rsyslogd 7.2.7, compiled with: FEATURE_REGEXP: Yes FEATURE_LARGEFILE: No GSSAPI Kerberos 5 support: No FEATURE_DEBUG (debug build, slow code): No 32bit Atomic operations supported: Yes 64bit Atomic operations supported: Yes Runtime Instrumentation (slow code): No uuid support: Yes And postgresql support. Issue is that when run via cli database inserts work flawlessly. When run from the init script updates do not make it into the database and silently fail.
1) Please post your `emerge --info' output in a comment. 2) Does rsyslogd actually start properly?
Created attachment 362136 [details] emerge --info for rsyslog Here's the requested emerge log.
Created attachment 362140 [details] rsyslogd -n -d 2>&1 >rsyslogd.log Here's the output from rsyslogd without running it through the /etc/init.d script.
Hi, Maybe you should compare the command line ran by the init script and use the same with the -d and -n options to debug this issue. I have never used rsyslog with pgsql so I'm a bit clueless. Also, did you try with latest rsyslog ?
Hi, there was recently a bug report for the ompgsql module in the rsyslog mailing list: http://thread.gmane.org/gmane.comp.sysutils.rsyslog/12959 Sounds like that could be the same bug you experienced because when using "-n" rsyslog won't fork so this error won't be triggered (the same thing you noticed). This was fixed in v7.4.10 - not yet in tree. See bug 501978.
Thx @Thomas D. Please reopen if you still have the problem with newer rsyslog >=7.4.10