daily latency spike

Paul B. Henson henson at acm.org
Mon Nov 2 20:42:46 EST 2015


> From: Russ Allbery
> Sent: Sunday, November 01, 2015 8:25 PM
> 
> I assume you've already looked for daily cron jobs?  It's common for LDAP
> servers to take a nightly LDIF dump for backups, for example, which would
> cause symptoms like this.

The latency occurs every 24 hours from when the service was last restarted;
I'm pretty sure I don't have any cron jobs that are scheduled based on the
last start time of the kdc process :). We do indeed do a daily LDAP dump for
backup purposes, but that has a fixed time to run. Given the way it occurs
every 24 hours after the service starts like clockwork, it feels like
something within the kdc itself? I was hoping somebody would jump up
immediately and say "oh yeah, it's such and such", barring that, I guess
we're going to need to start tracing the process activity during the latency
window and see if anything jumps out.

Thanks.



More information about the Kerberos mailing list