Here is another unrelated example I was talking about. https://hudson.apache.org/hudson/job/Ant_BuildFromPOMs/buildTimeTrend
So, more than doubled in the last 3 days compared with other build times. Note as I mentioned before, this is not tied to just one machine, but both Ubuntu slaves and the Solaris slave exhibit this behaviour, the same I stated for the archives builds sebb mentions. So, not tied to a particular machine means not disk related. (Surely?) Gav... > -----Original Message----- > From: Sim IJskes - QCG [mailto:s...@qcg.nl] > Sent: Tuesday, 23 November 2010 10:39 PM > To: builds@apache.org > Subject: Re: [hudson] archive phase problems > > On 11/23/2010 01:25 PM, Sim IJskes - QCG wrote: > > Looks like a totally saturated disk subsystem (to me, from a > distance). > > Is there sar data from that machine? > > When you have a jobs that prints a timestamp just before the archive > stage, you can correlate this with the sar data. You just have to look > at the runq size or the waiting/blocked percentage to see if the I/O > was > the culprit. Is there idle=0% or very low, then its CPU bound. > > However, you have to find out how the work is divided between slave and > master, maybe all the data is first sent to the master, and archived > there. > > Gr Sim
