From: Alan Sparks (asparks@quris.com)
Date: Thu Jan 23 2003 - 19:51:11 EST
OK, I'm desperate. Been working this intermittent problem on Solaris
7 for months. Now it follows me into my new installs of Solaris 8.
I have recently patched (recommended/security patch bundles) Solaris 8
and Solaris 7 boxes (most of them Netra T1s). While NFS mounts onto
those boxes are generally working OK, they intermittently do not show
the new attributes (or contents) of files updated on the NFS
filesystem. Ever. At least until something happens, like an attempt
to delete a file through the mount. This seems to force a cache flush
or something, since suddenly the new attributes show up.
Mounts are TCP based, version 3. All machines sync'ed to a common NTP
server. No NFS timeouts or anything like that. I've seen the problem
on filesystems mounted from a Sun server, and from a NetApp filer
(suggesting client problem).
Setting noac or actimeo=0 on the mounts aren't curing the problem.
Neither is setting nfs_disable_rddir_cache=1 on the solaris 7
/etc/system. The Solaris 8 boxes have a pretty generic /etc/system.
Has anyone else had problems like this? Is there any subtle NFS
patches not included in the recommended patch bundle I should know
about?
Any debug suggestions quite welcome. I'll summarize whatever I get.
Thanks in advance.
-Alan
-- Alan Sparks, Sr. UNIX Administrator asparks@quris.com Quris, Inc. (720) 836-2058 _______________________________________________ sunmanagers mailing list sunmanagers@sunmanagers.org http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:25:42 EDT