You are viewing a plain text version of this content. The canonical link for it is here.
Posted to bugs@httpd.apache.org by bu...@apache.org on 2009/12/23 17:49:13 UTC
DO NOT REPLY [Bug 48440] New: Segmentation Fault
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
Summary: Segmentation Fault
Product: Apache httpd-2
Version: 2.2.14
Platform: Sun
OS/Version: Solaris
Status: NEW
Severity: critical
Priority: P2
Component: All
AssignedTo: bugs@httpd.apache.org
ReportedBy: aurabhi@yahoo.com
I have a apache 2.2.14 running Solaris 5.10
Apache is built using SunStudio 12 Update 1.
On this setup I have a custom module and wlproxy(mod_wl_22.so)
When I request pages which are delivered via wlproxy(mod_wl_22.so) I see
segmentation faults. The requests are server properly but at the end of every
request I see the following error
[Wed Dec 23 21:08:52 2009] [notice] child pid 11624 exit signal Segmentation
fault (11)
Following is what I see when I opened the core in dbx
bash-3.00# dbx httpd /tmp/apache2-gdb-dump/core
For information about new features see `help changes'
To remove this message, put `dbxenv suppress_startup_message 7.7' in your
.dbxrc
Reading httpd
core file header read successfully
Reading ld.so.1
Reading libm.so.2
Reading libaprutil-1.so.0.3.9
Reading libexpat.so.0.1.0
Reading libapr-1.so.0.3.9
Reading libuuid.so.1
Reading libsendfile.so.1
Reading librt.so.1
Reading libsocket.so.1
Reading libnsl.so.1
Reading libpthread.so.1
Reading libc.so.1
Reading libaio.so.1
Reading libmd.so.1
Reading libc_psr.so.1
Reading libCrun.so.1
Reading mod_alias.so
Reading mod_wl_22.so
Reading mod_cwmp.so
t@1 (l@1) program terminated by signal SEGV (Segmentation Fault)
0xff0465c8: __pollsys+0x0008: blu _cerror ! 0xfefa34c0
(dbx) where
current thread: t@1
=>[1] __pollsys(0x4, 0x0, 0xffbf9820, 0x0, 0xffbf9820, 0x0), at 0xff0465c8
[2] _pollsys(0xffbf97b8, 0x0, 0xffbf9820, 0x0, 0x0, 0x0), at 0xff03965c
[3] _pselect(0xffbf97b8, 0xff0723d0, 0xff0723d0, 0x0, 0xffbf9820, 0x0), at
0xfefe6b10
[4] _select(0x0, 0x0, 0x0, 0x0, 0xffbf9888, 0xf4240), at 0xfefe6e88
[5] apr_sleep(0x0, 0xf4240, 0x1, 0x1, 0x0, 0x40000), at 0xff224fe0
[6] ap_wait_or_timeout(0x11000, 0xffffffff, 0xffbf9a30, 0xa93e8, 0x7,
0xffbf99b4), at 0x4a9e8
[7] server_main_loop(0x9e5cc, 0x3f, 0x0, 0x9d324, 0x1f, 0x20), at 0x6ba6c
[8] ap_mpm_run(0x20, 0x9ea40, 0x9d320, 0x0, 0x20, 0x60), at 0x6c840
[9] main(0x9bb98, 0x9cc00, 0x7c608, 0x9ce1c, 0x9cc00, 0x7c504), at 0x299dc
Help me out please.
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
DO NOT REPLY [Bug 48440] Segmentation Fault
Posted by bu...@apache.org.
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
--- Comment #4 from aurabhi <au...@yahoo.com> 2009-12-28 02:51:30 UTC ---
I could get the stack using dbx of a particular thread which crashes. I see
that apr_file_read throws the SEGV.
The problem is the content from the file are read and request gets processed. I
get response back on my client but I see this SEGV error always. At times I see
that request times out.
I tried using std file io instead of apr_file io. In which case I see an error
in __fseek();
(dbx) cont
t@65 (l@65) signal SEGV (no mapping at the fault address) in apr_file_read at
0xff212fcc
0xff212fcc: apr_file_read+0x0020: ld [%i0 + 32], %g3
(dbx) where
current thread: t@65
=>[1] apr_file_read(0xffffffff, 0x1c7c38, 0xfad7bb0c, 0x0, 0xbc6, 0x0), at
0xff212fcc
[2] apr_file_read_full(0xffffffff, 0x1c7c38, 0xbc6, 0xfad7bb88, 0xfad7bb0c,
0x0), at 0xff211be4
[3] my_input_filter(0x1ba8b0, 0x1bbc98, 0xbc7, 0x0, 0x1c7c38, 0x0), at
0xfeee46bc
[4] ap_discard_request_body(0x1b2cf0, 0x4, 0x0, 0xc, 0x0, 0x0), at 0x60334
[5] ap_finalize_request_protocol(0x1b2cf0, 0x1b2cf0, 0xfeee5ab4, 0x14,
0x1258e0, 0x1b2cf0), at 0x32e3c
[6] my_handler(0xfffffffe, 0x4, 0x0, 0xc, 0x0, 0x0), at 0xfeee3a38
[7] ap_run_handler(0x1b6050, 0x7, 0x126470, 0x9dc80, 0x12649c, 0x1), at
0x40dc4
[8] ap_invoke_handler(0x1b6050, 0x0, 0x126544, 0x0, 0x1258e0, 0x0), at
0x413dc
[9] ap_process_request(0x1b6050, 0x4, 0x0, 0x9cc00, 0x0, 0x1b6050), at
0x5d898
[10] ap_process_http_connection(0x1ac2d8, 0x1b6050, 0xafde0, 0x9cd98,
0x80000000, 0x9cc00), at 0x5a418
[11] ap_process_connection(0x1ac2d8, 0x1ac028, 0x1268b8, 0x0, 0x9e4a0,
0x9e400), at 0x490bc
[12] worker_thread(0x129958, 0x3be, 0x1afff8, 0x1ac2d8, 0x9d340, 0x3e), at
0x6aad0
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
DO NOT REPLY [Bug 48440] Segmentation Fault
Posted by bu...@apache.org.
