Instabilities in 11634.911 (DD4Hep) workflow comparisons

[makortel]

We've observed differences in the DD4Hep workflow 11634.911 comparisons in tests of a few PRs that should not affect results of the DD4Hep workflow. This issue is to collect pointers to those comparisons.

[cmsbuild]

A new Issue was created by @makortel Matti Kortelainen.

@Dr15Jones, @perrotta, @dpiparo, @makortel, @smuzaffar, @qliphy can you please review it and eventually sign/assign? Thanks.

cms-bot commands are listed here

[makortel]

assign geometry

[cmsbuild]

New categories assigned: geometry

@Dr15Jones,@cvuosalo,@civanch,@ianna,@mdhildreth,@makortel you have been requested to review this Pull request/Issue and eventually sign? Thanks

[makortel]

Observed in #35068 (comment) and #34995 (comment)

[makortel]

Here is another occurrence https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-cf3e63/18431/summary.html

[civanch]

@cvuosalo , is the problem back or it is another one?

[cvuosalo]

The instability appears to be random and rare. It is strange that wf 11634.912 does not show it. The difference between the two workflows is that 11634.911 runs the algorithms and calculates the reco geometry, while 11634.912 reads the already calculated algorithm results and reco geometry out of the DB.

[cvuosalo]

I ran workflow 11634.911 thirty times in CMSSW_12_1_X_2021-09-20-1100 with identical results each time. It appears the instability has gone away.