Using Credhub behind Concourse, database growing?

What version of the credhub server you are using?

1.6.5 via https://github.com/starkandwayne/bucc/releases/tag/v0.4.1

What version of the credhub cli you are using?

1.6.0

If you were attempting to accomplish a task, what was it you were attempting to do?

Backup my BOSH + Credhub

bucc bbr backup

What did you expect to happen?

• I expected it to happen relatively quickly. I’m only use the credhub for < 100 items across my BOSH + Concourse.

What was the actual behavior?

It took 4 mins to backup credhub database:

[bbr] 2018/03/16 22:38:21 INFO - Backing up credhub on bosh/0...
[bbr] 2018/03/16 22:42:56 INFO - Finished backing up credhub on bosh/0.

The credhub backup is over 2G

-rw-rw-r-- 1 jumpbox jumpbox 2.3G Mar 16 22:46 bosh-0-credhub.tar

Is there an audit log or similar inside my database that’s growing regularly because Concourse is polling it? Suggestions for operators to cleanup their DB to constrain its size?

[mdelillo]

@drnic @poblin-orange @gberche-orange It seems like you all figured this out by now. We had an issue where audit logs were quickly filling up our database.

In version 1.9+, the audit logs now go to a file instead. In versions 1.6 and 1.7, we added a property (audit_logs.days_retained) that will only keep around logs from the last X days.

[gberche-orange]

Seems related to story 154256575: feature request: automate cleanup of database tables used in audit operations

It’s fairly easy to fill these up as we’ve seen in the past. It would be a nicer operator experience if it didn’t need to be done manually and in response to failure that over-full databases cause.

cited in Slack: https://pivotal.slack.com/archives/C133WF954/p1515686364000363

which mentions:

Jan 31, 5:12 pm @cojackson respond to inform them that audit tables are going away entirely