Messages in this thread | | | Date | Thu, 25 Mar 1999 21:47:15 +0200 | From | Meelis Roos <> | Subject | Re: 2.2.1 SMP: NFS can't get a request slot |
| |
[digging out an old message from l-k archive]
>> After seeing these messages, nfs on this SMP 2.2.1 machine no longer >> works. The problem only seems to crop up if the machine accesses its own >> nfs server (ie: export a directory and then access the directory via amd's >> automounter on the same machine).
TM> The question then arises as to why the RPC layer gets congested when TM> accessing your own nfs server. I have no problems with this on my own TM> machine, so I would first and foremost suspect amd. Can you repeat the TM> problem using just the stock NFS client + server?
I'm having the same problem with pristine 2.2.4. Also with amd (latest am-utils package from redhat, other places may have newer ones).
I tried to add anouther amd mount point but got the configuration file wrong at the first try. So amd had 1 working and 1 nonworking mount point (mount point type parameter was wrong in my conf. file). And when I tell amd to stop (via /etc/rc.d/init.d/amd stop), the process just hangs. It's sitting there right now: 2 processes in rpc_execute, one amd and one shell script to unload amd. And syslog says nfs: task 13820 can't get a request slot
So probably amd messed something up and the nfs client can't talk to amd. The message seems to be a only consequence of the amd problem and the real problem is probably the hanging amd.
-- Meelis Roos (mroos@tartu.cyber.ee)
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |