RT newb here. My team is currently testing RT to see if (and how well)
it would serve us. So far so good, with one exception: every now and
then, the web interface is completely unresponsive for a couple of
minutes. When I get into the system’s console, I see Apache taking
100% of processors during the spike. When the usage falls, everything
goes back to normal: behavior is a lot faster and it keeps going on
like that for several hours until the next spike.
Note that I don’t think the database is being hit during these periods
as I’m watching the MySQL server (SHOW PROCESSLIST) and I can see the
connection pool there, but completely idle. Besides, right now we have
one queue with two tickets, so not much data.
The actual setup is Ubuntu Server 10.04.1 and RT from packages, which
is 3.8.7. The web server is Apache and MySQL the RDBMS (on another
server.) The hardware is a VMware ESXi virtual machine with 1 GB of
RAM and two ~ 2 GHz processors assigned to it. I initially thought of
swapping or caching issues but a quick check on the VMware host shows
no resource overcommitment (and other VMs performing as usual.)
What the best way to track what’s going on?