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

Mailing List Archive: Linux-HA: Pacemaker

set cib_writes_enabled = TRUE during process running

 

 

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


ikedaj at intellilink

Nov 19, 2009, 10:12 PM

Post #1 of 4 (306 views)
Permalink
set cib_writes_enabled = TRUE during process running

Hi,

I run PostgreSQL as one of resources of Pacemaker.
Its log file become really large, and the disk for /var was full
unfortunately.
cib process failed to write cib.xml, and there was the bottom message in
ha-log.

ERROR: cib_diskwrite_complete: Disabling disk writes after write failure

cib.xml was 0 bite at that time.
Even after recovering of the disk full, cib didn't start to write the file.
It seems that cib_writes_enabled = FALSE,
so cib didn't write CIB on disk.

Is it possible to set cib_writes_enabled = TRUE during pacemaker running?
Is the restart of pacemaker only way to recover this situation?

Thanks,
Junko

_______________________________________________
Pacemaker mailing list
Pacemaker [at] oss
http://oss.clusterlabs.org/mailman/listinfo/pacemaker


andrew at beekhof

Nov 19, 2009, 11:12 PM

Post #2 of 4 (276 views)
Permalink
Re: set cib_writes_enabled = TRUE during process running [In reply to]

On Fri, Nov 20, 2009 at 7:12 AM, Junko IKEDA <ikedaj [at] intellilink> wrote:
> Hi,
>
> I run PostgreSQL as one of resources of Pacemaker.
> Its log file become really large, and the disk for /var was full
> unfortunately.
> cib process failed to write cib.xml, and there was the bottom message in
> ha-log.
>
> ERROR: cib_diskwrite_complete: Disabling disk writes after write failure
>
> cib.xml was 0 bite at that time.
> Even after recovering of the disk full, cib didn't start to write the file.
> It seems that cib_writes_enabled = FALSE,
> so cib didn't write CIB on disk.
>
> Is it possible to set cib_writes_enabled = TRUE during pacemaker running?

hmmm, i hadn't thought about that situation to be honest.
might be something good to add.

you might be able to attach to the running cib with gdb and set it that way...

> Is the restart of pacemaker only way to recover this situation?
>
> Thanks,
> Junko
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker [at] oss
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>

_______________________________________________
Pacemaker mailing list
Pacemaker [at] oss
http://oss.clusterlabs.org/mailman/listinfo/pacemaker


andrew at beekhof

Nov 19, 2009, 11:12 PM

Post #3 of 4 (266 views)
Permalink
Re: set cib_writes_enabled = TRUE during process running [In reply to]

On Fri, Nov 20, 2009 at 7:12 AM, Junko IKEDA <ikedaj [at] intellilink> wrote:
> Hi,
>
> I run PostgreSQL as one of resources of Pacemaker.
> Its log file become really large, and the disk for /var was full
> unfortunately.
> cib process failed to write cib.xml, and there was the bottom message in
> ha-log.
>
> ERROR: cib_diskwrite_complete: Disabling disk writes after write failure
>
> cib.xml was 0 bite at that time.
> Even after recovering of the disk full, cib didn't start to write the file.
> It seems that cib_writes_enabled = FALSE,
> so cib didn't write CIB on disk.
>
> Is it possible to set cib_writes_enabled = TRUE during pacemaker running?

hmmm, i hadn't thought about that situation to be honest.
might be something good to add.

you might be able to attach to the running cib with gdb and set it that way...

> Is the restart of pacemaker only way to recover this situation?
>
> Thanks,
> Junko
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker [at] oss
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>

_______________________________________________
Pacemaker mailing list
Pacemaker [at] oss
http://oss.clusterlabs.org/mailman/listinfo/pacemaker


ikedaj at intellilink

Nov 19, 2009, 11:43 PM

Post #4 of 4 (275 views)
Permalink
Re: set cib_writes_enabled = TRUE during process running [In reply to]

Hi,

On Fri, 20 Nov 2009 16:12:07 +0900, Andrew Beekhof <andrew [at] beekhof>
wrote:

> hmmm, i hadn't thought about that situation to be honest.
> might be something good to add.

ok, I see.
cib could restart it write process after the restart of pacemaker,
so I'll recover this situation with this way for now.

I create an enhancement bugzilla for this.
http://developerbugs.linux-foundation.org/show_bug.cgi?id=2229

Thanks,
Junko

_______________________________________________
Pacemaker mailing list
Pacemaker [at] oss
http://oss.clusterlabs.org/mailman/listinfo/pacemaker

Linux-HA pacemaker 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.