On Sun, Mar 25, 2012 at 8:08 PM, Juhani Connolly <[email protected]> wrote: > Still stuck with a severely under-performing cluster where with every > version change the problem seems to change. >
Understood. Going from 0.90 to 0.92, there were nearly 1k changes made. > I'm curious to know if anyone has devised a systematic approach to finding > issues in their configurations. > Do these help: http://hbase.apache.org/book.html#trouble.casestudy and http://gbif.blogspot.com/2012/03/hbase-performance-evaluation-continued.html > What tools do people look to, and what are they most suspicious of > initially? Ganglia or opentsdb w/ a big of log splunking and some thread dumping is usually how a few of us proceed. At extreme, added debug logging. > I'm concerned I may be overlooking something major as after going > through hardware saturation checks and service logs, returning everything to > the simplest configuration possible and trying various other configurations, > I'm now mainly diving around the source code and stack dumps from the > master/region servers/datanodes/namenode and making no progress, and I don't > feel this is the right approach, but can't see any other alternative. > Let us try and give you feedback on the other thread you started. St.Ack
