[p4] Newly-slow checkpoints

Matt Janulewicz perforce-user-forum at forums.perforce.com
Thu Mar 16 13:55:01 PDT 2017


Posted on behalf of forum user 'Matt Janulewicz'.

Twenty two minutes. How quaint!  :)

I can't overstate the importance and convenience of running a good
monitoring system on all your Perforce servers (and other servers.) We run
Zabbix so we have a nice historical record of CPU usage, disk usage, i/o, etc.
When something like this happens, checking Zabbix is usually the first thing I
do.

If you catch your checkpoint in the throes of taking longer than usual, and
you're on a Unix-like system, 'iotop' and 'iostat' are two
things that are useful. Nine times out of ten this will likely be a busy disk
with high iowait due to other processes (as mentioned by Sambwise, backups are a
popular one) taking up resources.

Also, you're running this on an offline copy of your database, right? Not on
a live, running server ...?



--
Please click here to see the post in its original format:
  http://forums.perforce.com/index.php?/topic/5247-newly-slow-checkpoints


More information about the perforce-user mailing list