Skip to content

Releases: AliceO2Group/O2DPG

prod-202109-21: fix unit of default memory limit

Choose a tag to compare

@sawenzel sawenzel released this 21 Sep 18:17
This was in bytes which lead to overscription
on some systems. Now corrected to be in MB.

prod-202109-20

Choose a tag to compare

@sawenzel sawenzel released this 20 Sep 07:52
Integrate FV0 workflow

prod-202107-22

Choose a tag to compare

@sawenzel sawenzel released this 22 Jul 14:12
Properly quote configKeyValues; Be clear that full paths required for…

prod-202107-05: Tmp disabling confkey check

Choose a tag to compare

@sawenzel sawenzel released this 05 Jul 11:53
We need a more detailed treatment; The condition
to have to have confkeys does not always apply (boxgen for instance).

prod-202106-15

Choose a tag to compare

@sawenzel sawenzel released this 15 Jun 11:11
simworkflow: option for early cleanup / disable OpenMP for TPC clusters

prod-202106-04

Choose a tag to compare

@sawenzel sawenzel released this 04 Jun 07:24
adjusting few resource fields

prod-202106-02: Update o2dpg_sim_workflow.py

Choose a tag to compare

@sawenzel sawenzel released this 02 Jun 11:36
dcabe31
add missing list initialization

prod-202105-09

Choose a tag to compare

@sawenzel sawenzel released this 09 May 17:43
AOD merging as part of workflow

prod-202104-30: Adjustin IRate; improvements in pipeline runner

Choose a tag to compare

@sawenzel sawenzel released this 30 Apr 20:38
Several improvements in pipeline runner:

* fix a bug with "skipping task" ... now much faster
* provide option to make core-dump checkpoint on failure
  and upload to ALIEN
* offer possibility to re-schedule a failed task when
  it matches a certain condition
  (for instance random failures that happen very rarely)

prod-202104-27: Fix stall problem in pipeliner runner in environments with nice>0

Choose a tag to compare

@sawenzel sawenzel released this 27 Apr 06:41
This is fixing a stall problem in the pipeline runner, which was
observed on systems (GRID), where the runner was launched with an
increased nice value (>0).

In this scenario, allocated resources per task were not correctly
"un-accounted' and so the runner stalled at some moment since it couldn't
find free resources.

This commit is also adding some debugging mechanism with webhooks.
Later integration into the standard logging system would be nice.