Pipes do not seem to work well on Linux. I could not get consistent stability with them, and changed to writing to 'buffer' files that are named by the current time to include seconds. I then operate on the data. I also preformatted the data using templates so that it resembles what I already want for the data. Richard
-----Original Message----- From: syslog-ng-admin@lists.balabit.hu [mailto:syslog-ng-admin@lists.balabit.hu] On Behalf Of Greg Hartung Sent: Wednesday, January 22, 2003 9:47 AM To: syslog-ng@lists.balabit.hu Subject: [syslog-ng]1.5.24 hangs, Red Hat Linux 7.1 2.96-98
I upgraded from 1.4.14 to 1.5.24 because I needed support for named pipes. Now the daemon hangs every few hours or so (with or without the use of pipes). All the output files (about 90 devices) have the same last timestamp. It happens a dozen or more times a day with no obvious pattern in the timing. Does seem to be proportional to volume (doesn't hang overnight). The output actually stops mid-message. Many of the files are 2GB or more. kill -HUP or syslog-ng restart seems to clear the problem up (continues writing to the same large files). syslog-ng conf file is about 10k with about 20 filter/log/destination combo's. Any ideas? If you want to see the config, let me know, I just didn't want to clutter the list...
Thanks, Greg
_______________________________________________ syslog-ng maillist - syslog-ng@lists.balabit.hu https://lists.balabit.hu/mailman/listinfo/sysl> og-ng
Frequently asked questions at http://www.campin.net/syslog-ng/faq.html