On Tue, 7 Mar 2006, James C. McPherson wrote: > Ashok Nair wrote: >> This is the pstack o/p for the first thread that shows up in "pstack >> corefile" >> >> ----------------- lwp# 1 / thread# 1 -------------------- >> ff01f180 _read (5, ffbff9ab, 1, 175378, ffffffe0, 175395) + c >> 0005d15c ap_mpm_pod_check (1648f0, 58418, 59138, 175358, 175280, 2596f5) + >> 34 >> 00059de4 child_main (0, 57bc8, 0, 0, 0, 0) + 474 >> 0005a06c make_child (102670, 0, 0, fffffff8, ffffffe0, 174865) + 13c >> 0005a234 startup_children (2, 2, 5, 100878, f0, 0) + 6c >> 0005af40 ap_mpm_run (100878, 12c928, 102670, 102670, 0, 0) + 2d0 >> 00067a08 main (5, ffbffca4, ffbffcbc, eb400, 0, 0) + d38 >> 0004c378 _start (0, 0, 0, 0, 0, 0) + 108 > > Hi Ashok, > I goooooogled for ap_mpm_pod_check and came up with these > two links:
The Apache POD is used to send signals between the Apache Parent and child processes, and it typically used along with the scoreboard generation to tell a child to die after a restart or graceful shutdown is issued. If you are interested in learning more about the POD, you can review pod.h and pod.c. To see what is causing the problem, I would recommend installing mod_backtrace (there is a Solaris specific patch for Jeff's module on my website.). Hope this helps, - Ryan -- UNIX Administrator http://daemons.net/~matty
