Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Linux-HA: Users

why 'crm' and 'pengine' direcotries are empty on one working/healthy node ?

 

 

Linux-HA users RSS feed   Index | Next | Previous | View Threaded


ibnerazi at yahoo

Nov 25, 2009, 11:00 AM

Post #1 of 4 (878 views)
Permalink
why 'crm' and 'pengine' direcotries are empty on one working/healthy node ?

OS: SLES 10 SP2

rpm -qa |grep heartbeat
heartbeat-pils-2.1.4-0.15.5
heartbeat-2.1.4-0.15.5
heartbeat-cmpi-2.1.4-0.15.5
heartbeat-ldirectord-2.1.4-0.15.5
heartbeat-stonith-2.1.4-0.15.5

its a two node active/passive test cluster.
both nodes are working fine... if active node shutdowns, the passive node
becomes active and runs all the resources properly.

but on one node the /var/lib/heartbeat/crm and /var/lib/heartbeat/pengine
is empty... but the node is working quite fine(i.e this node can
resources etc)..  how/why ?

on the other node both 'crm' and 'pengine' directories are populated.

whats wrong ?

Regards



_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems


ibnerazi at yahoo

Nov 26, 2009, 11:30 PM

Post #2 of 4 (827 views)
Permalink
Re: why 'crm' and 'pengine' direcotries are empty on one working/healthy node ? [In reply to]

--- On Wed, 11/25/09, Muhammad Sharfuddin <ibnerazi [at] yahoo> wrote:
> OS: SLES 10 SP2
>
> rpm -qa |grep heartbeat
> heartbeat-pils-2.1.4-0.15.5
> heartbeat-2.1.4-0.15.5
> heartbeat-cmpi-2.1.4-0.15.5
> heartbeat-ldirectord-2.1.4-0.15.5
> heartbeat-stonith-2.1.4-0.15.5
>
> its a two node active/passive test cluster.
> both nodes are working fine... if active node shutdowns, the passive node
> becomes active and runs all the resources properly.
>
> but on one node the /var/lib/heartbeat/crm and /var/lib/heartbeat/pengine
> is empty... but the node is working quite fine(i.e this node can
> resources etc).. how/why ?
>
> on the other node both 'crm' and 'pengine' directories are populated.
>
>whats wrong ?

if I only start heartbeat on the node where 'crm' and 'pengine' directories are empty, crm_mon/hb_gui does not show any resources, but when I also start the service on the node where 'pengine' and 'crm' directories are populated, crm_mon/hb_gui starts showing resources from both nodes.

Regards






_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems


andrew at beekhof

Nov 26, 2009, 11:51 PM

Post #3 of 4 (830 views)
Permalink
Re: why 'crm' and 'pengine' direcotries are empty on one working/healthy node ? [In reply to]

On Fri, Nov 27, 2009 at 8:30 AM, Muhammad Sharfuddin <ibnerazi [at] yahoo> wrote:
> --- On Wed, 11/25/09, Muhammad Sharfuddin <ibnerazi [at] yahoo> wrote:
>> OS: SLES 10 SP2
>>
>> rpm -qa |grep heartbeat
>> heartbeat-pils-2.1.4-0.15.5
>> heartbeat-2.1.4-0.15.5
>> heartbeat-cmpi-2.1.4-0.15.5
>> heartbeat-ldirectord-2.1.4-0.15.5
>> heartbeat-stonith-2.1.4-0.15.5
>>
>> its a two node active/passive test cluster.
>> both nodes are working fine... if active node shutdowns, the passive node
>> becomes active and runs all the resources properly.
>>
>> but on one node the /var/lib/heartbeat/crm and /var/lib/heartbeat/pengine
>> is empty... but the node is working quite fine(i.e this node can
>> resources etc).. how/why ?
>>
>> on the other node both 'crm' and 'pengine' directories are populated.
>>
>>whats wrong ?
>
> if I only start heartbeat on the node where 'crm' and 'pengine' directories are empty, crm_mon/hb_gui does not show any resources, but when I also start the service on the node where 'pengine' and 'crm' directories are populated, crm_mon/hb_gui starts showing resources from both nodes.

Did you try reading the logs?
_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems


ibnerazi at yahoo

Nov 27, 2009, 4:26 AM

Post #4 of 4 (873 views)
Permalink
Re: why 'crm' and 'pengine' direcotries are empty on one working/healthy node ? [In reply to]

> --- On Thu, 11/26/09, Andrew Beekhof <andrew [at] beekhof> wrote:
>> On Fri, Nov 27, 2009 at 8:30 AM, Muhammad Sharfuddin
>> <ibnerazi [at] yahoo> wrote:
>>> --- On Wed, 11/25/09, Muhammad Sharfuddin
>>> <ibnerazi [at] yahoo> wrote:
>>>
>>> OS: SLES 10 SP2
>>>
>>> rpm -qa |grep heartbeat
>>> heartbeat-pils-2.1.4-0.15.5
>>> heartbeat-2.1.4-0.15.5
>>> heartbeat-cmpi-2.1.4-0.15.5
>>> heartbeat-ldirectord-2.1.4-0.15.5
>>> heartbeat-stonith-2.1.4-0.15.5
>>>
>>> its a two node active/passive test cluster.
>>> both nodes are working fine... if active node shutdowns, the passive node
>>> becomes active and runs all the resources properly.
>>>
>>> but on one node the /var/lib/heartbeat/crm and /var/lib/heartbeat/pengine
>>> is empty... but the node is working quite fine(i.e this node can
>>> resources etc).. how/why ?
>>>
>>> on the other node both 'crm' and 'pengine' directories are populated.
>>>
>>> whats wrong ?
>>
>> if I only start heartbeat on the node where 'crm' and 'pengine'
>> directories are empty, crm_mon/hb_gui does not show any resources,
>> but when I also start the service on the node where 'pengine' and
>> 'crm' directories are populated, crm_mon/hb_gui starts showing
>> resources from both nodes.
>>>

Hi Andrew Beekhof,

> Did you try reading the logs?
Not before, but yes now.
issue resolved, thanks for pointing me towards logs.
actually by mistake 'crm' and 'pengine' were owned by root.

from /var/log/messages

Cannot open /var/lib/heartbeat/crm/cib.xml for writing: Permission denied
write_cib_contents: Changes couldn't be written to disk
cib_diskwrite_complete: Disk write failed: status=256, signo=0, exitcode=1
cib_diskwrite_complete: Disabling disk writes after write failure
Cannot chdir to [/var/lib/heartbeat/cores/hacluster]: Permission denied
crm_is_writable: /var/lib/heartbeat/pengine must be owned and r/w by user hacluster
Cannot chdir to [/var/lib/heartbeat/cores/hacluster]: Permission denied
Cannot open /var/lib/heartbeat/pengine/pe-input-0.bz2 for writing: Permission denied
write_last_sequence: Cannout open series file /var/lib/heartbeat/pengine/pe-input.last for writing

issue now resolved by granting owner ship to 'hacluster.haclient'

Regards
Muhammad Sharfuddin




_______________________________________________
Linux-HA mailing list
Linux-HA [at] lists
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Linux-HA users RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.