Pptransfer failing and retrying

General
1

Hi,
I am receiving both CPU time limit exceeded
2022-08-20T10:53:44Z CRITICAL - failed/XCPU errors on my PPTRANSFER and also sometimes fails with this message:
md5sum: WARNING: 1 computed checksum did NOT match
[ERROR] Checksum verification failed.
The pptransfer automatically reruns after this and has succeeded but the final cycle has now failed 4 times and so I’m unsure whether to proceed. The XCPU fail is happening on both u-cp350 and u-co989 and the checksum fail is only occurring so far on u-cp350. The time out is happening at 3 hours.

Thanks,
Hannah

2

Hi Hannah

I can not see a reason for the checksum failure nor for the slow transfer - what happens if you trigger the pptransfer task now?

Grenville

3

Hi Grenville,

It has now run fine after releasing! Is there anything I can do to stop this occurring again? The wallclock time is set to 3H50M but the PPtransfers were all failing at 3 hours so I’m wondering if I need to extend this for future runs / how I extend this to run for the final 50 minutes if needed?

Best,
Hannah

4

Hannah

As far as I can see, the time is set to 3H (see archer2.rc [[POSTPROC_RESOURCE]])

I can’t make any sensible suggestions for how to prevent this, since we don’t know why it happened.

Grenville