UNIX for Dummies Questions & Answers

Hi reader,

I'm making a tool out of korn shell script that is running on a HP-UX server. But everytime i invoke the tool, it gets killed after a while (mid-process). I have tried re-running it a couple of times but each invocation ending up the same way ..

following is a snippet of the o/p i get when the scripts runs :

-------------------------------------------------------------------
. . .

2) FTP-ing all files to remote exceed :
> FTP initiated to : 47.103.130.35
The files will be copied to NODATA_smtxv19bk_17_May_2011_13:18:45 on remote exceed
This may take a while...please wait
..........................................................................................
.............................................
/home/casa/clakkad/bin/one_shot_build.sh[1015]: 7177 Killed

-----------------------------------------------------------------------
The line in bold is the sys generated error , rest is all from the script.

I need to find out if there is any way i can find out why the process is being killed ? any log that is generated when system kills a process or something!!

Will realy appreciate any help/lead that i can get

thanks already

Chinmayee

Could you please post the script

sure the ftp server on one side is timing you out most likely. check your ftp config on both sending and receiveing sides

no, i had already checked if FTP was being timed-out.. but not so. Just as soon as the FTP completed the process used to die.

I was transfering a big file to the other end.. could this be a problem ? As in the process being too resource intensive, the server chose to close it.

I'm actually only looking for a way to find the reason behind the closure so that i can somehow prevent its cause from the script. Is there any log that is generated when a process gets killed ?

It's the script which is being killed.
I'd look for a "kill" command within the script itself or failing that some sort of process clean-up cron which is killing the script after a certain period of time.


(There is no timeout as such with HP-UX "ftp". It can be implemented by backgrounding the "ftp" and then killing the "ftp" process if it takes too long.
That is not what has happened here.)

If I remember correctly, as it's been a couple of decades, when a process is killed it is returned via the common register. If you don't have access to this return code, as most who aren't coding in assembly are, then you will most likely have to hunt down the global variable that is assigned the failure code.

It seemed to me that we had that problem and one of the script wizards just made some kind of global variable before it was spawned and when it was killed that error was found there. If my memory serves this is documented in the whatever mechanism that you spawn it by. If you follow. I know it returns or places that return error in a bucket, but I can't remember if we passed it or just had to make a known global variable.

Best of luck, please tell us and I will hunt for it when I get a chance.

global variable! great... i will try and track it down. but i would appreciate help as well, cause i dont know scripting that well to know which variable to look for .

btw, if it helps , this FTP is being implemented through an EXPECT script which is being invoked from an hierarchy of ...lets say.. parent scripts. The orde is something like :

script 1 > script 2 > script 3 > ftp script.

here the script being killed is 'script 1'. perhaps so many sub-shell invocations is causing the problem and as suggested previously by methyl a process clean-up cron mighth be killing the main script. But i ran the same tool (scripts) on another HP-UX server and it didnt get killed and worked to completion.

This is a resource shortage or something issue!! then again i have find a way like the global var thing to try and prevent this.

PS :-
the above may not exactly be a good coding style what with so many sub-shell invocations but thats how i have worked it out for now.

thank you all