New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Phase1 Pixel DQM Bugfixes #14760
Phase1 Pixel DQM Bugfixes #14760
Conversation
A new Pull Request was created by @schneiml (Marcel Schneider) for CMSSW_8_1_X. It involves the following packages: DQM/SiPixelPhase1Common @cmsbuild, @dmitrijus, @vanbesien, @davidlange6 can you please review it and eventually sign? Thanks. cms-bot commands are list here #13028 |
what is the reason for a crash now? |
I have honestly no idea what this sequence does, but to me this is a sign that it is not DQM related. |
The handle seems to silenty get invalid at some point. A copy would be even safer.
…f the HistogramManager.
46c08b9
to
e5dd360
Compare
Pull request #14760 was updated. @cmsbuild, @dmitrijus, @vanbesien, @davidlange6 can you please check and sign again. |
On 6/3/16 6:51 AM, Marcel Schneider wrote:
Probably not, but I guess the only way is to check in a test build with
|
Are there instructions on how to do a clean revert (with cmssw packages etc.)? |
@slava77 I am fairly certain the new crash was introduced by this #14608 .
There are two bugs here: one is that the Relevant lines in output:
notice the |
Having debugged these kind of cases (too) many times, I'd much prefer to get an early crash with a clear error message with as much useful information as possible to help fixing the cause. I think mapping this case to empty sequence would add confusion and make debugging more difficult. |
On 6/7/16 2:16 AM, Marcel Schneider wrote:
Thanks for checking.
|
@dmitrijus My guess is that this PR also fixes this issue, at least for On 07.06.2016 15:46, Dmitrijus wrote:
|
The refs got invalid when the vector grows, which caused cms-sw#14800.
Pull request #14760 was updated. @cmsbuild, @dmitrijus, @vanbesien, @davidlange6 can you please check and sign again. |
+1 |
The tests are being triggered in jenkins. |
This pull request is fully signed and it will be integrated in one of the next CMSSW_8_1_X IBs after it passes the integration tests. This pull request requires discussion in the ORP meeting before it's merged. @slava77, @davidlange6, @Degano, @smuzaffar |
@dmitrijus I was about to ping you for this one. As far as I see this should fix all known issues with the Phase1 DQM as of now. |
+1 |
This PR fixes some smaller bugs that were found in the first version merged in #14586 (also @fioriNTU ).
edm::ESHandle
to TT, and that seemed to silently become invalid on rare occasions. The fix is to use a bare pointer, which is a bad solution but works. While I can understand that the Handle or also the pointer gets invalid, what I do NOT understand is why this happens during the DQM booking phase (maybe also elsewhere, but definitely there) and only in one out of 6 plugins, that all run the same code. Maybe there is more to it. (The plugin in question is SiPixelPhase1RecHits, the symptom is that many histograms are not booked, and changing to pointers fixes it).