dan@rna.UUCP (Dan Ts'o) (10/02/90)
When running a large program on a MVAX II under Ultrix 3.0, I get "panic: IO error in push" quite consistantly. What is the problem ? Cheers, Dan Ts'o 212-570-7671 Dept. Neurobiology dan@rna.rockefeller.edu Rockefeller Univ. rna!dan@nyu.edu 1230 York Ave. tso@rockvax.bitnet NY, NY 10021
grr@cbmvax.commodore.com (George Robbins) (10/02/90)
In article <1046@rna.UUCP> dan@rna.UUCP (Dan Ts'o) writes: > > When running a large program on a MVAX II under Ultrix 3.0, I get > "panic: IO error in push" quite consistantly. What is the problem ? The first guess is that you have bad spots lurking in your swap space. Check your console output and/or uerf(8)... -- George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr but no way officially representing: domain: grr@cbmvax.commodore.com Commodore, Engineering Department phone: 215-431-9349 (only by moonlite)
alan@shodha.enet.dec.com ( Alan's Home for Wayward Notes File.) (10/03/90)
In article <1046@rna.UUCP>, dan@rna.UUCP (Dan Ts'o) writes: > > When running a large program on a MVAX II under Ultrix 3.0, I get > "panic: IO error in push" quite consistantly. What is the problem ? There is a very good chance that an disk I/O error occured while writting a page to the page/swap device. Check the error log for disk errors. You might also want to make a pass with radisk on the disks that have the page/swap space on them. > > Dan Ts'o 212-570-7671 -- Alan Rollow alan@nabeth.enet.dec.com
roj@dodads.dco.dec.com (Roger Masse) (10/03/90)
In article <1046@rna.UUCP>, dan@rna.UUCP (Dan Ts'o) writes: |> |> When running a large program on a MVAX II under Ultrix 3.0, I get |> "panic: IO error in push" quite consistantly. What is the problem ? !>> From: grr@cbmvax.commodore.com (George Robbins) !>> The first guess is that you have bad spots lurking in your swap space. !>> Check your console output and/or uerf(8)... George's comments would agree with the kernel messages description of the problem: IO err in push File: /sys/vm/vm_swp.c Routine: swdone Problem: A hard I/O error occurs as a page kluster is being transferred. The routine detected a hard I/O error had occurred as a page kluster was being transferred. This error typically indicates a hardware problem. Regards, Roger Masse' 8301 Professional Place Digital Equipment Corporation Landover, Maryland, 20785 Washington ULTRIX Resource Center Internet: roj@dco.dec.com uucp: decvax!decuac!roj Disclaimer: My opinions do not necessarily represent anything.
jenst@dxcsjr.cern.ch (Jens Trier Rasmussen) (10/04/90)
roj@dodads.dco.dec.com (Roger Masse) writes:
:George's comments would agree with the kernel messages description of the
:problem:
: IO err in push
: File: /sys/vm/vm_swp.c
: Routine: swdone
: Problem: A hard I/O error occurs as a page kluster
: is being transferred.
: The routine detected a hard I/O error had
: occurred as a page kluster was being
: transferred. This error typically indicates
: a hardware problem.
:Regards,
: Roger Masse'
:
: 8301 Professional Place Digital Equipment Corporation
: Landover, Maryland, 20785 Washington ULTRIX Resource Center
: Internet: roj@dco.dec.com uucp: decvax!decuac!roj
: Disclaimer: My opinions do not necessarily represent anything.
I don't know. I have a DECstation2100 running Ultrix4.0 (rev 179) and I've had
3 crashes all giving this type of message on the console. I have checked my
uerf log and there were no hardware errors. I have also used rzdisk to
scan for bad blocks, none of them either. Someone told me that the fact
that I'm running with 2 swap partitions with different size could lead to
the problem.
regards
Jens Trier Rasmussen
CERN-CN/CS
jenst@cernvax.cern.ch