Stacktrace handler revisited #11329
Stacktrace handler revisited #11329
Conversation
We've been unable to figure out why GDB sometimes does not print out the traceback of the thread invoking clone(). Accordingly, this patch launches a helper thread at the beginning of the process; the signal handler will notify helper thread to print out a traceback and wait for completion. (cherry picked from commit 99b873e)
(cherry picked from commit 240c5b0)
Make sure that the child thread only talks to pipes in the *current* process; before, the forked child process would unexpectedly talk to the child thread of the parent process, triggering an exception.
|
A new Pull Request was created by @bbockelm (Brian Bockelman) for CMSSW_7_6_X. Stacktrace handler revisited It involves the following packages: FWCore/Services @cmsbuild, @smuzaffar, @Dr15Jones can you please review it and eventually sign? Thanks. |
|
please test |
|
@bbockelm did you test when forking that a crash in a child process would give a traceback? |
|
The tests are being triggered in jenkins. |
|
Just tested this. stacktrace printer works fine in the child process. However, it seems that the parent gets wedged in Xrootd-land when the child SEGV's. Doesn't appear to be related to this, however. |
|
+1 |
cmsbuild
added
core-approved
fully-signed
and removed
core-pending
pending-signatures
labels
|
This pull request is fully signed and it will be integrated in one of the next CMSSW_7_6_X IBs (tests are also fine). This pull request requires discussion in the ORP meeting before it's merged. @davidlange6, @Degano, @smuzaffar |
|
This pull request is fully signed and it will be integrated in one of the next CMSSW_7_6_X IBs (tests are also fine). This pull request requires discussion in the ORP meeting before it's merged. @davidlange6, @Degano, @smuzaffar |
|
+1 |
This is the second version of the improved stacktrace. The first version had improper handling around forked processes (the helper thread would communicate with the wrong process).