Fedora Account System
Red Hat Associate
Red Hat Customer
Description of problem: rsyslog started consuming 100% of my CPU. Version-Release number of selected component (if applicable): rsyslog-7.4.2-2.fc20.x86_64 How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: I do not see any log file increasing in size. When the problem started I was getting messages like these: Jan 1 09:25:39 cascavel systemd-journald[9212]: Failed to write entry (22 items, 636 bytes), ignoring: Invalid argument Jan 1 09:25:49 cascavel systemd-journald[9212]: Failed to write entry (22 items, 636 bytes), ignoring: Invalid argument Jan 1 09:25:59 cascavel systemd: httpd.service: Got notification message from PID 938, but reception only permitted for PID 0 Jan 1 09:26:29 cascavel systemd: httpd.service: Got notification message from PID 938, but reception only permitted for PID 0 Jan 1 09:26:39 cascavel systemd-journald[9212]: Failed to write entry (22 items, 636 bytes), ignoring: Invalid argument Jan 1 09:26:59 cascavel systemd: httpd.service: Got notification message from PID 938, but reception only permitted for PID 0
I tried just to reboot several times, and rsyslogd was always back eating all of my CPU... Then, I deleted all files in /var/log/journal/f4355f24cef24df19056ceeed66fa774/ and reboot again. This was the only way, because the system became completely unresponsive. Until now, everything is fine. No clue about what caused this issue, but if this is a recurring problem, it will be impossible keep using F20...
Thanks for the info. There's already another bug for this issue, so add any new findings there. Do I understand correctly that purging /var/log/journal/* stopped the cpu utilization for the time being? Please let me know if it happens again. If the problem reoccurs (too frequently), there's a workaround described in the other bug. *** This bug has been marked as a duplicate of bug 1047039 ***