Happy to report I recently installed an RT test platform with very few hitches
- my sincere thanks to the authors and contributors to what appears to be a
VERY powerful addition to our toolset.
My group is testing the system - refining queue lists, keyword heirarchy, the
usual up-front workload I’m sure…
Have a few questions on tweaking RT, and on some [to me] confusing areas in
the docs. Apologize if any of this is covered in the archives already!
Our test platform is RT 2.0.14 on a SuSE 7.3 PII 366 laptop with 128Meg RAM,
14G HD, mysql backend.
Not a screamer by any means, but it’s been up to the task thus far.
Our production box [not installed yet] is a Solaris8 Netra t1 440Mhz, 512MB
RAM & two 9G drives.
I’m shy about doing any sort of raid/mirroring configuration here [in future
we’ll almost certainly need to add an external disk array] - so I plan on
putting the database partition and the ~/rt2/WebRT stuff on separate drives to
glean a bit of performance improvement. Any recommendations on partition
allocation, sizing, naming based upon default destination directories when
compiling/installing apache/perl/mysql/RT from source?
Most of the user community seem to be running RT on linux/Intel. Any good/bad
experiences with Solaris/Sparc? Any Solaris version-specific issues I should
be aware of? [we’ve got every release from 2.5.1 through 2.8 available]
We could do RT on linux/intel - but our onhand spare PC’s are older PIII-500
machines with small drives [6G], and my mandate is to deploy the box without
buying any new hardware. We’ve got a couple of netra T1’s sitting around, so
why not?
Comments / Suggestions are very welcome.