14

Noticing that for a VM box, getting in the logs that rsyslogd process gets HUPed. Finding no ideas except a few post in some forums saying this is for logrorate. Any ideas how to fix/troubleshoot this..

messages-20141011:2014-10-10T04:02:02.054134-06:00 udr-oradl01 rsyslogd: [origin software="rsyslogd" swVersion="5.8.12" x-pid="364" x-info="http://www.rsyslog.com"] rsyslogd was HUPed

messages-20141011:2014-10-11T04:02:02.079917-06:00 udr-oradl01 rsyslogd: [origin software="rsyslogd" swVersion="5.8.12" x-pid="739" x-info="http://www.rsyslog.com"] rsyslogd was HUPed

messages-20150124:2015-01-24T04:02:01.497596-07:00 udr-oradl01 rsyslogd: [origin software="rsyslogd" swVersion="5.8.12" x-pid="819" x-info="http://www.rsyslog.com"] rsyslogd was HUPed
19

There's nothing to fix. This is normal operation. When the logfile is rotated by logrotate, it tells rsyslog to close and re-open its logfile file handles by sending it a HUP signal.

If rsyslog doesn't do that, then it will continue to write its log output to the OLD logfiles, not to the new ones because the file handles will still connect to the old files.

This is true even if the old files are deleted (because, on unix, a file isn't truly deleted until the last file handle holding it open is closed)....which is why files don't always release the space they are using up back to the filesystem immediately if some process has an open file handle to them.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.