errpt - paging space problem

[shoux]

Hi all !

Can anyone help me what is the error means? It seems there is insufficient paging space size but the percentage use is only 1% (lsps –s). This message appears every 5 minutes.

Thank you in advance

Regards
Shoux

#lsps -s

Total paging Space Percent Used
2560mb 1%

# errpt -a
LABEL: CORE_DUMP_FAILED
IDENTIFIER: 45C7A35B

Date/Time: Thu Jan 20 14:26:46 WAUS
Sequence Number: 55748
Machine Id: 003044DB4C00
Node Id: SEAIB_SERVER
Class: S
Type: PERM
Resource Name: SYSPROC

Description
SOFTWARE PROGRAM ABNORMALLY TERMINATED

Probable Causes
INTERNAL SOFTWARE ERROR
SYSTEM RUNNING OUT OF PAGING SPACE

User Causes
USER GENERATED SIGNAL
Failure Causes
CORE DUMP FAILED - SEE A REASON CODE BELOW

Recommended Actions
DEFINE ADDITIONAL PAGING SPACE
RERUN THE APPLICATION PROGRAM
IF PROBLEM PERSISTS THEN DO THE FOLLOWING
CONTACT APPROPRIATE SERVICE REPRESENTATIVE

Detail Data
SIGNAL NUMBER
11
USER'S PROCESS ID:
113340
REASON CODE
1
USER ID
369
PROCESSOR ID
7
CORE FILE NAME
/tmp/TEST.0105/core

 

[alexhu]

I believe the way the core dump program works is when it detects a failure it dumps its file to the paging area. From there it is copied to the designated filesystem. Clearly 2560mb is not enough for the size of core file you have.

What size of main memory do you have?

Alex

 

[MoshiachNow]

Looks to me like your core directory is defined to /tmp - whicj normally will not be enough for any core file.
This does not look related to your paging space settings.

Long live king Moshiach !

http://www.7for70.com/

 

[plamb]

check and see where your primary dump location is.
default primary dump device is
/dev/hd6. The default secondary dump device is /dev/sysdumpnull

sysdumpdev -e Estimates the size of the dump (in bytes) for the current running system. If
the dump will be compressed, then the size shown is the estimate of the size
after compression.


do a man sysdumpdev for more info

 

[baanman]

Hi,
find which user is causing this error by finding user of userid = 369 .
You can do that with;
lsuser ALL |grep id=369

Probably the cause of this error is that the user don't have write premission on /tmp/TEST.0105/core file.


Baanman

 

[shoux]

Hi all !

thanks for the response. here is some info wich can help
to cater the problem. thanks

1. size of main memory is 8GB

2. #sysdumpdev -l
primary /dev/lg_dumplv
secondary /dev/sysdumpnull
copy directory /var/adm/ras
forced copy flag TRUE
always allow dump TRUE
dump compression OFF

3. # sysdumpdev -e
0453-041 Estimated dump size in bytes: 1189085184

4. #df -kt
Filesystem 1024-blocks Used Free %Used Mounted on
/dev/hd3 851968 460548 391420 55% /tmp

5. #ls -l
drwxrwsr-x 29 bin bin 184832 Jan 26 14:46 tmp


shoux

 

[MoshiachNow]

Looks like you better change /tmp access rights to 777.

Long live king Moshiach !

http://www.7for70.com/

 

[plamb]

hmm the est dump size is 1.189g /tmp is .85g and free is .4g
unless my math is bad (need more coffee) I see a problem

 

[RodKnowlton]

This is a core dump, not a system dump.

Find out what uid 369 has in their crontab that's running every five minutes and core dumping (or trying to). The user is probably not in the bin group, so they can't create the core dump. Changing perms to 777 on /tmp should eliminate the errors, but the user should be notified that their process is croaking.



Rod Knowlton
IBM Certified Advanced Technical Expert pSeries and AIX 5L
CompTIA Linux+
CompTIA Security+

 

[guelph]

RodKnowlton is right, but you should add 't' to /tmp using chmod u+t /tmp

 

[RodKnowlton]

But if users can't delete each other's temp files, what will I do for entertainment?

Seriously, though, guelph is right. Without the 't', you're open to all sorts of race condition vulnerabilities.



Rod Knowlton
IBM Certified Advanced Technical Expert pSeries and AIX 5L
CompTIA Linux+
CompTIA Security+

 

[biondo68]

hi all,

Can anyone help me what is the error means? In errpt i find this error, every day

oslevel 5.2 maintanance level 4

thanks

biondo 68
Description
platform_dump processing failure

Detail Data
command name
platform_dump
Line number
960
Message
The file system type may not be any.
Indicator event sequence number
0
Dump file name