Red Hat Bugzilla – Bug 162843
nfsv4 kernel error on fedora 4: xdr error 10008
Last modified: 2007-11-30 17:11:09 EST
+++ This bug was initially created as a clone of Bug #142813 +++
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.3)
Description of problem:
when trying to mount a nfs4 filesystem on another remote fedora core 3
(FC3), i got the following error repeated on the client syslog :
Dec 10 15:46:27 filer2 kernel: decode_getfattr: xdr error 10008!
Dec 10 15:46:43 filer2 kernel: decode_getfattr: xdr error 10008!
... while the mount command seems to be "freezed" (but interruptible
by ^C). So any of my attempts to mount with nfs4 failed.
Here is the config on both nodes:
kernel 2.6.9 compiled from tarballs from kernel.org, all nfs option
compiled-in the kernel itself.
Fedora Core 3 with unmodified tools (FC3 seems to natively support nfs4)
both portmaps on client and server listen only to localhost
(vanemery's doc allow access to portmap only to localhost, so i think
it does the same)
Server: nfs export of a XFS partition
$ cat /etc/exports
The mount command on the client was:
mount -t nfs4 -o ro,intr 10.0.0.58:/ /home/
i used the AUTH_SYS method described step by step on the vanemery tutorial
except the "acl" mount option because xfs doesn't accept it. (i think
xfs supports ACLs natively)
anyone has an idea about the "xdr error 10008" error ?
Thanks in advance,
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1.Mount an nfsv4 filesystem
Additional info: I`v got same bug in fc4 with nfs-utils-1.0.7-8
"decode_getfattr: xdr error 10008" means the server is
waiting for some type internal request to complete
(like an uid mapping request to rpc.idmapd). When this happens
make sure idmapd is started and if it is, try restarting
it (i.e. server rpcidmapd restart). The problem should
After system reboot problem never repeated. Sorry.