On Tuesday 16 November 2010 19:21:00 [email protected] wrote: > On Tue, 16 Nov 2010, Marc Schiffbauer wrote: > > Hi all, > > > > Can rsyslog distinguish between "DB-Service not available" and single > > "SQL Statement failed" conditions? > > > > Background is that I want rsyslog to to retry for a infinite numer of > > times if the postgresql is down but I do not want it to DOS the > > PG-server until the end of time if a single SQL statements fails. > > > > Is this possible? >
Hi David, > it is possible in theory at least. When we were developing the new batched > (vector) interface for database plugins this was one of the issues that we > discussed. This is implemented in the om database module because it can > tell the difference between the two types of errors. > > unfortunantly I don't remember the details of making this work, but I > think that if you set rsyslog to retry forever, but it the om module will > not keep retrying a bad statement. > > I would suggest giving it a try. setup a output format that does not > auto-quote the parameters and then try sending 'good' log messages > through to verify things are working, then send a 'bad' log message > through (say something with a quote in it) and see if that jams things up > or if it gets dropped (hopefully with an error being logged by rsyslgo) > and then following log messages go through. > I tried this out. Bad statements will be tried again and again in 4.6.4 :-/ So I guess I will have to deal with that by using as many casts as possible in the SQL statements. Thanks for your help. -Marc _______________________________________________ rsyslog mailing list http://lists.adiscon.net/mailman/listinfo/rsyslog http://www.rsyslog.com
