• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 914
  • Last Modified:

How to configure a dump utility on AIX 5.3/6.1?

I am new to working with AIX. We are trying to find a Linux Net Dump like utility for AIX. Also, any guides or information on how dumps work within AIX would be greatly appreciated!
1 Solution
Hi again,

AIX doesn't have a net dump facility.

Here kernel dumps always go to disk. You can define dedicated dump devices (primary/secondary)
or you can have the dump destination default to swap (paging) space.

If the dump goes to swap, the system tries to copy it to a defined destination at restart time.

I always recommend creating dump devices, so the dump doesn't get lost in case the copy destination was too small.

Creating dump devices consists in setting up a logical volume (or two) in rootvg and then making this/these known to AIX as dump destination(s).

1. Making an LV

First, estimate the dump size:

sysdumpdev -e

Divide the value displayed by (1024*1024), then divide the result by the partition size of rootvg to get the required number of partitions (rounded up).
Determine the partition size with

lsvg rootvg

It's displayed at the right under "PP size".

Example: sysdumpdev -e gives  "Estimated dump size in bytes: 314363084". This value divided by (1024*1024) gives 299,
this divided by a PP size of e.g. 32 gives 10 (rounded up).

Now create the LV:

mklv -t sysdump -y dumplv1 rootvg 10 hdisk#
with 10 being the number of partitions to allocate (see above). Specifying a desired hdisk# (e.g. hdisk0) is optional. If you have two or more disks in rootvg, it makes sense to create a secondary dump device on the other hdisk, e.g. with hdisk0 and hdisk1:

mklv -t sysdump -y dumplv1 rootvg 10 hdisk0mklv -t sysdump -y dumplv2 rootvg 10 hdisk1

2) Configure sysdump

sysdumpdev -p -P dumplv1
sysdumpdev -s -P dumplv2

means primary, -s means secondary and -P means "make the configuration permanent across reboots".

Now you have two dump devices.

If a kernel panic occurs, a dump will be invoked automatically (first to primary, and to secandary if primary is occupied or unavailable).
The following is dumped during the dump process.  
  • list of currently running processes and related information about the process
  • currently mounted filesystems, inode table and open file table
  • currently configured ttys and their status
  • memory buffers for data  
  • system buffers  
  • system variables and statistics
  • kernel's own record of process it is currently running.
You can start a dump manually by issuing

sysdumpstart -porsysdumpstart -s

with -p and -s meaning primary or secondary, respectively.

Attention: sysdumpstart will first "freeze" the system, then take the dump, then it will reboot!

If there is enough space to copy the dump to the /var/adm/ras directory, then it will be copied directly during reboot.
By default it is copied to /var/adm/ras/vmcore.x.

This destination is configurable:

sysdumpdev -d /var/adm/ras

Once the dump is copied to a file, you can examint it using the kdb tool.

I think explaining kdb would go far beyond the scope of your question, or am I wrong?

Anyway, "man kdb" gives some explanation,
and here is the AIX Troubleshooting Redbook: http://www.redbooks.ibm.com/redbooks/pdfs/sg245496.pdf

If you need more information about dumps and troubleshooting, please let me know!


Sultaana43Author Commented:
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Free Tool: Subnet Calculator

The subnet calculator helps you design networks by taking an IP address and network mask and returning information such as network, broadcast address, and host range.

One of a set of tools we're offering as a way of saying thank you for being a part of the community.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now