Am 08.04.2011 15:53, schrieb Ralf Hildebrandt: > * Timo Sirainen <t...@iki.fi>: >> On Thu, 2011-04-07 at 16:33 +0200, Kádár Tamás (KTamas) wrote: >> >>> However once in a while, someone goes rogue, we >>> install a new version of thunderbird, someone accidentally sets up an >>> email account in thunderbird mistakingly leaving the download >>> everything option on. This causes high IO on our server, and a single >>> user can quickly more or less kill the whole server in mere minutes, >>> load average quickly spiking to 30-40-50 and everything becomes >>> sloooooow (obviously). >> >> So a single process that is reading files fast enough from disk can >> cause disk IO to spike in a way that makes all other processes wait for >> available disk IO? I don't think that's a common problem. Typically only >> that one process would be waiting for disk IO and the load would >> increase by 1 (or maybe a little bit more, but definitely not by 50).. >> Maybe you should look into Linux I/O schedulers and see if you're using >> the wrong one. > > Sounds like a filesystem problem or a scheduler issue, yes.
perhaps try echo "deadline" > /sys/block/sda/queue/scheduler -- Best Regards MfG Robert Schetterer Germany/Munich/Bavaria
