Bad drive + flaky raid = unhappy server
Nevermind.. have it sort of figured out..
(One bad drive + one semi-bad-drive-flaking-out-in-a-weird-way for those who really want to know what I was babbling about in here.. )
« Are null objects equal? | Main | Obscure code of the week »
Nevermind.. have it sort of figured out..
(One bad drive + one semi-bad-drive-flaking-out-in-a-weird-way for those who really want to know what I was babbling about in here.. )
Comments
When you're saying reboot, is it a "shutdown -r", or a forced reboot "/sbin/reboot -nf"?
A shutdown -r will hang if there are processes blocked in I/O operations (check with ps -o pid,command,state | grep B$). I've seen this happen on a borked ReiserFS.
I've never seen a server ignore a reboot -nf.
Posted by: Sérgio Carvalho | June 7, 2005 11:33 AM