TrigMC software release coordination: PTD's logbook - page for December 2018

Notes

Use the shift date in reverse chronological order to enter any new Merge Requests.

For each shift date show the NEW MRs, and any related relevant comments/questions; note down when I (or some one else) merged/closed each request, and mark the MR as done/closed by ticking the box ( [x]). If an MR takes longer than a day to merge, leave it listed under the original shift date, but add comments prefaced with the date of the comment/action. Only MRs that are still active should have the box unticked ( [ ]).

Therefore each MR will only appear once in the log, under the date of the day when I first became aware of it.

18/12/2018

The entire ATLAS Athena code base was made public (Open Source) earlier today...! For details, see the talk at today's ATLAS Weekly meeting; in particular slides 6, and 16-17. The "old" gitlab repository (with the code base up until today) has been copied to atlas/athenaprivate1; the gitlab repository for continuous (now in public!) development of ATLAS software, remains at atlas/athena. The MRs for the Open Source era will start (have started, already!) at number 20000, to differentiate from the pre-public MRs.

4/12/2018

Progress with 21.0-TrigMC merge into 21.3 (for info only)

10.33am For info only (no action required of me): the MR for the merging of 21.0-TrigMC into 21.3 is !16441 ("Merge 21.0-TrigMC into 21.3 and enable HLT online packages") and was opened yesterday by Rafal. The MR page has lots of details (eg on how merge conflicts were resolved). The pipeline failed on "required tests" so, at the moment, the merge hasn't been done yet.

5/12/2018 10.10am This has not been merged yet. No new discussion on the MR page as to why.

4.10pm There has been more discussion about this MR posted on the page. Conflicts still present. Rafal's advice is to wait for the pipeline (now running) to conclude, before deciding to merge.

9.04pm The CI pipeline has failed the "required tests".

7/12/2018 1.34pm This has now been merged.

[x] !16447 ATLINFR-2731: fix for arguments with blanks (opened by Oleg Zenin)

10.49am MR was opened yesterday night. Pipeline failed "make" (other steps OK). Not ready to merge.

12.27pm Oleg has checked the error that caused the pipeline failure during "make": it turns out it is the "usual" error due to a symbolic link duplication. In principle not-related to the MR. Of course, there is technically the possibility that the failure is masking some other subsequent error... So I will try to re-trigger a build/pipeline. Note that the L1 shifter still hasn't changed the "review-pending-L1" label.

5/12/2018 10.05am CI pipeline now completed successfully. Merged OK.

3/12/2018

Phasing out of Run-2 Trigger branches

11.44am According to the plans outlined in Rafal's talk last week (here; see also the November 2018 page of this logbook), now that 21.0-TrigMC is up-to-date with 21.0 and 21.1, it is time to merge 21.0-TrigMC into 21.3 (the branch that will be used throughout the Long Shutdown, to develop Run-3 releases). I have searched for the corresponding MRs for 21.3... but does not seem to have been opened yet.

11.46am I have also searched for any recent MRs (less than a day old) that are labelled "alsoTargeting:21.0-TrigMC": found 11 (!), all approved & merged about 40-45 mins ago (see screenshot at the bottom of this logbook page). Therefore can expect the corresponding automatic MRs (into 21.0-TrigMC) will be opened tonight. Maybe Rafal is waiting for these to go through, before merging 21.0-TrigMC into 21.3?

12.16pm I have posted a comment/question to Rafal on the above... (see !14043) ... and then promptly realised that the 11 MRs I mentioned above were all applied to TrigMC quite a while ago (they have MR nos. in the range 9000 - 15000, whereas we are now on 16000+). For some reason the corresponding GitLab pages were "updated" today, and therefore they turned up on my regular search of alsoTargeting:21.0-TrigMC MRs... So I guess we are still on track to open the MR to merge 21.0-TrigMC into 21.3 at some point later today...

[x] !16412 time prints the full cmd line before real..user..sys (for branch 21.0-TrigMC) (opened by Oleg Zenin)

00.31am This MR was opened yesterday. CI pipeline failed 14 h ago due to externals, after having run for 3h30. New pipeline job was automatically started 1 hour ago by the ATLAS Robot; still running...

11.24am CI pipeline has now concluded successfully. Review-approved label added a couple of hours ago. Merged OK.

[x] Check nightlies

11.31am Quick check of the GIT and ATN nightlies: all looks OK; all tests completed, and same success rate as in previous nightlies (64%). The spurious compilation-time error that was seen in the nightly build last week has disappeared.

-- PedroTeixeiraDias - 28 Nov 2018

• List of MRs alsoTargeting:21.0 TrigMC (screenshot 2018-12-03):