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
--- Comment #2 from aurabhi <au...@yahoo.com> 2009-12-23 18:59:21 UTC ---
Is there a way to get a dump from the child which exits due to the SEGV or a
way to force apache to dump core for every SEGV?
Any help would be really gr8. I am totally new to this field.
Thank you!
(In reply to comment #1)
> Could this be Bug 48030?
>
> Your dbx information is unhelpful: it just shows the parent process. If it's
> not the same as PR 48030, you should try to diagnose the role of your
> third-party module.
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
DO NOT REPLY [Bug 48440] Segmentation Fault
Posted by bu...@apache.org.
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
--- Comment #6 from aurabhi <au...@yahoo.com> 2009-12-28 08:14:54 UTC ---
(In reply to comment #5)
> (In reply to comment #4)
> > I could get the stack using dbx of a particular thread which crashes. I see
> > that apr_file_read throws the SEGV.
> >
> > The problem is the content from the file are read and request gets processed. I
> > get response back on my client but I see this SEGV error always. At times I see
> > that request times out.
> >
> > I tried using std file io instead of apr_file io. In which case I see an error
> > in __fseek();
> >
> >
> > (dbx) cont
> > t@65 (l@65) signal SEGV (no mapping at the fault address) in apr_file_read at
> > 0xff212fcc
> > 0xff212fcc: apr_file_read+0x0020: ld [%i0 + 32], %g3
> > (dbx) where
> > current thread: t@65
> > =>[1] apr_file_read(0xffffffff, 0x1c7c38, 0xfad7bb0c, 0x0, 0xbc6, 0x0), at
> > 0xff212fcc
> > [2] apr_file_read_full(0xffffffff, 0x1c7c38, 0xbc6, 0xfad7bb88, 0xfad7bb0c,
> > 0x0), at 0xff211be4
> > [3] my_input_filter(0x1ba8b0, 0x1bbc98, 0xbc7, 0x0, 0x1c7c38, 0x0), at
> > 0xfeee46bc
>
> This looks like a bug in my_input_filter in a way that it calls
> apr_file_read_full with invalid parameters. my_input_filter is no code from
> httpd. Please remove this filter and let us know if the segmentation fault goes
> away. If not please provide a stack trace without my_input_filter.
my_input_filter is a filter that I have added which modifies the BB. I can not
take this filter away as this my application requires this to talk to a
weblogic apache plugin.
If the seg fault was where I do apr_file_read_full then request flow should
fail at that point but whats happening is the requests complete and the
response is sent back to the client. I can see all the debug messages that I
have included in the debug log. I have handlers for each of the phases
including the log_transaction phase and I see all the handlers being invoked
and processed successfully.
Has this got something to do with
https://issues.apache.org/bugzilla/show_bug.cgi?id=48029?
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
DO NOT REPLY [Bug 48440] Segmentation Fault
Posted by bu...@apache.org.
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
--- Comment #5 from Ruediger Pluem <rp...@apache.org> 2009-12-28 07:50:51 CET ---
(In reply to comment #4)
> I could get the stack using dbx of a particular thread which crashes. I see
> that apr_file_read throws the SEGV.
>
> The problem is the content from the file are read and request gets processed. I
> get response back on my client but I see this SEGV error always. At times I see
> that request times out.
>
> I tried using std file io instead of apr_file io. In which case I see an error
> in __fseek();
>
>
> (dbx) cont
> t@65 (l@65) signal SEGV (no mapping at the fault address) in apr_file_read at
> 0xff212fcc
> 0xff212fcc: apr_file_read+0x0020: ld [%i0 + 32], %g3
> (dbx) where
> current thread: t@65
> =>[1] apr_file_read(0xffffffff, 0x1c7c38, 0xfad7bb0c, 0x0, 0xbc6, 0x0), at
> 0xff212fcc
> [2] apr_file_read_full(0xffffffff, 0x1c7c38, 0xbc6, 0xfad7bb88, 0xfad7bb0c,
> 0x0), at 0xff211be4
> [3] my_input_filter(0x1ba8b0, 0x1bbc98, 0xbc7, 0x0, 0x1c7c38, 0x0), at
> 0xfeee46bc
This looks like a bug in my_input_filter in a way that it calls
apr_file_read_full with invalid parameters. my_input_filter is no code from
httpd. Please remove this filter and let us know if the segmentation fault goes
away. If not please provide a stack trace without my_input_filter.
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
DO NOT REPLY [Bug 48440] Segmentation Fault
Posted by bu...@apache.org.
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
Nick Kew <ni...@webthing.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |NEEDINFO
--- Comment #1 from Nick Kew <ni...@webthing.com> 2009-12-23 09:27:16 UTC ---
Could this be Bug 48030?
Your dbx information is unhelpful: it just shows the parent process. If it's
not the same as PR 48030, you should try to diagnose the role of your
third-party module.
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org
DO NOT REPLY [Bug 48440] Segmentation Fault
Posted by bu...@apache.org.
https://issues.apache.org/bugzilla/show_bug.cgi?id=48440
--- Comment #3 from Ruediger Pluem <rp...@apache.org> 2009-12-24 07:25:15 CET ---
http://httpd.apache.org/docs/2.2/mod/mpm_common.html#coredumpdirectory
http://httpd.apache.org/dev/debugging.html
--
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@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org