Cron return codes

[Manuel G]

Hi all:

Our system returned a rc = 134 error from cron last night.

We were searching itrc site for some rc error document or something related but we didn??t find anything. We??ve got from HP-technicians a sort erors list (attached) but ir doesn??t covers rc=134.

Does anyone know if there is such document or
Does anyone know what rc=134 error means?

Thanks.

[Thierry Poels_1]

Hi,

do you mean cron exited with this error?

or that a program (from your crontab) exited with the eroror code? In this case the return code reflects to the exit code of the specified script/program. It can be user defined; = not standard.

regards,
Thierry.

[Roger Baptiste]

<Does anyone know what rc=134 error means?>>

Since there doesn't seem to be a corresponding error code for 134, you can look at the script itself and see where it returns in the error. Some pointers:
Was this script running correctly before?
Are you able to execute it manually?
is there any in-built error coding scheme?

HTH
raj

[Frank Slootweg]

As far as I know, this means that the job exited with signal 134-128, i.e. signal 6, i.e. SIGABRT.

I have found no good reference to this 'RETURN VALUE /return code above 128' issue. It is implicitly documented in wait(2). As far I know In The Old Days (tm) it was explicitly documented in that manual page.

[My reference: S1100002718]

[Manuel G]

Hi again:

I can clarify some of your questions; Thierry we detected error code rc=134 on user job mail. All error codes from our cron jobs are written in user??s mail file (I guess this is the default).

cron runs a user defined PRO*C process that:
1) it runned correctly ever (cron runs it every 30 mt)
2) runs manually.
3) doesn??t have a built in code handle.


I think Frank is correct. We have discover some swap problems. Sometimes swapspace ocupation reaches 100%; when we try to run a process at this moment it aborts indicating not enough swapspace and generates a core.
rc=134 could have some relation to swap problems, we??ll continue investigating on this way.

I??ll post the news.

Thanks for your help.
Manu.

[harry d brown jr]

If you have some space disk space, you add filesystem swap (with overhead costs).

live free or die
harry

[Manuel G]

We first try to figure out how used memory is growing. It could be a lack of memory?
We want to install glance or ??load? for sniffing.
Any other suggestion?
Thanks for your help.

[Bernie Vande Griend]

If you install Glance try to get Glance Plus so you get the Measureware pieces too. scopeux (part of Measureware) can then record a history of memory, cpu, disk preformance so you can go back and see what was happening at certain times. If you don't want to use GlancePlus/Measureware, then you can use sar to record a history.

But the first thing to check is that you have twice as much swap space as physical memory.
The command swapinfo -tm will show you your swap devices and how much is being used currently.

[Manuel G]

Ok Bernie,
You??re right, we have 256MB RAM and 512 MB swap space.

I will open a new discuss with our swap problems.

Thanks.