The ticket https://fedorahosted.org/389/ticket/48266 creates new subentries (per replica) in order to do keep alive updates for replication.
There is a keep alive entry per replica, so when the replicaID is removed from the topology the entry become useless.
It could be part of CLEANALLRUV, to delete that subentry (related to the replica).
Per triage, push the target milestone to 1.3.6.
attachment 0001-Ticket-48278-cleanAllRUV-should-remove-keep-alive-en.patch
Looks good.
Note: the following comment is beyond this ticket's purpose... :) So, please push your proposed patch as is.
Some time in the future, we may want to categorize all the log messages and add some indication if the message is for success or benign or severe? {{{ 2049 cleanruv_log(task, rid, CLEANALLRUV_ID, "No Keep-Alive entry to remove (%s)", 2050 keep_alive_dn); <== benign or even a success
2053 cleanruv_log(task, rid, CLEANALLRUV_ID, "Failed to delete Keep-Alive entry (%s) " 2054 "Error (%d) This entry will need to be manually removed", 2055 keep_alive_dn, rc); <== severe (the word "Error" indicates it.)
2058 cleanruv_log(task, rid, CLEANALLRUV_ID, "Removed Keep-Alive entry (%s)", keep_alive_dn); <== success. }}}
Replying to [comment:5 nhosoi]:
Looks good. Note: the following comment is beyond this ticket's purpose... :) So, please push your proposed patch as is. Some time in the future, we may want to categorize all the log messages and add some indication if the message is for success or benign or severe?
Some time in the future, we may want to categorize all the log messages and add some indication if the message is for success or benign or severe?
I'm assuming you mean error logging in general, and not just cleanAllRUV.
So, I like this idea, and we could even add a log level for these types. For example, only log "severe" messages regardless if it's replication logging, or plugin, or trace functions, etc.
I think there was also a proposal to do something like this, and get our error levels inline with syslog's levels.
We can just add a secondary level:
{{{ slapi_log_error(SLAPI_LOG_REPL, LOG_FATAL, repl_subsystem, "Error\n"); }}}
Thoughts?
b0fc822..a2d97e0 master -> master commit a2d97e0 Author: Mark Reynolds mreynolds@redhat.com Date: Fri Sep 2 11:46:30 2016 -0400
Replying to [comment:6 mreynolds]:
I'm assuming you mean error logging in general, and not just cleanAllRUV. Yep, that's what I meant. So, I like this idea, and we could even add a log level for these types. For example, only log "severe" messages regardless if it's replication logging, or plugin, or trace functions, etc. I think there was also a proposal to do something like this, and get our error levels inline with syslog's levels. We can just add a secondary level: {{{ slapi_log_error(SLAPI_LOG_REPL, LOG_FATAL, repl_subsystem, "Error\n"); }}} Thoughts?
I'm assuming you mean error logging in general, and not just cleanAllRUV. Yep, that's what I meant.
Sounds nice! It'd be easy to add/modify. We could have some candidate keywords, like "Error" and "Benign" and so on...
Metadata Update from @nhosoi: - Issue assigned to mreynolds - Issue set to the milestone: 1.3.6.0
389-ds-base is moving from Pagure to Github. This means that new issues and pull requests will be accepted only in 389-ds-base's github repository.
This issue has been cloned to Github and is available here: - https://github.com/389ds/389-ds-base/issues/1609
If you want to receive further updates on the issue, please navigate to the github issue and click on subscribe button.
subscribe
Thank you for understanding. We apologize for all inconvenience.
Metadata Update from @spichugi: - Issue close_status updated to: wontfix (was: Fixed)
Login to comment on this ticket.