apr-bugs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 48029] httpd-2.2.14 hangs in port_getn
Date Wed, 21 Oct 2009 18:52:01 GMT
https://issues.apache.org/bugzilla/show_bug.cgi?id=48029

--- Comment #2 from Jan Schmidt <httpd.apache@jan-o-sch.net> 2009-10-21 11:52:00 UTC
---
First: Yes, the amount of time passing until the request "completes" roughly
fits the config's Timeout value.

Unfortunately getting a truss of what is going on is not possible. I forget to
mention why we used dtrace to see what is going on - and that bit might be
interesting, too: When running with truss, everything works fine. Reproducibly.
Might be a hint to some kind of weird timing issue.

I was expecting the return value showed by dtrace for "libc.so.1`_portfs+0x15"
in [2] (which is 1), was the return value of port_getn.

When trussing the already hanging child process in [8], I see it sleeping in
port_getn. dtrace states it is in _portfs. pstack by the way also states
_portfs. Thus I guess port_getn is the high level name for _portfs.

If that is correct, then your explanation with 11 and EAGAIN does not add up,
as dtrace shows port_getn really returned 1. However, 1 would mean EPERM, which
is no documented return code for port_getn, either.

To gather more information, we could play around with dtrace somewhat further,
if we knew what we are looking for.

-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@apr.apache.org
For additional commands, e-mail: bugs-help@apr.apache.org


Mime
View raw message