I am playing with the interval setting, and still messages seem to get stuck or delivery continues only when firing a new message a new client, in which case the previous messages remain in the disk queue but never get forwarded, so they are "discarded" in a sense.
Any idea how to move forward? -Sivan On Wed, Nov 24, 2010 at 8:50 AM, Rainer Gerhards <[email protected]> wrote: >> -----Original Message----- >> From: [email protected] [mailto:rsyslog- >> [email protected]] On Behalf Of [email protected] >> Sent: Tuesday, November 23, 2010 7:56 PM >> To: rsyslog-users >> Subject: Re: [rsyslog] messages stuck on a client, using RELP between >> client and server, after a restart of the server. >> >> On Tue, 23 Nov 2010, Sivan Greenberg wrote: >> >> > it is already set to -1 as i understood this controls infinite >> retries >> > but what about intervals ? >> >> retry count is set to -1, but you don't have retry interval set. >> interval >> sets how frequently you retry (or at least that's how I read the >> documentation) > > Yes, that's right. There are two settings. Rsyslog does not try for each and > every message, but suspends the action for some time, in order to guard the > rest of the system against failure (think about the timeout associated with > trying to connect to a gone-down remote system). The suspension window is > increased after each unsuccessful retry in sequence (up to some upper limit). > Think of it as graceful degradation of some system components in order to > preserve overall health. > > Rainer >> >> David Lang >> >> > On Tue, Nov 23, 2010 at 8:20 PM, <[email protected]> wrote: >> >> look at ActionResumeInterval, it looks like what you are needing. >> >> >> >> David Lang >> >> >> >> On Tue, 23 Nov 2010, Sivan Greenberg wrote: >> >> >> >>> Hi all, >> >>> >> >>> I have two rsyslog nodes set up as follows: >> >>> >> >>> CLIENT: >> >>> >> >>> >> >>> $ModLoad omrelp >> >>> $ModLoad imuxsock >> >>> >> >>> $template myhostid,"%TIMESTAMP:::date-rfc3339% CLIENT157 >> >>> %syslogtag:1:32%%msg:::sp-if-no-1st-sp%%msg%" >> >>> $WorkDirectory /root/rsyslog/spool >> >>> $MainMsgQueueSaveOnShutDown on >> >>> $MainMsgQueueType LinkedList >> >>> $MainMsgQueueFileName mainq >> >>> $MainMsgQueueSize 5 >> >>> $MainMsgQueueCheckpointInterval 1 >> >>> $MainMsgQueueHighWatermark 4 >> >>> $MainMsgQueueLowWatermark 2 >> >>> $MainMsgQueueMaxDiskSpace 1g >> >>> >> >>> $WorkDirectory /root/rsyslog/spool >> >>> $ActionQueueType Disk # making the queue pure disk type >> >>> $ActionQueueSyncQueueFiles on >> >>> $ActionQueueFileName forwq >> >>> $ActionResumeRetryCount -1 # infinite retires on insert failure >> >>> :msg, contains, "lead" :omrelp:10.200.10.181:10514;myhostid >> >>> >> >>> >> >>> SERVER: >> >>> $ModLoad omrelp >> >>> $ModLoad imuxsock >> >>> $ModLoad imrelp >> >>> $ModLoad ommysql >> >>> >> >>> $InputRELPServerRun 10514 >> >>> >> >>> $template myhostid,"%TIMESTAMP:::date-rfc3339% ReplicaFrontEnd109 >> >>> %syslogtag:1:32%%msg:::sp-if-no-1st-sp%%msg%" >> >>> $WorkDirectory /root/rsyslog/spool >> >>> $MainMsgQueueSaveOnShutDown on >> >>> $MainMsgQueueType LinkedList >> >>> $MainMsgQueueFileName mainq >> >>> $MainMsgQueueSize 5 >> >>> $MainMsgQueueCheckpointInterval 1 >> >>> $MainMsgQueueHighWatermark 4 >> >>> $MainMsgQueueLowWatermark 2 >> >>> $MainMsgQueueMaxDiskSpace 1g >> >>> >> >>> $WorkDirectory /root/rsyslog/spool >> >>> $ActionQueueType Disk # making the queue pure disk type >> >>> $ActionQueueSyncQueueFiles on >> >>> $ActionQueueFileName forwq >> >>> $ActionResumeRetryCount -1 # infinite retires on insert failure >> >>> :msg, contains, "lead" :ommysql:127.0.0.1,Syslog,rsyslog,rsyslog >> >>> :msg, contains, "lead" /var/log/leads.log >> >>> >> >>> >> >>> ------------- >> >>> >> >>> Message are delivered fine in normal case where server is always >> up, >> >>> however, rsyslog according to the configuration should support >> >>> resending messages in the disk queue if they have been stuck there >> due >> >>> to the server being down. So when I intentionally stop the server, >> and >> >>> bring it up again ($ reboot) messages waiting in the client's disk >> >>> queue seem to never get delivered to the server. Also I confirmed >> that >> >>> messages are still in the disk queue at the client side and rsyslog >> >>> never seems to try and send them again after it lost connection to >> the >> >>> server; >> >>> 1) Is there a way to set the interval by which rsyslog retries? >> >>> 2) If not how to fix this? maybe upgrading the 5.6.0 can solve >> this? >> >>> >> >>> Many thanks, >> >>> >> >>> -Sivan >> >>> _______________________________________________ >> >>> rsyslog mailing list >> >>> http://lists.adiscon.net/mailman/listinfo/rsyslog >> >>> http://www.rsyslog.com >> >>> >> >> _______________________________________________ >> >> rsyslog mailing list >> >> http://lists.adiscon.net/mailman/listinfo/rsyslog >> >> http://www.rsyslog.com >> >> >> > _______________________________________________ >> > rsyslog mailing list >> > http://lists.adiscon.net/mailman/listinfo/rsyslog >> > http://www.rsyslog.com >> > > _______________________________________________ > rsyslog mailing list > http://lists.adiscon.net/mailman/listinfo/rsyslog > http://www.rsyslog.com > _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com
