Fedora Account System
Red Hat Associate
Red Hat Customer
Description of problem: I am using nfsv4 towards a Centos-5 server. The server is mounted using kerberos. The relevant part of the fstab looks as follows: nest.pheasant:/home1 /home/home1 nfs4 sec=krb5 0 0 At some point in time (mainly after some heavy nfs activity) the kernel keeps repeating "NFS: v4 server returned a bad sequence-id error" It does not recover. Reboot seems to be the only recovery possible. The Fedora 9 kernel occasionally also gives this type of errors, but that kernel does recover from the error, while the fedora 10 kernel does not. Version-Release number of selected component (if applicable): kernel-2.6.27.7-134.fc10.x86_64 nfs-utils-1.1.4-4.fc10.x86_64 How reproducible: Some heavy nfs activity. I noticed this error especially while installing a virtual machine in VirtualBox, but I have also seen the errors while using evolution Steps to Reproduce: 1. heavy nfs activity 2. 3. Actual results: errors keep repeating in /var/log/messages Expected results: system recovers (or even better, no such errors at all) Additional info:
I forgot to show the error message on the server side. Here are the errors that the Centos 5 server reports (from a recent Fedora 9 client, as I am back to Fedora 9 for the time being as Fedora 10 is not usable): NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 2, got 3) NFSD: preprocess_seqid_op: bad seqid (expected 2, got 3) NFSD: preprocess_seqid_op: bad seqid (expected 2, got 3) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306) NFSD: preprocess_seqid_op: bad seqid (expected 8305, got 8306)
With the latest kernel : Linux travel.pheasant 2.6.27.9-159.fc10.x86_64 #1 SMP Tue Dec 16 14:47:52 EST 2008 x86_64 x86_64 x86_64 GNU/Linux the hangings disappeared. this may be related to the kernel update on my Centos server machine taht now runs: Linux nest.pheasant 2.6.18-92.1.22.el5 #1 SMP Tue Dec 16 11:57:43 EST 2008 x86_64 x86_64 x86_64 GNU/Linux
So we can close this bug?
The bug hit again today. A reboot solved the issue. I still do not know exactly how to trigger it. Here is an exempt from the Fedora 10 messages file: Dec 29 10:27:13 travel ntpd[3052]: synchronized to 81.171.44.131, stratum 2 Dec 29 10:27:11 travel ntpd[3052]: time reset -1.518701 s Dec 29 10:27:11 travel ntpd[3052]: kernel time sync status change 0001 Dec 29 10:30:51 travel ntpd[3052]: synchronized to 81.171.44.131, stratum 2 Dec 29 10:58:01 travel yum: Updated: gstreamer-plugins-ugly-0.10.10-1.fc10.x86_64 Dec 29 12:01:04 travel kernel: NFS: v4 server returned a bad sequence-id error on an unconfirmed sequence ffff880053409a28! Dec 29 12:01:04 travel kernel: NFS: v4 server returned a bad sequence-id error on an unconfirmed sequence ffff880053409628! Dec 29 12:01:04 travel kernel: NFS: v4 server returned a bad sequence-id error on an unconfirmed sequence ffff880053409a28! Dec 29 12:01:04 travel kernel: NFS: v4 server returned a bad sequence-id error on an unconfirmed sequence ffff880053409628! Dec 29 12:01:04 travel kernel: NFS: v4 server returned a bad sequence-id error on an unconfirmed sequence ffff880053409a28! and then it goes on and on..... and this is what the Centos logfile has: Dec 29 11:48:27 nest nmbd[4557]: Cannot get workgroup name. Dec 29 11:56:25 nest kernel: NFSD: preprocess_seqid_op: bad seqid (expected 27286, got 27287) Dec 29 11:56:25 nest kernel: NFSD: preprocess_seqid_op: bad seqid (expected 2, got 3) Dec 29 12:01:02 nest xinetd[4160]: START: hotwayd pid=8583 from=10.0.0.1 it repeats the Dec 29 12:01:03 nest kernel: NFSD: preprocess_seqid_op: bad seqid (expected 27286, got 27287) many times Centos is now running the Xen kernel: Linux nest.pheasant 2.6.18-92.1.22.el5xen #1 SMP Tue Dec 16 12:26:32 EST 2008 x86_64 x86_64 x86_64 GNU/Linux. Apart from this nothing changed
I have for a long time been running a standard 2.6.28 kernel on my Centos box and did not have the problem any more. I now started the stock Centos kernel 2.6.18-128.2.1.el5xen (x86_64) to play around with Xen and the problem re-appeared. It appears to be a Centos/RHEL5 problem, not Fedora. Should I file a new BZ towards RHEL5?
yes... Close this one and please open a RHEL5 bz...
Today I installed a Fedora 11 client in my network and noticed this bug. My server is also a NFSv4 with Kerberos: Server CentOS: kernel-2.6.18-128.2.1.el5 Client Fedora: kernel-2.6.29.6-217.2.3.fc11.x86_64
Closing this BZ. Have not seen the bug after installing dzickus 2.6.18-162.el5xen experimental RHEL kernel. Will report it on RHEL kernel only if the problem re-appears
Created a BZ for RHEL5: https://bugzilla.redhat.com/show_bug.cgi?id=517629