httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ishita Kapadiya <ishim...@gmail.com>
Subject Re: [users@httpd] Segmentation fault error
Date Wed, 30 May 2012 21:59:56 GMT
Thanks William. This is the stack output i got from mdb when core got
dumped. Also, when i google "ap_mpm_pod_check" then got to know that
there are lots of people who hits Apache bug.

Please refer this URL - http://www.gossamer-threads.com/lists/apache/bugs/414768

Also, i ran pfiles against the core file and i got

        data model = _ILP32  flags = MSACCT|MSFORK
 /1:    flags = 0
        sigmask = 0xffffbeff,0x0000ffff  cursig = SIGSEGV
 /2:    <defunct>
 /3:    flags = STOPPED  lwp_park(0x4,0x0,0x0)
        why = PR_SUSPENDED
        sigmask = 0xffbe6007,0x0000fff7
.. (rest of lines truncted)

I don't know much about debugging as i am not from developer
background but whatever o/p i got, looks like some bug with Apache
2.2.22.

Please revert.

On Wed, May 30, 2012 at 4:43 PM, William A. Rowe Jr.
<wrowe@rowe-clan.net> wrote:
> You didn't dump the offending stack, you dumped the first stack.  It's highly
> unlikely there was a segfault in _read.
>
> You need to dump all the thread stacks, and work out the offending one; this is
> usuallly designated <<< FAULT or some other indication of where the fault occured.
>
> On 5/30/2012 10:32 AM, Ishita Kapadiya wrote:
>> Thanks Nick.
>> I have compiled Apache for myself both the time and both are 32-bit.
>> The same modules all other instances are using and thus i am not sure
>> what is different with this instance that causing Segmentation fault
>> error.
>>
>> I tried to dig more into it and here is what i got -
>>
>> mdb core
>>> ::stack
>> libc.so.1`_read+0xc(6, ffbff5e3, 1, 0, 10b4, fef73ac0)
>> ap_mpm_pod_check+0x18(d6800, 68764, 68f8c, 1b7ec0, 0, 1)
>> child_main+0x2d4(0, 682dc, 0, fee58000, fef73700, fedf2a00)
>> make_child+0x128(9bc00, 0, 1, 9cc00, 9b400, 9c800)
>> ap_mpm_run+0x740(fe720058, 4, 0, a, 1, 0)
>> main+0x77c(a7810, 99c00, 9bc00, 9bc00, a5808, 0)
>> _start+0x5c(0, 0, 0, 0, 0, 0)
>>
>> pstack core
>>  fef45874 _read    (6, ffbff5e3, 1, 0, 10b4, fef73ac0) + c
>>  0006b714 ap_mpm_pod_check (d6800, 68764, 68f8c, 1b7ec0, 0, 1) + 18
>>  000697b4 child_main (0, 682dc, 0, fee58000, fef73700, fedf2a00) + 2d4
>>  00069930 make_child (9bc00, 0, 1, 9cc00, 9b400, 9c800) + 128
>>  0006a160 ap_mpm_run (fe720058, 4, 0, a, 1, 0) + 740
>>  00029bc8 main     (a7810, 99c00, 9bc00, 9bc00, a5808, 0) + 77c
>>  00028f7c _start   (0, 0, 0, 0, 0, 0) + 5c
>>
>> pmap core
>> 00010000     448K r-x--  /abc/apache-2.2.22/bin/httpd
>> 00080000      32K r-x--
>> 00096000      24K rwx--  /abc/apache-2.2.22/bin/httpd
>> 0009C000      16K rwx--  /abc/apache-2.2.22/bin/httpd
>> 000A0000    6528K rwx--    [ heap ]
>> FAB7A000       8K rw---    [ stack tid=36 ]
>> ...
>> (removed rest of the lines to avoid length)
>>
>> mdb /abc/apache-2.2.22/bin/httpd
>>
>>> ::dis ap_mpm_pod_check!head
>> ap_mpm_pod_check:               save      %sp, -0x78, %sp
>> ap_mpm_pod_check+4:             ld        [%i0], %o1
>> ap_mpm_pod_check+8:             call      +0x2d0e0      <PLT:apr_os_file_get>
>> ap_mpm_pod_check+0xc:           add       %fp, -0x14, %o0
>> ap_mpm_pod_check+0x10:          ld        [%fp - 0x14], %o0
>> ap_mpm_pod_check+0x14:          add       %fp, -0x15, %o1
>> ap_mpm_pod_check+0x18:          call      +0x2cdc4      <PLT:read>
>> ap_mpm_pod_check+0x1c:          mov       1, %o2
>>
>> Please help me what could be the problem? It's affecting my production
>> env and i really don't know what to do next?
>>
>> On Tue, May 29, 2012 at 7:15 PM, Nick Kew <nick@webthing.com> wrote:
>>>
>>> On 29 May 2012, at 23:11, Ishita Kapadiya wrote:
>>>
>>>> Hi All,
>>>>
>>>> I am using this configurations -
>>>>
>>>> Solaris sparc 10/apache 2.2.22/openssl 1.0.0g/simteminder sso/mod-jk 1.30
>>>
>>> Did you compile everything yourself?
>>>
>>> If yes, could any compile options have changed?  E.g. between 32-bit and 64-bit,
>>> or something less obvious but just as important?
>>>
>>> If no, what suppliers do your binaries come from, and have you checked with them?
>>>
>>> --
>>> Nick Kew
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
>>> For additional commands, e-mail: users-help@httpd.apache.org
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
>> For additional commands, e-mail: users-help@httpd.apache.org
>>
>>
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
> For additional commands, e-mail: users-help@httpd.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org


Mime
View raw message