Simon wrote: > Hi There, > > We are seeing a large spike in load average on our web server (from > 0.xx to 10-15.xx) but there does not seem to be anything hogging the > CPUs or anything... Am i missing something here? What can i do to > check other issues. > > Thanks > > Simon > > > > top - 15:33:25 up 5 days, 5:28, 1 user, load average: 10.74, 10.23, 10.92 > Tasks: 128 total, 1 running, 127 sleeping, 0 stopped, 0 zombie > Cpu(s): 26.2% us, 29.5% sy, 0.0% ni, 40.0% id, 4.0% wa, 0.0% hi, 0.3% si > Mem: 1036100k total, 943536k used, 92564k free, 45664k buffers > Swap: 1951888k total, 89692k used, 1862196k free, 488912k cached
That looks very similar to a problem I had with an early 2.6 kernel on a Dell 1RU server -- there was some sort of odd hardware problem between the motherboard and the kernel, and a later kernel update cleared the problem. This was a number of years ago, and iostat showed a lot of I/O going on when it wasn't really expected on that machine, but fixing it really was more of a "might as well try the latest kernel" type of thing, nothing direct was ever found. Nate -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
