You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by rb...@covalent.net on 2001/02/06 15:47:23 UTC
Re: core dump from mod_include/cgi change?
I never saw this in my testing yesterday. The ht variable looks wrong to
me, you could try looking into that. I am willing to bet dollars to
donuts that the problem is the order of your modules. I didn't think of
this yesterday.
This can be fixed by editing mod_cgi and mod_cgid, and making sure that
the Successor field for their post_config_hook is "Mod_include". The
problem is that we create the hash table in mod_include's post_config and
use it in mod_cgi(d)'s. If mod_include is called after mod_cgi(d), then
the things will crap out.
Ryan
On Wed, 7 Feb 2001, Brian Havard wrote:
> Yesterday, after a cvs up I started getting an access violation at start-up
> that looks related to the recent change to mod_include. Does this look
> familiar to anyone?
>
> (gdb) bt
> #0 0x4ae9c in find_entry (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
> at apr_hash.c:261
> #1 0x4af70 in apr_hash_set (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
> at apr_hash.c:303
> #2 0x15307 in ap_register_include_handler (tag=0x321c6 "exec",
> func=0x31e50 <handle_exec>) at mod_include.c:2746
> #3 0x32222 in cgi_post_config (p=0x961ec, plog=0xc048c, ptemp=0xc44cc,
> s=0x96f4c) at mod_cgi.c:986
> #4 0x34541 in ap_run_post_config (pconf=0x961ec, plog=0xc048c,
> ptemp=0xc44cc,
> s=0x96f4c) at config.c:129
> #5 0x36cd8 in ap_post_config_hook (pconf=0x961ec, plog=0xc048c,
> ptemp=0xc44cc, s=0x96f4c) at config.c:1629
> #6 0x389f8 in main (argc=3, argv=0x288ffc0) at main.c:426
>
>
> configured with only --with-debug --enable-status, after a cvsclean
>
> --
> ______________________________________________________________________________
> | Brian Havard | "He is not the messiah! |
> | brianh@kheldar.apana.org.au | He's a very naughty boy!" - Life of Brian |
> ------------------------------------------------------------------------------
>
>
_______________________________________________________________________________
Ryan Bloom rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------
Re: core dump from mod_include/cgi change?
Posted by rb...@covalent.net.
This should be fixed now.
Ryan
On Tue, 6 Feb 2001 rbb@covalent.net wrote:
>
> I'll fix it right now. The order is determined by the ap_hook_post_config
> call. You'll see it in my commit. That's probably easier than trying to
> explain it.
>
> Ryan
>
> On Wed, 7 Feb 2001, Brian Havard wrote:
>
> > On Tue, 6 Feb 2001 06:47:23 -0800 (PST), rbb@covalent.net wrote:
> >
> > >I never saw this in my testing yesterday. The ht variable looks wrong to
> > >me, you could try looking into that. I am willing to bet dollars to
> > >donuts that the problem is the order of your modules. I didn't think of
> > >this yesterday.
> > >
> > >This can be fixed by editing mod_cgi and mod_cgid, and making sure that
> > >the Successor field for their post_config_hook is "Mod_include". The
> > >problem is that we create the hash table in mod_include's post_config and
> > >use it in mod_cgi(d)'s. If mod_include is called after mod_cgi(d), then
> > >the things will crap out.
> >
> > It's compiled fully static so I'm not sure what determines the order. I
> > get:
> >
> > F:\Apache\httpd-2.0>httpd.exe -l
> > Compiled in modules:
> > mod_access.c
> > mod_auth.c
> > mod_include.c
> > http_core.c
> > mod_mime.c
> > mod_log_config.c
> > mod_negotiation.c
> > mod_dir.c
> > mod_imap.c
> > mod_actions.c
> > mod_userdir.c
> > mod_alias.c
> > mod_env.c
> > mod_setenvif.c
> > spmt_os2.c
> > mod_status.c
> > mod_autoindex.c
> > mod_asis.c
> > mod_cgi.c
> >
> >
> >
> > >On Wed, 7 Feb 2001, Brian Havard wrote:
> > >
> > >> Yesterday, after a cvs up I started getting an access violation at start-up
> > >> that looks related to the recent change to mod_include. Does this look
> > >> familiar to anyone?
