2009/2/23 gert <[email protected]>: > > On Feb 18, 11:17 am, Graham Dumpleton <[email protected]> > wrote: >> I inadvertantly saw one crash when doing something else, but can't >> otherwise duplicate it. Let me know if you ever work out a narrow use >> case or example which reliably shows it. >> >> I'm going to go back to more important things for the time being. >> >> Graham >> >> 2009/2/17 Graham Dumpleton <[email protected]>: >> >> > 2009/2/17 gert <[email protected]>: >> >> Here is the deal, first you let the a interpreter crash (500) syntax >> >> error or something, then when you want to clean up that specific >> >> crashed interpreter, boom segfault :) >> >> >> [Tue Feb 17 00:49:22 2009] [info] mod_wsgi (pid=7085): Destroy >> >> interpreter '127.0.0.1|/appwsgi/www/invoice/invoice.wsgi'. >> >> [Tue Feb 17 00:49:22 2009] [info] mod_wsgi (pid=7085): Cleanup >> >> interpreter ''. >> >> [Tue Feb 17 00:49:22 2009] [info] mod_wsgi (pid=7085): Terminating >> >> Python. >> >> [Tue Feb 17 00:49:23 2009] [notice] child pid 7085 exit signal >> >> Segmentation fault (11) >> >> > Thanks for trying to narrow it down. I'll try with exceptions at >> > global scope on first import and within request handler itself. Likely >> > I am not clearing some exception state properly, or leaving a >> > reference to object which has since been destroyed. On shutdown >> > probably checking the value and so crashing. >> > > [Sun Feb 22 14:44:01 2009] [info] mod_wsgi (pid=27188): Destroy > interpreter '127.0.0.1|/appwsgi/www/invoice/invoice.wsgi'. > [Sun Feb 22 14:44:01 2009] [info] mod_wsgi (pid=27188): Cleanup > interpreter ''. > [Sun Feb 22 14:44:01 2009] [info] mod_wsgi (pid=27188): Terminating > Python. > > [Sun Feb 22 14:44:01 2009] [info] [client 80.201.198.141] mod_wsgi > (pid=27187, process='', application='127.0.0.1|/appwsgi/www/invoice/ > invoice.wsgi'): Loading WSGI script '/usr/httpd/www/appwsgi/www/ > invoice/invoice.wsgi'., referer: > http://91.121.53.159/appwsgi/www/invoice/invoice.htm?1-admin > > [Sun Feb 22 14:44:01 2009] [error] ('SELECT uid,sid,exp FROM sessions > WHERE sid=? AND exp>?', ('d4db489597f2eeab', '2009-02-22 13:44:01')) > [Sun Feb 22 14:44:01 2009] [error] [('gert', 'd4db489597f2eeab', > '2009-02-22 14:43:52')] > [Sun Feb 22 14:44:01 2009] [error] ('SELECT gid FROM groups WHERE > uid=? AND gid=?', ('gert', 'admin')) > [Sun Feb 22 14:44:01 2009] [error] [('admin',)] > [Sun Feb 22 14:44:01 2009] [error] ('UPDATE sessions SET exp=? WHERE > sid=? ', ('2009-02-22 14:44:01', 'd4db489597f2eeab')) > [Sun Feb 22 14:44:01 2009] [error] ('SELECT bid FROM shop_orders WHERE > oid=?', ('1',)) > [Sun Feb 22 14:44:01 2009] [error] [(1,)] > [Sun Feb 22 14:44:01 2009] [error] ('UPDATE shop_orders SET bid=? > WHERE oid=?', ('3', '1')) > > [Sun Feb 22 14:44:02 2009] [notice] child pid 27188 exit signal > Segmentation fault (11) > > Can you completely comment out the "Python has shutdown." part and see > if you are getting segfaults ? > > notice the 27188 pid id shutdown getting skipped and crashes after pid > 27187 is done.
They are separate processes and so messages could interleave like that. There is nothing particularly special about that. Graham --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "modwsgi" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/modwsgi?hl=en -~----------~----~----~----~------~----~------~--~---
