On 04/28/2014 06:48 PM, Kaleb S. KEITHLEY wrote: > On 04/25/2014 06:23 PM, Joe Julian wrote: >>> >> Just to keep all this discussion in one place: >> >> #gluster [10:01] <kkeithley> JoeJulian: I'll try running cppcheck to see >> how long it takes. I'm setting up a bunch of machines to do things like >> routinely run things like coverity, cppcheck, valgrind, etc. If cppcheck >> takes too long perhaps we just want a daily run instead of a run every >> time someone commits > > Seems to take about 15 minutes. At least in the VM I set up on a > fairly beefy server. > > Off hand though I'm not sure we want to add another 15 minutes of > tests to every commit. Justin has indicated that there's a hook in > Jenkins to run cppcheck. I propose a weekly (or daily) run of > master/head ? does that seem reasonable? > IMO Cppcheck should be part of regression tests. So that issues will be caught when patches are sent. It would be like "prevention is better than cure" :). Cppcheck can be executed multi threaded so a beefy server will definitely help. > Related to that, we need Jenkins slaves. Are we satisfied with using > Rackspace VMs for this? Are we making any headway with adding slaves > for things like gluster regression tests? Those being perhaps the > biggest bottleneck. > > I have a beefy server with a big disk array that we could send to the > Phoenix lab and put in the DMZ, accessible from the outside, but I > don't know if we have any space there for more gear. Is that something > people think is worth pursuing, or should we just stick to Rackspace VMs? >
