The email script doesn't run on GitHub. We have a GitHub web hook that fires out to the Open MPI hostgator instance (our web hosting provider). That fires up a PHP script (i.e., GitHub calls https://...open-mpi.org/...php) and funnels it the information about the commits that were just pushed to the branch.
The PHP script basically does: git pull git log ... email << results_of_git_log However, our HostGator instance is really tuned for *web hosting*, not *script hosting*. I've seen cases where the PHP script timed out / Host Gator killed it because it took too long. I wonder if that's happening here. It might be time to move our gitdub.php script stuff to Open MPI's AWS infrastructure, where no such disk and time limits exist... > On Feb 8, 2017, at 10:24 AM, Brice Goglin <brice.gog...@inria.fr> wrote: > > FWIW, I didn't get the commit email either, and I am pretty sure it's > not the first time it happens. There are no archives for this ML, do we > have a way to see the logs of the emailing script that runs on github ? > > Brice > > > > Le 08/02/2017 16:19, Jeff Squyres (jsquyres) a écrit : >> On Feb 7, 2017, at 12:12 PM, Jeff Squyres (jsquyres) <jsquy...@cisco.com> >> wrote: >>> Fair enough, but the test itself is just a switch/case statement -- it's >>> not an actual test to see if the system supports binding or not. Hence, >>> hedging the warning message a little seemed reasonable. >> I see you actually reverted my commit (somehow I didn't get an email about >> that -- I only noticed it by chance today on GitHub.com). >> >> 1. You reverted an actual grammar fix: "support" -> "supported". >> >> 2. I don't think that "likely" is bad to have. Like I said above, the test >> itself is just a switch/case test based on a hard-coded list of OSs. The >> test does not *actually* test to see if the system supports binding. So >> weakening the language a little to say "likely" is not necessarily a bad >> thing. >> >> Sure, in some (most? all?) cases, the likelihood of not supporting binding >> will be 100%. But a) that doesn't mean the use of "likely" is incorrect, >> and b) allows for the possibility of not supporting binding to be less than >> 100% in some future / unpredicted system. >> >> "Always" (and words/phrasing like it) is a very, very strong word. It >> should be avoided when possible. >> > > _______________________________________________ > hwloc-devel mailing list > hwloc-devel@lists.open-mpi.org > https://rfd.newmexicoconsortium.org/mailman/listinfo/hwloc-devel -- Jeff Squyres jsquy...@cisco.com _______________________________________________ hwloc-devel mailing list hwloc-devel@lists.open-mpi.org https://rfd.newmexicoconsortium.org/mailman/listinfo/hwloc-devel
