I run a 32bit ASE15.5 server on SUN Solaris machine (Solaris 10, single data disk). Recently, I got the error message below whenever clients run a big db job (like a query to a big table (>5GB) with data sorting, or a job use tempdb a lot):
"kernel Spid 148 has exceeded its timeout value while issuing a write for virtual device 8 ('/dev/sybase/temp/tempdb.dat'). A system call has not responded for 60 seconds, and Adaptive Server will be impacted until it returns. Contact your operating system vendor for assistance."
Then the server become not accessible, and I have to kill the server an d re-start. I contact Sybase TS Team, I was told this is OS problem. But I am not convinced because the server work just fine after re-start. And this problem is reproducable.
My tempdb is 8GB with 400MB log. The queried tables which caused the problem range from 2BG to 12GB.
Do anyone have any idea about what causes this problem and how can I fix it? Thanks a lot.