1834161 Members
2346 Online
110064 Solutions
New Discussion

core dump

 
HERMANS DIRK
Occasional Advisor

core dump

After that we installed hp-ux 10.20 ace 5, we have a core dump when we do a 'swinstall'.
The error message is: "FATAL ERROR: X Failed BadLength (poly request too large or internal Xlib length error) Abort(coredump)"
Is there anywhone that know a solution?
?lkqqdsf8çà&à"'é('&((-
13 REPLIES 13
Cheryl Griffin
Honored Contributor

Re: core dump

Check for a bad patch:
# swlist -l fileset -a state |grep PH

Everything should be configured vs. installed transient or corrupt.
If installed:
# swconfig \*

If transient or corrupt, select the patch for installation again.

There is a problem with the Xlib patch. More on this in a minute.
Cheryl
"Downtime is a Crime."
Cheryl Griffin
Honored Contributor

Re: core dump

What is the date from the ACE media? (Or a part number.) I am attempting to locate a list of what's on the media.
"Downtime is a Crime."
HERMANS DIRK
Occasional Advisor

Re: core dump

the ENU is 01-00-00.
?lkqqdsf8çà&à"'é('&((-
Alex Glennie
Honored Contributor

Re: core dump

QU : what H/W is the swinstall gui displaying on, an hpux system or a PC ?

QU : has this just started failing ? anything else affected ?

QU : "file core" = ?
Paddy_1
Valued Contributor

Re: core dump

please use the following c file to compile and print the output as it gives us more details as to why it crashed.
The sufficiency of my merit is to know that my merit is NOT sufficient
HERMANS DIRK
Occasional Advisor

Re: core dump

*it's on a J5600.
*the swinstall command isn't failling all the time. It's for example possible to run it twice and the third time it failed. The same happend with sam, sometimes it failing.
*attached you will find the core file.
?lkqqdsf8çà&à"'é('&((-
HERMANS DIRK
Occasional Advisor

Re: core dump

*attatched the output from the c file.
?lkqqdsf8çà&à"'é('&((-
Cheryl Griffin
Honored Contributor

Re: core dump

If you have the part number it would make this easier. Without the part number I did a search of X patches which were released on ACE media and came up with the following patches:
PHSS_19792
PHSS_19791
PHSS_19793
PHSS_16648 ('01)
PHSS_11136 ('01)

All release on Dec '99 or '01 ACE.
Check the status of these patches.
Did you "match what target has" when they were installed?
Did you change any options which may have forced a reinstall of a patch?
Were any errors recorded in the swagent.log?

"Downtime is a Crime."
Paddy_1
Valued Contributor

Re: core dump

That was a prelim program to see what it diesd from.
Please run the core against this one too for a more detailed output and we can zero in on what happened.While this will tell what exactly happened then we can propose a solution
The sufficiency of my merit is to know that my merit is NOT sufficient
HERMANS DIRK
Occasional Advisor

Re: core dump

The part number from the CD is B3782-10456. I checked the patchlist, but non of this patches where installed.
?lkqqdsf8çà&à"'é('&((-
HERMANS DIRK
Occasional Advisor

Re: core dump

I tried to compile the C program but i had a lot of error messages.
?lkqqdsf8çà&à"'é('&((-
Alex Glennie
Honored Contributor

Re: core dump

Tried downlaoding core file attachment ... looks like there's been some corruption, so can you run "file core" and post the output.

You say this is intermittent, any other applications running at the time of the crash ? if yes what and what do they do ?

also check /var/dt/Xerrors for clues and the syslog.log just in case upon next crash.

Sounds as if X resources are becomming short of supply and I was expecting you to be running this on a PC Xserver,so I'd also urge you to apply the latest Xserver patch and also consider the latest motif and CDE patches too :

PHSS_28364 & PHSS_27877
HERMANS DIRK
Occasional Advisor

Re: core dump

We noticed that ther was firmware difference between this J5600 and another.
We did a upgrade from firmware release 4.9 to 5.0.
This was no solution. We also swapped the disks to an identional machine but the fault stayed on our J5600.
?lkqqdsf8çà&à"'é('&((-