Everyone,
Let me start out by telling you I know that locking and NFS is a generally
bad idea, for whatever reason my architecture (e.g. it's not important nor
changeable) does not give me the ability to alter it.
So now to the problem. I'm having just a challenging time with Oracle 8i
on Linux trying to save database (dbf) files on an NFS mounted filesystem.
I was wondering if anyone had any input into the situation with this
version of Linux, Oracle, NFS v2 (using UDP) on NetApp (latest rev) or
ECCS network attached disk storage.
I've heard everything from "don't use oracle", to "use solaris" on through
"don't store the databases on the NFS" as well as "a pending kernel update
in 2.3 will help make this problem go away (heard this from one of the
vendors of the NAS)".
What I'd like to know is if anyone has encountered any situation similar
and any and all advice you can provide to me personally (I will send a
digest back to the list based on what I receive, help keep the Internet
beautiful, do your part to cut down on spam eh?).
I'm sure there is not one thing that by itself would make this problem go
away, but if anyone has had experience with mount options, nfs version
usage (v2 vs v3), protocol usage (tcp vs udp), portmap/rpc timeout values
and or nfsd timeout (not sure right at this moment where timeouts are
set).
Thanks in advance for your time.
Sincerely,
Andrew Daniels
---
Manager, Operations
-====---====---====---====---====---====---====---====---====---====---====-
to unsubscribe email "unsubscribe linux-admin" to [EMAIL PROTECTED]
See the linux-admin FAQ: http://www.kalug.lug.net/linux-admin-FAQ/
