[5.4] kernel panic when the referral server is not responding
Environment
- Red Hat Enterprise Linux 5 Update 4
- i686 (and probably x86_64 too).
- kernel version: 2.6.18-164.el5 and 2.6.18-164.9.1.el5
Issue
- [5.4] kernel panic when the referral server is not responding.
- The kernel panic occurs after some seconds when a nfsv4 volume is mounted.
Resolution
- Patch from Upstream: 0001-Fix-a-referral-error-Oops.patch
- RHEL 5.6 Erratum: http://rhn.redhat.com/errata/RHSA-2011-0017.html
Root Cause
- There is a bug in fs/nfs/client.c
Diagnostic Steps
- Following steps can easily reproduce the issue:
A. configure
1. export nfs share directory at server side.
# cat /etc/exports
/nfsroot/nfsv4 127.0.0.1(rw,fsid=0,insecure,no_root_squash,nohide,crossmnt)
/nfsroot/nfsv4/dir1 127.0.0.1(rw,insecure,no_root_squash,nohide,crossmnt,refer=/work@100.100.100.100)B. steps
step1: # mkdir -p /nfsroot/nfsv4/dir1
step2: # service nfs restart
step3: # mount /dev/hda1 /nfsroot/nfsv4/dir1
step4: # mount -t nfs4 -o timeo=10,retrans=1 127.0.0.1:/ /mnt
step5: # ll /mntNote: In step3, a system seems to have a panic when some iso is loop mounted instead of an actual partition like hda1.
This solution is part of Red Hat’s fast-track publication program, providing a huge library of solutions that Red Hat engineers have created while supporting our customers. To give you the knowledge you need the instant it becomes available, these articles may be presented in a raw and unedited form.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
