Core Dump Vs Crash Dump

[Kanagaraj]

Dear All,

Whar are the differents between core dump and crash dump.
core dump---Application use
crash dump--- server panic/hanged condition it will create(by pressing toc button or run command TC in MP>CM> prompt..

But I have no idea,What is the use of these both??
Please help me...

[Johnson Punniyalingam]

>>But I have no idea,What is the use of these both??<<

Core dump - Application use ,, Used to dump some Application core files, If was there any Application realted issues.

crash dump -> This Used for both Application & System hardware & software failure.
(in short system may crash and crash files are dumped for either due Application problem and System panic )

>>crash dump--- server panic/hanged condition it will create(by pressing toc button or run command TC in MP>CM> prompt..
<<

Above mentioned can be done ,

while you testing your crash dump or if your server panic or hang , You can perform the TC, Server will perfrom transfer of control and crash files created,


Hope this helps,

Thanks,
Johnson

[Jeeshan]

actually some times the application core dump is not all time necessary for further investigation.

to investigate system crash information has kept in crash dump files. it will give you full information what has happened while system crash, like black box.


check this link

http://forums13.itrc.hp.com/service/forums/questionanswer.do?threadId=41692

[George_Dodds]

These can be sent off to the hardware/software vendor (HP, oracle, etc) who have software that can analyse these dumps.

[Taifur]

Hi D.Kanagaraj,

For crash dump and core dump, you can follow the below link for better understanding,

crash dump
http://docs.hp.com/en/5992-4580/ch03s07.html


core dump

http://docs.hp.com/en/B3909-90029/ch05s04.html


If it helps, pls assign point.

Rgds//
Taifur

[Laurent Menase]

In fact both don't tell what happened, but tell what was the state of the application - for core dump - or of the system -for crashdump-.

HP support and lab experts can find out the direct cause of the panic, - what was dereferenced- and knowing application or kernel structure, behavior and code imagine scenarios which can lead to the hang/panic/memory leak. - the root cause-.

Sometimes it can be quick to find the root cause starting from the direct cause, but sometimes, it is a puzzling nightmare when the conjunction of multiple events is needed, and when the crash happens a long time ago after the root cause.

It is nevertheless the only way HP support can use to start to analyze a panic or system cause hang.

- So be sure to well configure your dumping area, and well dimension /var/adm/crash
- be sure to keep your crash dump files in case there is a need to analyze them.
- if you need to reboot your system because an application is hang unkillable, and you can't wait for support instruction then take a TOC at least it will be a starting point for kernel investigations.

[Kanagaraj]

Thanks to all for your reply,I am clear about core dump and crash dump.