Hello managers,
a while ago i asked for help with the messages
NFS3 server x00 not responding still trying
NFS3 server x00 ok
appearing up to 50 times per day in the logs of my AlphaServers
(being NFS-client to x00) 'hanging' that machine for (many) seconds
each time.
After some more detailed investigations (a helpful guy from dec-support
gave me a script called 'sys_check' which produces a very helpful hmtl-output
- HINT!!! ;-) we found the solution:
My nfs-exported AdvFS was badly fragmented and thus was steadily running
short of BMT space (with UFS-words: was running short of inodes).
A first run of only 1 hour of the defragment-program (from AdvFS-utilities)
stopped those nasty messages very quickly and now i'm going to move that
AdvFS to a newly created one with better fitting BMT-space.
Or in other words: if you have AdvFS-filesystems with usually more than about
40.000 files in it, you should very quickly learn about the '-x' and '-p'
options for 'mkfdmn' and 'addvol', if you happen to be an ignorant like me ;-)
With regards,
Bernt Christandl
----------------------------------------------------------------------
- Bernt Christandl / Max Planck Institut - Extraterrestrische Physik -
- D-85740 Garching / Phone: +49/89/3299-3342 / Fax: +49/89/3299-3569 -
- email: beb_at_mpe.mpg.de -
----------------------------------------------------------------------
Received on Mon Jan 19 1998 - 15:34:31 NZDT