Hi, in the past couple moths we have seen many reports that indicate that we are still having pexpect hangs, especially with Maxima. Since none of those happen on systems where we usually have direct access it is very difficult the debug the problem. A while ago we introduced some settings that will make debugging these somewhat easier since the traffic of such failures is logged on demand.
So if you encounter timeouts and they are clearly not caused by a slow system please do the following. First export SAGE_PEXPECT_LOG=yes this environment variable will cause all pexpect traffic to be logged to files in $DOT_SAGE/pexpect_log, i.e. usually $HOME/.sage/ pexpect_log. If you don't use bash you have to use setenv in case you use [t]csh/ksh, but I would assume there is something wrong with you for using such a shell in the first place :p Then identify all tests that give you timeouts during a "make check". const.tex is a common one it seems and so is calculus.py. For example now run ./sage -t -verbose devel/doc/const/const.tex <SNIP> At some point this should hang while telling you which test is running. Now look in $DOT_SAGE/pexpect_log. You should see something like the following: singular-31329-14103328-2.log gap-31329-14105248-2.log pari-31329-73925648-2.log pari-31329-73925168-2.log maxima-31329-27545184-2.log maxima-31329-15330912-2.log Identify the youngest file (ls -latr) and upload it to some place where it can be downloaded. Since those files tend to be large do not attach them to tickets. Then copy and paste the last 20 or so lines from the doctest output into a sage-devel thread with a meaning ful summary, give the system info and anything else that might be useful and post a link to the log in that message. One of us will take a closer look and once we have identified the problem we will open a ticket and work on a fix. I will not put this info into the wiki since it belongs in the documentation, so I will add it post Sage 3.4. Since we are also switching to ecl in Sage 3.4 I expect various other pexpect related problem to crop up in and post 3.4, so the above should come in handy. Cheers, Michael --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
