusterlab rep)
>>
>> At the first moment everything seems to work well
>> login into console:
>> Cluster have qourum green point
>> vmstore1 (dc) online(dc)standby (with green point)
>> vmstore2 online-standby with green point
>> Resources with green point
>>
>> I also can chance status to active for both nodes.
>>
>> My problems occurs when i try to install a resource.
>> At first with the gui:
>> *The gui fails when i try to use Resources to add primitive class ocf -
>> only a kill -9 corosync and the a new login into hb_gui is possible
>> after i try this .
>> *
>>
>
> Please first try the crm subshell if the GUI does not work.
>
>
>> THe configuration is standard. Resource defaults are empty and i tried
>> stonith enabled and disabled.
>>
>
> Please set stonith-enabled=alse"
>
>
>> *When i let stonith enabled and i try to install a resource the
>> commandline cibadmin (Ipaddr2) i tells me that it can not run this
>> resource because no stonith resource is configured. when i disable
>> stonith it seems to insert the ressource into the crm because it shows
>> me the complete config file with the resource added. BUt when i try
>> crm_resource -L then the system tells me that no resource added.
>> I am using the script from the book Clusterbau: Hochverfügbarkeit mit
>> pacemaker.. think the origninal english title should be: buliding
>> cluster: with pacemaker.. I am using the newest edition of this book
>> from Dr.Michael Schwartzkopff.*
>>
>
> Thanks ;-)
>
> Any more descritpive error reports? output of crm_mon -1f?
>
>
>> I dont know how to go on and i am a little bit frustradet because i
>> thougt i was near the target and at the moment i feel miles away
>> What i think possible problems are:
>> 1. Maybe i have problems with the user rights because the
>> /usr/lib/ocf/resource.d/heartbeat belongs to the user and group root
>> wtih drwxr-xr-x
>>
>
> Unlikely.
>
>
>> 2. Maybe the path to the heartbeat resources is wrong maybe corosync
>> wants another path to the resources
>>
>
> Very Unlikely.
>
>
>> 3. Maybe i had to configure resource defaults first but i dont know how.
>>
>
> Should work out of the box.
>
>
>> 4. Maybe I must configure stonith resource firse but i dont know how.
>>
>
> set stonith-enabled=alse"
>
>
>> 5. MaybeThe crmadmin couldnt wright the rosource in database because i
>> have a right problem
>>
>
> no.
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20100315/1351739b/attachment-0002.htm>
More information about the Pacemaker
mailing list