Description of problem: I am trying to mount /home from my Debian NFS server, but keep getting the error message "mount to NFS server `172.26.101.48` failed: server is down". But the server is obiviousiy up, as I am writing this from a thin client whose server mounts /home from the same spot... The server is up, I am able to ping it, ssh to it, and rpcinfo -p shows theese: portmapper (ver 2, port 111, tcp and udp), rquotad (vers. 1 and 2, tcp and udp, port 830 and 833), status (ver. 1, udp and tcp, port 32768), nfs (version 2, tcp/udp, port 2049, mountd (version 1 and 2, tcp/udp, port 955 and 958) I have also tried to re-export it from the thin-client server I mentioned, but with no luck (also tried insecure): mount: 172.26.100.253:/home failed, reason given by server: Permission denied At home i have a FC2-t3 test-computer which mounts a NFS export on my desktop (noauto, users). That works great, so NFS is not totaly broken. Version-Release number of selected component (if applicable): As suplied with fedora core 2 test 3 How reproducible: Try to mount an nfs export Steps to Reproduce: 1. 2. 3. Actual results: Error message Expected results: Get mounted Additional info: As I said, it works at home. Also seen several mailing list postings by searching the web for "mount to NFS server" "failed: server is down" (google). And nfs does not show up in /proc/filesystems normal? (nothing is mounted so maybe if its a loadable module it is normal?)
Use the '-v' option on the mount command... It will show the RPC error (if any) that is being returned.
This might be a duplicate of 12294. Is the original poster using autofs to mount /home, or is it an fstab mount?
I am using fstab, and then using mount -a to mount the stuff. Pretty shure (at least 99%) that i am using the correct options in fstab - the line looks something like 172.26.101.48:/home /home nfs defaults 0 0 (at least it that what i think it should look like - but I am not on the site rigth now, and won't be until friday. I've asked someone who should be on site tomorow to take a look, but i do no know if it will be done.)
in the original comment rpcinfo -p is only showing nfs (version 2, tcp/udp, port 2049). Is this truly the case or just a typo? If so does mount -o vers=2 work?
no, it was no typo :D Works now, but why didn't it detect it automatically?
Is this still an issue with more recent FC kernels and autofs rpms?
this is an fstab mount. I still have to specify -o vers=2 when mounting manually. But something changed during the upgrade of nfs-utils earlier this week - suddenly stat() became *very* slow. I am going to open another bug at that one.
how are things with the 2.6.9 based errata kernel ?
The vers=2 stuff or the stat (that was LDAP... I don't know what was wrong, it migth have been the server... "Fixed" it by enabeling nscd.. But we are almost finished swithching out that old crap box with a new server - debian sarge - and it don't have any problems with nfs v3...)