> > >>
> > >> (gdb) bt
> > >> #0 0x4ae9c in find_entry (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
> > >> at apr_hash.c:261
> > >> #1 0x4af70 in apr_hash_set (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
> > >> at apr_hash.c:303
> > >> #2 0x15307 in ap_register_include_handler (tag=0x321c6 "exec",
> > >> func=0x31e50 <handle_exec>) at mod_include.c:2746
> > >> #3 0x32222 in cgi_post_config (p=0x961ec, plog=0xc048c, ptemp=0xc44cc,
> > >> s=0x96f4c) at mod_cgi.c:986
> > >> #4 0x34541 in ap_run_post_config (pconf=0x961ec, plog=0xc048c,
> > >> ptemp=0xc44cc,
> > >> s=0x96f4c) at config.c:129
> > >> #5 0x36cd8 in ap_post_config_hook (pconf=0x961ec, plog=0xc048c,
> > >> ptemp=0xc44cc, s=0x96f4c) at config.c:1629
> > >> #6 0x389f8 in main (argc=3, argv=0x288ffc0) at main.c:426
> > >>
> > >>
> > >> configured with only --with-debug --enable-status, after a cvsclean
> >
> > --
> > ______________________________________________________________________________
> > | Brian Havard | "He is not the messiah! |
> > | brianh@kheldar.apana.org.au | He's a very naughty boy!" - Life of Brian |
> > ------------------------------------------------------------------------------
> >
> >
>
>
> _______________________________________________________________________________
> Ryan Bloom rbb@apache.org
> 406 29th St.
> San Francisco, CA 94131
> -------------------------------------------------------------------------------
>
>
_______________________________________________________________________________
Ryan Bloom rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------
Re: core dump from mod_include/cgi change?
Posted by rb...@covalent.net.
I'll fix it right now. The order is determined by the ap_hook_post_config
call. You'll see it in my commit. That's probably easier than trying to
explain it.
Ryan
On Wed, 7 Feb 2001, Brian Havard wrote:
> On Tue, 6 Feb 2001 06:47:23 -0800 (PST), rbb@covalent.net wrote:
>
> >I never saw this in my testing yesterday. The ht variable looks wrong to
> >me, you could try looking into that. I am willing to bet dollars to
> >donuts that the problem is the order of your modules. I didn't think of
> >this yesterday.
> >
> >This can be fixed by editing mod_cgi and mod_cgid, and making sure that
> >the Successor field for their post_config_hook is "Mod_include". The
> >problem is that we create the hash table in mod_include's post_config and
> >use it in mod_cgi(d)'s. If mod_include is called after mod_cgi(d), then
> >the things will crap out.
>
> It's compiled fully static so I'm not sure what determines the order. I
> get:
>
> F:\Apache\httpd-2.0>httpd.exe -l
> Compiled in modules:
> mod_access.c
> mod_auth.c
> mod_include.c
> http_core.c
> mod_mime.c
> mod_log_config.c
> mod_negotiation.c
> mod_dir.c
> mod_imap.c
> mod_actions.c
> mod_userdir.c
> mod_alias.c
> mod_env.c
> mod_setenvif.c
> spmt_os2.c
> mod_status.c
> mod_autoindex.c
> mod_asis.c
> mod_cgi.c
>
>
>
> >On Wed, 7 Feb 2001, Brian Havard wrote:
> >
> >> Yesterday, after a cvs up I started getting an access violation at start-up
> >> that looks related to the recent change to mod_include. Does this look
> >> familiar to anyone?
