On 28 April 2013 20:56, Ken Causey ken@kencausey.com wrote:
-------- Original Message -------- Subject: Re: Problems with Build.squeak.org From: Frank Shearar frank.shearar@gmail.com Date: Sun, April 28, 2013 7:39 am To: Ken Causey ken@kencausey.com Cc: Squeak Hosting Support box-admins@lists.squeakfoundation.org
So it was just a case of not enough time. The builds take around 40 minutes to run, which isn't bad considering it's 3k tests run three times.
But more importantly, how does it look from an unkilled process standpoint? I saw some messages in Jenkins talking about leaking file descriptors, so I reckon we probably still have a bit of a problem?
frank
It looks fine. When I started checking today there were a number of processes running and it took a while for it to clear out but once it did there was nothing left hanging about.
ExternalProcesses seems to run a lot, is it triggered by commit to trunk or the like? Builds less than an hour apart seem a bit over the top to me. Is it possible to limit it to no more than once every 6 hours or something like that? I ask not because I think it is that important at this point but because I expect it will become more relevant in the future.
It's set to run after SqueakTrunk, but I just saw that somehow it was also configured to run after SqueakTrunkPerformance. I removed that, which will halve the number of times it runs. No idea how that happened. It's a bit much to be a typo, and I know some builds had their configurations altered during the reboot.
SqueakTrunk's set to run after a change to the build scripts, or when something's committed to Trunk. The latter condition's checked by a Groovy script that polls every 15 minutes.
I'll make sure the other ExternalPackages builds have the correct upstream repositories when I switch them on (right now).
Thanks for keeping an eye on things, Ken!
frank
Ken