NCP exports for NSS volumes are lost on OES11SP1 after implementing Patch SLESSP2-Kernel 7277 (kernel-default-3.0.58-0.6.2.1.x86_64.rpm)

  • 7011764
  • 10-Feb-2013
  • 12-Mar-2013

Environment

SUSE Linux Enterprise Server 11 Service Pack 2 (SLES11SP2)
Linux Kernel (x86-64) 7277 (kernel 3.0.58-0.6.2.1) for SLES11SP2
Novell Open Enterprise Server 11 Linux Support Pack 1 (OES11SP1)
January 2013 Scheduled Maintenance for OES11SP1 Patch

Situation

After installing a OES11SP1 server and / or updating the server with all available updates, including the January 2013 Scheduled Maintenance for OES11SP1 and the latest Kernel patch for SLES11SP2 (slessp2-kernel version 7277) the NCP exports for all NSS volumes, including the admin volume(s) are lost.
There are also a couple instances where the NCP exports for POSIX volumes are lost or no longer working as expected.

From that moment on the error: '-672' are shown in the ncpcon and ncpserv log files.
Also the error: "IPCThread received request from uid -1" is shown.

Resolution

This issue is resolved with SUSE Linux Enterprise 11 SP2 Patch slessp2-kernel 7374 (kernel 3.0.58-0.6.6.1 x86_64) or later, which has been made available trough the regular update channels.

This specific patch is also available for download via the Novell Patch Finder.

Cause

The culprit would appear to be slessp2-kernel 7277 (kernel 3.0.58.0.6.2.1 x86_64) released for SLES11SP2.



Additional Information

If the server was reconfigured to prevent the implementation of kernel patches or later kernel versions using zypper locks, these need to be removed.

This is established by executing as root:
zypper rl -t patch slessp2-kernel
zypper rl kernel-default* (or the flavor of your kernel, determined with uname -r ).
Verify that the locks were removed, the following command should not return any locks regarding the kernel:
zypper ll