Patch to fix workflow errors from HGCalTypes changes#50039
Patch to fix workflow errors from HGCalTypes changes#50039kpedro88 wants to merge 3 commits intocms-sw:masterfrom
Conversation
|
cms-bot internal usage |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-50039/47881
|
|
A new Pull Request was created by @kpedro88 for master. It involves the following packages:
@Dr15Jones, @bsunanda, @civanch, @cmsbuild, @kpedro88, @makortel, @mdhildreth can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
|
please test |
|
+1 Size: This PR adds an extra 24KB to repository Comparison SummarySummary:
|
|
@kpedro88 The order was changed to conform with the specification in XML file. Earlier order and the specification did not conform with the ones in the XML files. Please let me try to check these and provide a more stable solution |
|
hold |
|
Pull request has been put on hold by @kpedro88 |
|
so the list of names is in the wrong order? This was an inconsistency that I noticed but was not the actual cause of workflow failure. Let me revert that commit and see if it fixes the comparison |
This reverts commit 995949e.
|
please test |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-50039/47886
|
|
Pull request #50039 was updated. @Dr15Jones, @bsunanda, @civanch, @kpedro88, @makortel, @mdhildreth can you please check and sign again. |
|
+1 Size: This PR adds an extra 24KB to repository Comparison SummarySummary:
|
|
please test workflows 34551.85,34634.114,34634.21,34634.703,34634.99,34634.9921,34634.999 lets run the workflows which are failing in 16.1.X IB |
|
A cross-check in #49984 (comment) was successful. |
|
REMINDER @sextonkennedy, @mandrenguyen, @ftenchini: This PR was tested with #49984, please check if they should be merged together |
|
It looks like I misunderstood #50039 (comment). In any case, this PR seems to be superseded by #50043. |
|
+1 Size: This PR adds an extra 24KB to repository Comparison SummaryThere are some workflows for which there are errors in the baseline: Summary:
|
PR description:
Changes in #49763 led to disruption in Phase 2 workflows, as reported in #49795. It was eventually found that the error could occur even in brand-new MC (consuming no old input files), meaning that it must be the result of an actual inconsistency, not just an incompatibility between releases.
I found one bug in #49763 (inadvertent change in the order of the list of constants), which is fixed here. However, this bug did not turn out to be the cause of the workflow errors.
Only reverting the changes in the values of the constants from #49763 fixes the workflows. I still do not know why this is the case. Every usage of these values that I can find compares them to the actual named constants, not raw/hardcoded versions of the values. But there must be some comparison somewhere that does use raw values and therefore becomes inconsistent with the change. For now, all we can do is revert the change. If the new values were in fact correct, they can be restored at a later time once all usage is understood and corrected.
PR validation:
Ran on the example failing workflow from #49795 (comment). Also ran on a privately produced sample of MC from the same generator fragment that showed the same error. The workflows run successfully with the changes in this PR.
If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:
Not a backport, should not need to be backported because this change was only introduced in 16_1_X.