> >>
> >> (gdb) bt
> >> #0 0x4ae9c in find_entry (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
> >> at apr_hash.c:261
> >> #1 0x4af70 in apr_hash_set (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
> >> at apr_hash.c:303
> >> #2 0x15307 in ap_register_include_handler (tag=0x321c6 "exec",
> >> func=0x31e50 <handle_exec>) at mod_include.c:2746
> >> #3 0x32222 in cgi_post_config (p=0x961ec, plog=0xc048c, ptemp=0xc44cc,
> >> s=0x96f4c) at mod_cgi.c:986
> >> #4 0x34541 in ap_run_post_config (pconf=0x961ec, plog=0xc048c,
> >> ptemp=0xc44cc,
> >> s=0x96f4c) at config.c:129
> >> #5 0x36cd8 in ap_post_config_hook (pconf=0x961ec, plog=0xc048c,
> >> ptemp=0xc44cc, s=0x96f4c) at config.c:1629
> >> #6 0x389f8 in main (argc=3, argv=0x288ffc0) at main.c:426
> >>
> >>
> >> configured with only --with-debug --enable-status, after a cvsclean
>
> --
> ______________________________________________________________________________
> | Brian Havard | "He is not the messiah! |
> | brianh@kheldar.apana.org.au | He's a very naughty boy!" - Life of Brian |
> ------------------------------------------------------------------------------
>
>
_______________________________________________________________________________
Ryan Bloom rbb@apache.org
406 29th St.
San Francisco, CA 94131
-------------------------------------------------------------------------------
Re: core dump from mod_include/cgi change?
Posted by Brian Havard <br...@kheldar.apana.org.au>.
On Tue, 6 Feb 2001 06:47:23 -0800 (PST), rbb@covalent.net wrote:
>I never saw this in my testing yesterday. The ht variable looks wrong to
>me, you could try looking into that. I am willing to bet dollars to
>donuts that the problem is the order of your modules. I didn't think of
>this yesterday.
>
>This can be fixed by editing mod_cgi and mod_cgid, and making sure that
>the Successor field for their post_config_hook is "Mod_include". The
>problem is that we create the hash table in mod_include's post_config and
>use it in mod_cgi(d)'s. If mod_include is called after mod_cgi(d), then
>the things will crap out.
It's compiled fully static so I'm not sure what determines the order. I
get:
F:\Apache\httpd-2.0>httpd.exe -l
Compiled in modules:
mod_access.c
mod_auth.c
mod_include.c
http_core.c
mod_mime.c
mod_log_config.c
mod_negotiation.c
mod_dir.c
mod_imap.c
mod_actions.c
mod_userdir.c
mod_alias.c
mod_env.c
mod_setenvif.c
spmt_os2.c
mod_status.c
mod_autoindex.c
mod_asis.c
mod_cgi.c
>On Wed, 7 Feb 2001, Brian Havard wrote:
>
>> Yesterday, after a cvs up I started getting an access violation at start-up
>> that looks related to the recent change to mod_include. Does this look
>> familiar to anyone?
>>
>> (gdb) bt
>> #0 0x4ae9c in find_entry (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
>> at apr_hash.c:261
>> #1 0x4af70 in apr_hash_set (ht=0xf49cc, key=0x321c6, klen=5, val=0x31e50)
>> at apr_hash.c:303
>> #2 0x15307 in ap_register_include_handler (tag=0x321c6 "exec",
>> func=0x31e50 <handle_exec>) at mod_include.c:2746
>> #3 0x32222 in cgi_post_config (p=0x961ec, plog=0xc048c, ptemp=0xc44cc,
>> s=0x96f4c) at mod_cgi.c:986
>> #4 0x34541 in ap_run_post_config (pconf=0x961ec, plog=0xc048c,
>> ptemp=0xc44cc,
>> s=0x96f4c) at config.c:129
>> #5 0x36cd8 in ap_post_config_hook (pconf=0x961ec, plog=0xc048c,
>> ptemp=0xc44cc, s=0x96f4c) at config.c:1629
>> #6 0x389f8 in main (argc=3, argv=0x288ffc0) at main.c:426
>>
>>
>> configured with only --with-debug --enable-status, after a cvsclean
--
______________________________________________________________________________
| Brian Havard | "He is not the messiah! |
| brianh@kheldar.apana.org.au | He's a very naughty boy!" - Life of Brian |
------------------------------------------------------------------------------