Tim Donnelly schrieb: > I tried Philippe's soulution and got no joy. > > To test the theory that it was a name resolution problem, I inserted a third > mount into the fstab file. I placed this before /proddb and /prodobj. The > first filesystem failed to mount, but the second and third worked as one > would expect. I also changed the fstab entry for the first filesystem to > point to the IP address instead of the hostname. No luck. > > In looking at the nfs start up script in /etc/init.d I noticed a comment > about it sometimes being helpful to mount NFS devices in the backgroup with > an & and a sleep time. I modified the startup script as mentioned, but still > the first mount is "ignored" while subsequent mounts work fine. > > Google is not helping, does anyone else have any ideas? > > Thanks
I'm running a SUSE 10.2 server and had running SLES 10 server before. When I've exported a file system like this: /share *(rw,root_squash,async,no_subtree_check) there was no problem to mount it from any client, but when exported it specific to one IP, IP range or subnet e.g. like this: /share 192.168.1.101(rw,root_squash,async,no_subtree_check) I couldn't mount it from 192.168.1.101. After that I've found a message in /var/log/warn of the server about something like: 'Attempt to mount from 192.168.1.101. Could not resolve the name via reverse-dns. Attempt dismissed' So I setup first the /etc/hosts with all my clients in, which helped! Now I'm running a local dns for all my clients on the server. This might help you or it might not. However check the logs on the SLES9 server. Thx Jan -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
