high CPU usage %sys
2007-12-25 10:19:00
> Hello Managers
>
> CPU usage : %sys about 50% for no apparent reason.
> Any idea, I don't have all the latest patches on my machine.
> when i try to power cycle the machine using init 6 the machine does not
> go down, and after 5 minutes or so i do a "reboot"
> after reboot still the %sys value is around 50.
>
Summary :
The problem was due to NFS.
stopped nfs it was all fine
then later on
ran snoop
to find out there was stale NFS file handle
so had to reboot the client machine
information about stale NFS file handle from sunsolve database
This means that a file or directory that your client has open has
been removed or replaced on the server. It happens most often when a
dramatic change is made to the file system on the server, for example
if it was moved to a new disk, or totally erased and restored. The
client should be rebooted to clear Stale NFS file handles.
Also, if the nfs server goes down and the client tries to write, you will get
this
stale nfs file handle.
Thanks to the following Sun Managers
Scott Clark <Scott.Clark@East.Sun.COM>
Casper Disk < Casper@holland.Sun.com
Reichert, Alan < aareichert@tasc.com>
Franczyk, Gary < gfrancyzk@carbomedics.com>
Edurado E. Silva < esilva@netcom.com>
Ram Kumar < ramk1@excite.com>
Kevin Sheehan <kevin@joltin.com>
Kris Briscoe <unixboy@aol.com>
Version 3: (14515923 calls)
I would check for network traffic. Is this machine perhaps an NFS server?
That and the high interrupt level are the clues I am going on here. NFS is
done in the kernel, and network traffic tends to be high in system CPU time
too.
> Any idea, I don't have all the latest patches on my machine.
netstat -i 5, and top should be very instructive.
Kevin
>
># nfsstat -s|more
>
>Server rpc:
>Connection oriented:
>calls badcalls nullrecv badlen xdrcall dupchecks dupreqs
>14515644 0 0 0 0 2 0
>Connectionless:
>calls badcalls nullrecv badlen xdrcall dupchecks dupreqs
>280 0 0 0 0 0 0
>
>Server nfs:
>calls badcalls
>14515924 0
>Version 2: (0 calls)
>null getattr setattr root lookup readlink read
>0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
>wrcache write create remove rename link symlink
>0 0% 0 0% 0 0% 0 0% 0 0% 0 0% 0 0%
>mkdir rmdir readdir statfs
>0 0% 0 0% 0 0% 0 0%
>Version 3: (14515923 calls)
Lots of NFS calls? Have you tried snoop?
Casper
Comments
Got something to say?
You must be logged in to post a comment.
