Hi, On 31.01.2011, at 22:43, Brice Figureau wrote:
> On 31/01/11 19:11, Udo Waechter wrote: >> [..] >> We use passenger 2.2.11debian-2 and apache 2.2.16-3, ruby1.8 from >> squeeze. > > I see a pattern here. It seems Micah (see a couple of mails above in > this thread) has about the same setup, except he's using mongrels. > > It would be great to try a non-debian ruby (hint: Ruby Enterprise > Edition for instance) to see if that's any better. > Well, since this behavior turned up with 2.6.3 I did not think about blaming it on another tool. Like ruby. I will try RubyEE though. > Do you use storeconfigs? Yes, ALOT! Nowadays with stompserver and puppetqd. I did switch it off already and that did not change a (performance) thing. > >> Puppetmaster works fine after restart, then after about 2-3 hours it >> becomes pretty unresponsive, catalog runs go upt do 120 seconds and >> more (the baseline being something about 10 seconds). > > With 160 hosts, a 30 min sleeptime, and a compilation of 10s, that means > you need 1600 cpu seconds to build catalogs for all your fleet. > With a concurrency of 8 core (assuming you use a pool of 8 passenger > app), that's 200s per core, which way less than the max of 1800s you can > accomodate in a 30 min time-frame. Of course this assumes an evenly > distributed load an perfect concurrency, but still you have plenty of > available resources. So I conclude this is not normal. Nope, like I said. We had pupetmaster running as a VM with 4 Cores and 4 Gigs 'o RAM. This worked fine since 0.22.x, now its twice as big (if this comparison holds) and performance is worse than ever. Also, we do not do 30 Minutes puppet runs. We do it every hour for workstations and every 2 hours for servers all with half of that time random sleep. The load on the server is pretty evenly distributed. Once or twice a day there are some peaks, but those are not critical at all. Thanks, udo.
smime.p7s
Description: S/MIME cryptographic signature
