Message boards : ATLAS application : Disk limit increased
Message board moderation

To post messages, you must log in.

AuthorMessage
David Cameron
Project administrator
Project developer
Project scientist

Send message
Joined: 13 May 14
Posts: 309
Credit: 9,881,693
RAC: 8,967
Message 39795 - Posted: 2 Sep 2019, 8:50:30 UTC

I have seen quite a few recent WU fail with DISK_LIMIT_EXCEEDED. It seems like some of the newer heavy tasks use more disk, so I have increased the limit from 8GB to 10GB.
ID: 39795 · Report as offensive     Reply Quote
Toby Broom
Volunteer moderator

Send message
Joined: 27 Sep 08
Posts: 581
Credit: 368,206,041
RAC: 164,391
Message 39932 - Posted: 15 Sep 2019, 18:32:24 UTC - in response to Message 39795.  

For my computers about 70% of task fail, ones that do not run past the report deadline so I aborted them are failing with the disk limit exceeded.

When I check the active task that I have running currently they still have the 8GB limit.
ID: 39932 · Report as offensive     Reply Quote
Aurum
Avatar

Send message
Joined: 12 Jun 18
Posts: 80
Credit: 35,637,488
RAC: 9,506
Message 39937 - Posted: 16 Sep 2019, 16:24:04 UTC

My computers are throwing off a lot of errors. I limited ATLAS WUs to 4c and the computers with 32 GB RAM and a 16 GB swap file. I don't seem to be running out of memory. There must be something else going on.
ID: 39937 · Report as offensive     Reply Quote
Toby Broom
Volunteer moderator

Send message
Joined: 27 Sep 08
Posts: 581
Credit: 368,206,041
RAC: 164,391
Message 39943 - Posted: 17 Sep 2019, 5:57:03 UTC

I had a couple work, maybe davids changes are being applied now,
ID: 39943 · Report as offensive     Reply Quote
Crystal Pellet
Volunteer moderator
Volunteer tester

Send message
Joined: 14 Jan 10
Posts: 893
Credit: 6,239,710
RAC: 2,231
Message 41384 - Posted: 27 Jan 2020, 17:36:14 UTC - in response to Message 39795.  

I have seen quite a few recent WU fail with DISK_LIMIT_EXCEEDED. It seems like some of the newer heavy tasks use more disk, so I have increased the limit from 8GB to 10GB.
27-Jan-2020 17:07:24 [LHC@home] Aborting task XsANDmcfGGwnsSi4apGgGQJmABFKDmABFKDm9a0XDmABFKDmelYfAn_0: exceeded disk limit: 9597.00MB > 9536.74MB
27-Jan-2020 17:07:56 [LHC@home] Computation for task XsANDmcfGGwnsSi4apGgGQJmABFKDmABFKDm9a0XDmABFKDmelYfAn_0 finished
27-Jan-2020 17:07:56 [LHC@home] Output file XsANDmcfGGwnsSi4apGgGQJmABFKDmABFKDm9a0XDmABFKDmelYfAn_0_r1608456720_ATLAS_result for task XsANDmcfGGwnsSi4apGgGQJmABFKDmABFKDm9a0XDmABFKDmelYfAn_0 absent

https://lhcathome.cern.ch/lhcathome/result.php?resultid=260591944

10000000000 is sometimes not enough. at least not if you want to use VBox snapshot mechanism.

1 snapshot fits, but before a previous snapshot can be deleted a new snapshot has to be created first, so there is a short period >10000000000 bytes in the slot directory.
ID: 41384 · Report as offensive     Reply Quote
Toby Broom
Volunteer moderator

Send message
Joined: 27 Sep 08
Posts: 581
Credit: 368,206,041
RAC: 164,391
Message 41386 - Posted: 27 Jan 2020, 18:23:18 UTC - in response to Message 41384.  

The snapshots are what cause problems for me, if I don't reboot or quit boinc then they run fine.
ID: 41386 · Report as offensive     Reply Quote

Message boards : ATLAS application : Disk limit increased


©2020 CERN