Hello,
El 27/05/14 23:05, Andrew Beekhof escribió:
But now, we get this line each 5 seconds on corosync syslog:
pacemakerd[23624]: notice: crm_add_logfile: Additional logging available in
/var/log/pacemaker.log
That message should only pop up when pacemakerd has just been started...
someone/thing is calling pacemakerd over and over.
Mmmm is strange. I have no other process than pacemakerd and its forked
processes, and pids are not changing.
But it could be related to pgsql RA. When I stop pgsql resource,
crm_log_init stops happening.
With pgsql RA started, this logs repeats:
----
May 28 12:13:47 lohap2 pacemakerd[25581]: info: crm_log_init:
Changed active directory to /opt/ha/var/lib/heartbeat/cores/root
May 28 12:13:47 lohap2 pacemakerd[25581]: info: crm_xml_cleanup:
Cleaning up memory from libxml2
May 28 12:13:47 lohap2 pgsql(pgsql)[25571]: DEBUG: PostgreSQL is running
as a hot standby.
----
¿Could it be originated by pgsql RA trying to log with ocf_log from
ofc-shellfuncs?
¿Can ocf_log shellfunc originate a call to crm_log_init on pacemakerd?
Also, I see ha_logd is not running. Is this normal?
And this also could be related (repeats each 60 seconds):
----
May 28 11:28:48 lohap1 stonith-ng[23321]: error: crm_abort:
crm_glib_handler: Forked child 13192 to record non-fatal assert at
logging.c:73 : Source ID 2493 was not found when attempting to remove it
May 28 11:28:48 lohap1 stonith-ng[23321]: crit: crm_glib_handler:
GLib: Source ID 2493 was not found when attempting to remove it
May 28 11:28:48 lohap1 stonith-ng[23321]: error: crm_abort:
crm_glib_handler: Forked child 13194 to record non-fatal assert at
logging.c:73 : Source ID 2494 was not found when attempting to remove it
May 28 11:28:48 lohap1 stonith-ng[23321]: crit: crm_glib_handler:
GLib: Source ID 2494 was not found when attempting to remove it
---
Our glib version: 2.40.0 from ubuntu 14.04LTS
And /var/log/pacemaker.log is at info level, but cib outputs LOTS of cib
messages like these (nearly each second):
I'll make the following change for rc2. There is really no need for query
operations to be logged at info.
With those removed, are the cib logs more acceptable?
diff --git a/cib/callbacks.c b/cib/callbacks.c
> [...]
Thank you, tested patch, and log is now more clean ;)
But with this priority we have no info about lrmd or pacemakerd actions.
The default value for PCMK_logpriority, when sending logs to syslog, is
'notice'... that should contain enough information to be useful without being
completely overwhelming.
its only the log file that is logged at level 'info' but you've already
disabled that with PCMK_debugfile=/dev/null.
Ok, thanks for the info.
Now I have logging as I wanted: Only log to syslog (corosync default),
priority INFO.
I have PCMK_debugfile=none, which seems better than /dev/null (seen from
sources).
Thank you
Bernardo
--
APSL
*Bernardo Cabezas Serra*
*Responsable Sistemas*
Camí Vell de Bunyola 37, esc. A, local 7
07009 Polígono de Son Castelló, Palma
Mail: bcabe...@apsl.net
Skype: bernat.cabezas
Tel: 971439771
_______________________________________________
Pacemaker mailing list: Pacemaker@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
Bugs: http://bugs.clusterlabs.org
