|Summary:||[Logs] Karaf logs are being rotated at 16M instead of 10G and maxrollover is set 7 leading to loss of log data|
|Product:||Red Hat OpenStack||Reporter:||Sai Sindhur Malleni <smalleni>|
|Status:||CLOSED NOTABUG||QA Contact:||Itzik Brown <itbrown>|
|Version:||13.0 (Queens)||CC:||aadam, jchhatba, jjoyce, jschluet, mkolesni, slinaber, tvignaud, vorburger|
|Target Milestone:||z1||Keywords:||Triaged, ZStream|
|Target Release:||13.0 (Queens)|
|Fixed In Version:||Doc Type:||If docs needed, set a value|
|Doc Text:||Story Points:||---|
|Last Closed:||2018-06-14 15:06:17 UTC||Type:||Bug|
|oVirt Team:||---||RHEL 7.3 requirements from Atomic Host:|
|Cloudforms Team:||---||Target Upstream Version:|
Description Sai Sindhur Malleni 2018-04-20 15:24:24 UTC
Description of problem: Karaf logs for ODL are being rotated every 16M. Earlier they were being rotated every 10G and looking puppet-opendaylight, it looks like the default is still 10G. Not sure if tripleo is missing it up to cause Version-Release number of selected component (if applicable): OSP 13 opendaylight-8.0.0-5.el7ost.noarch puppet-opendaylight-8.1.0-0.20180321182556.45c4db7.el7ost.noarch How reproducible: 100 % Steps to Reproduce: 1. Install ODL + OSP13 with defaults configuration 2. 3. Actual results: Log files are being rotated every 17M leading to 10s of karaf log files Expected results: Log files were previously being rotated every 10G Additional info: In etc/org.ops4j.pax.logging.cfg log4j2.appender.rolling.policies.size.size = 16MB
Comment 1 Sai Sindhur Malleni 2018-04-20 15:29:15 UTC
Looks like puppet-opendaylight tries to configure 10G by default? https://github.com/opendaylight/integration-packaging-puppet-opendaylight/blob/a6b1e01447b6e50b7b9b9250f106434b880c6e57/manifests/config.pp#L246
Comment 2 Michael Vorburger 2018-04-24 15:59:03 UTC
> External Bug ID: OpenDaylight Bug ODLPARENT-154 This issue does NOT have to await ODLPARENT-154 - it's (presumably) easy to do this locally in puppet-opendaylight - but it's really an indication that the default upstream could be better.
Comment 3 Sai Sindhur Malleni 2018-05-07 14:54:24 UTC
Hey Michael, Was wondering about the status of this downstream? Also looks like we are capping the rotated logs at 7 files. This lead to me losing a ot of valuable logs when running longevity tests over the weekend. Can we increase the max rollovers atleast if we cant increase the log file size?
Comment 4 Janki 2018-05-09 10:18:48 UTC
Puppet configures rollover at 2 with each file size of 10GB. Puppet also has a logic that if log_to_console: "logs to console + file" else: configure file with above said size and rollover and TripleO takes puppet to log to console. So the whole "else" section is not executed.
Comment 5 Sai Sindhur Malleni 2018-05-14 14:39:15 UTC
So who is setting the rollover to 7 in this case?
Comment 7 Janki 2018-05-16 12:22:33 UTC
I will edit the above explained if-else logic in puppet-odl to also configure file size and rollover even if log_mechanism mentioned in TripleO is "console".
Comment 8 Janki 2018-06-14 15:06:17 UTC
Since we are moving to logging to file in ODL, the logs files will be accordingly configured. This work is tracked at https://bugzilla.redhat.com/show_bug.cgi?id=1559055 No more changes needed for this bug. Hence closing it.