ello, I'm in the test-phase of my loggingserver. and syslog-ng started to create a logfile I don't understand why it did that. This is my conf, it's a fairly easy setup. (striped other unimportant stuff) options { long_hostnames(yes); sync(0); time_reap(10); use_time_recvd(yes); use_dns(yes); use_fqdn(yes); create_dirs(yes); log_fifo_size(10000); chain_hostnames(yes); }; source oldserver { udp(ip(xxx.xxx.xxx.xxx) port(514)); }; destination blackhole { file("/var/log/frontier/$YEAR$MONTH/$HOST"); }; destination blackhole_auth { file("/var/log/frontier/$YEAR$MONTH/$HOST-auth"); }; destination blackhole_crit { file("/var/log/frontier/$YEAR$MONTH/$HOST-crit"); }; destination blackhole_emerg { file("/var/log/frontier/$YEAR$MONTH/$HOST-emerg"); }; destination blackhole_err { file("/var/log/frontier/$YEAR$MONTH/$HOST-err"); }; destination blackhole_messages { file("/var/log/frontier/$YEAR$MONTH/$HOST-messages"); }; destination blackhole_warn { file("/var/log/frontier/$YEAR$MONTH/$HOST-warn"); } log { source(oldserver); destination(blackhole); }; log { source(oldserver); filter(f_auth); destination(blackhole_auth); }; log { source(oldserver); filter(f_crit); destination(blackhole_crit); }; log { source(oldserver); filter(f_emerg); destination(blackhole_emerg); }; log { source(oldserver); filter(f_err); destination(blackhole_err); }; log { source(oldserver); filter(f_messages); destination(blackhole_messages); }; log { source(oldserver); filter(f_warn); destination(blackhole_warn); }; The syslog-ng ran for the last 10 days without problems, however today I discovered three new logfiles. -rw------- 1 root root 119 Feb 1 10:43 last -rw------- 1 root root 60 Feb 1 05:45 last-messages -rw------- 1 root root 60 Feb 1 05:45 last-warn erwin:/var/log/frontier/200102# cat last Feb 1 05:45:23 last/ns2.xxx.net message repeated 2 times Feb 1 10:43:27 last/ns2.xxx.net message repeated 1 time erwin:/var/log/frontier/200102# cat last-messages Feb 1 05:45:23 last/ns2.xxx.net message repeated 2 times erwin:/var/log/frontier/200102# cat last-warn Feb 1 05:45:23 last/ns2.xxx.net message repeated 2 times The box the logs came from is a sparc10 with solaris5.8. The orginal logfile statements were: Feb 1 05:45:23 sparky last message repeated 2 times ... Feb 1 10:47:50 sparky last message repeated 1 time There are plenty of other "last message repeated..." lines in sparky's logfile, so I don't see a reason *why* those both got logged seperately. The first logentry repeated a warn-message, the other one a notice-message. Can anyone give me a hint please? The version I use is 1.4.10 on linux. (syslog-ng doesn't run yet on the sparc10. I export the logs via the goold old "*.crit,*.warn,etc,etc. @hostname way") -- Regards, Wiktor Wodecki, Unix Administration | Wapme-Systems AG Tel.: +49-211-748450 | Muensterstrasse 248 Fax: +49-211-74845176 | 40470 Duesseldorf E-Mail: wodecki@wapme-systems.de | http://www.wapme-systems.de
Hello folks, I traced this a bit further, and it seems as if syslog-ng treats "last message repeated xtime(s)" messages from the sun-syslogd as if they came from a seperate machine. As I wrote/pasted in my previous mail messages are forwarded via udp to an other machine and there sorted. Depending on the facility of the previous error message it's saved to various files (last-warn, last-crit, etc.). I'm going to run the syslog-ng on the sun (sunos5.8) during the next couple of days, I hope to get the problem fixed by this way. -- Regards, Wiktor Wodecki, Unix Administration | Wapme-Systems AG Tel.: +49-211-748450 | Muensterstrasse 248 Fax: +49-211-74845176 | 40470 Duesseldorf E-Mail: wodecki@wapme-systems.de | http://www.wapme-systems.de GnuPG-Fingerprint: 1024/E22253B9 084C79504D650E9246D148AFF3F73201E22253B9
participants (1)
-
Wiktor Wodecki