On Fri, Feb 08, 2008 at 01:01:06PM +0200, Janno Sannik wrote: > Sometimes it happends to us too. Although we fixed graceful initially, > we haven't figured out why this happens sometimes. > It tend's to happen more often if you make many graceful restarts in a > row. So it seems like they segfault if thay are in some kind on "state" > (maybe not yet on standby/ready) on the time graceful hits them. >
The only state I know that is kept in Apache (sometimes?) between restarts, which I've had problems with in the past (and I don't even use peruser), is the semaphores: # ipcs -s > Alexey Vlasov wrote: > >Hi. > > > >On 1 Dec 2007 г. 14:38:59, Alexey Vlasov wrote about "[peruser] graceful > >and > >peruser": > > > > > >>[Sat Dec 01 13:41:36 2007] [notice] child pid 16658 exit signal > >> > >Segmentation fault (11) > > > >>[Sat Dec 01 13:41:36 2007] [notice] child pid 10310 exit signal > >> > >Segmentation fault (11) > > > >>[Sat Dec 01 13:41:36 2007] [notice] child pid 12312 exit signal > >> > >Segmentation fault (11) > > > >>[Sat Dec 01 13:41:36 2007] [notice] child pid 12837 exit signal > >> > >Segmentation fault (11) > > > >>[Sat Dec 01 13:41:36 2007] [notice] child pid 14128 exit signal > >> > >Segmentation fault (11) > > > >These PIDs belong to processes of Multiplexers and Processors. Why after > >graceful restart do they end with segmentation fault? > >Does peruser always act this way, or only i have such problems with it? > >I use Apache 2.2.3/PHP 5.2.3(5), peruser 0.3.0. > > > >-- > >BRGDS. Alexey Vlasov. > > > >_______________________________________________ > >Peruser mailing list > >[email protected] > >http://www.telana.com/mailman/listinfo/peruser > > > > _______________________________________________ > Peruser mailing list > [email protected] > http://www.telana.com/mailman/listinfo/peruser _____________________________________ New things are always on the horizon. _______________________________________________ Peruser mailing list [email protected] http://www.telana.com/mailman/listinfo/peruser
