At 09:28 PM 21/04/00 +0200, bert hubert wrote:
>----- Forwarded message from Greg Baker <[EMAIL PROTECTED]> -----
>
>Date: Fri, 21 Apr 2000 10:51:00 -0500 (CDT)
>From: Greg Baker <[EMAIL PROTECTED]>
>Subject: Re: VM: killing...
>To: unlisted-recipients: ;(no To-header on input)
>Cc: [EMAIL PROTECTED]
>
>
>I'm getting this problem on a stock RedHat 6.2 (2.2.14-5.0) system,
>running RAID 1. It's usually preceded by these entries in syslog:
>
>Apr 21 10:26:01 case kernel: swap_duplicate: entry 246f3020,
>nonexistent swap file
>Apr 21 10:26:01 case kernel: VM: killing process calibre
>Apr 21 10:26:01 case kernel: swap_free: offset exceeds max
>Apr 21 10:26:01 case kernel: swap_free: Trying to free nonexistent
>swap-page
>Apr 21 10:26:01 case kernel: swap_free: offset exceeds max
>Apr 21 10:26:01 case kernel: swap_free: Trying to free nonexistent
>swap-page
>
>or...
>
>Apr 21 00:55:42 case kernel: swap_duplicate: entry 16dff014,
>nonexistent swap file
>Apr 21 00:55:42 case kernel: VM: killing process calibre
>
>This is with 768MB RAM on an Athlon 800Mhz system. I'm unable to
>benchmark performance of Linux vs. HPUX/Solaris due to the fact it can
>never complete the jobs we would use it for!
Out of interest, you may wish to try the latest RH kernel for 6.2 which is
2.2.14-6.1.1 I don't know if this may fix the bugs, but it's certainly not
going to hurt. You should be able to get the files from any mirror of RH
under /pub/redhat/updates/6.2/RPMS/i386/ in their tree (assuming that
redhat is mirrored in /pub/redhat that is).
RedHat apparently use the later RAID code patched back into the 2.2.x
series kernels, and this may cause all sorts of problems if you download
your own 2.2.x source and compile in RAID without patching it up to the
latest code.
--
-=[ Stuart Young (Aka Cefiar) ]=-------------------------------
| http://amarok.glasswings.com.au/ | [EMAIL PROTECTED] |
---------------------------------------------------------------
