Taz appears to be pretty badly overloaded. Access to the CVS repository
tends to be slow, both for updates and commits. It's getting to be a bit
of a problem, so we need to decide on the best way to fix it.
We could look at
o Upgrading Taz. RAID disks, more memory, faster CPUs, whatever.
o Partitioning the load. Taz is handling a lot of stuff; perhaps
moving all of the Apache XML stuff (CVS, web server, mailing lists)
to their own machine would make sense. Or it could be split
based on function rather than project, with CVS, mail and web
serving on separate machines, but with each handling all of the
projects.
o Some way of running local CVSs that mirror the main one on Taz.
Any other ideas? Or thoughts on how best to make it happen?
-- Andy Heninger
