apr-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From r..@rkbloom.net
Subject RE: SEGV in allocator_free
Date Fri, 19 Mar 2004 20:26:41 GMT

These are different scenarios, unless sslswamp is sending a signal to the Apache
server for some reason.  In the attached e-mail, somebody is trying to run pool
cleanup code from within a signal handler.  But running code within a signal
handler is always dangerous, and it generally shouldn't be done.  If you look at
what Apache does, it just sets a flag that is noticed later.  That flag causes
the shutdown.

As for what is happening in your case, we would need more information to help
debug it.  You mentioned that you were testing an SSL proxy.  Is the proxy
segfaulting or the origin server?  My bet is that there is a pool lifetime issue
in the mod_ssl code.  Pool lifetime with filters is kind of tricky.

Ryan

Quoting "Mathihalli, Madhusudan" <madhum@hp.com>:

> I've attached another mail that discusses a similar scenario.
> 
> In my case, the apr_pool_clear fails when sslswamp finishes.
> 
> -Madhu
> 
> >-----Original Message-----
> >From: Sander Striker [mailto:striker@apache.org]
> >Sent: Friday, March 19, 2004 10:51 AM
> >To: Mathihalli, Madhusudan
> >Cc: dev@apr.apache.org; dev@httpd.apache.org
> >Subject: RE: SEGV in allocator_free
> >
> >
> >On Fri, 2004-03-19 at 19:41, Mathihalli, Madhusudan wrote:
> >> Well - there might as-well be a bug in httpd (I don't deny that)
> >> 
> >> But shouldn't APR protect itself against NULL pointers in 
> >allocator_free ?
> >
> >And then what?  abort()?  Also note that this can only happen through
> >pool misuse (or a severe bug in the pools code).  So, running with
> >pool debugging enabled should point out where things go wrong.
> >
> >Sander
> >
> 
> 
> 
> >-----Original Message-----
> >From: Joe Orton [mailto:joe@manyfish.co.uk]
> >Sent: Thursday, January 29, 2004 3:44 AM
> >To: dev@apr.apache.org
> >Subject: Re: apr_pool_clear fails if the cleanup handler is still
> >running
> >
> >
> >On Wed, Jan 28, 2004 at 01:57:24PM -0800, Stas Bekman wrote:
> >> It doesn't seem like people are very excited about Colm's 
> >proposal. ;(
> >>
> >> I guess I should at least document that pool cleanups and 
> >server shut-down 
> >> don't go together. Which renders pool cleanups in apache as totally 
> >> unreliable, and shouldn't be used for anything crucial.
> >
> >It's not really a problem specific to cleanups: if the child was doing
> >any pool operation (be that allocation, destruction, etc) when the
> >SIGTERM arrives, it could also die horribly.
> >
> >The undocumented assumption is just that most use of APR from a signal
> >handler is totally unreliable.
> >
> >(I do like Colm's proposal, BTW, I'll follow up on that post)
> >
> >joe
> >
> 




Mime
View raw message