httpd-bugs mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 52291] New: Apache Child has Problems with LD_LIBRARY_PATH (php/oracle)
Date Tue, 06 Dec 2011 09:26:26 GMT
https://issues.apache.org/bugzilla/show_bug.cgi?id=52291

             Bug #: 52291
           Summary: Apache Child has Problems with LD_LIBRARY_PATH
                    (php/oracle)
           Product: Apache httpd-2
           Version: 2.2.19
          Platform: PC
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: All
        AssignedTo: bugs@httpd.apache.org
        ReportedBy: feigl@ertl-glas.at
    Classification: Unclassified


Hello,

i seen this issue since years on many different versions of apache2 and php.

At the moment i've seen this with:
Apache 2.2.16, PHP 5.3.3 (Oracle as php-pear-module)
Apache 2.2.3, PHP 5.2.0 (Oracle compiled with php)
Apache 2.2.3, PHP 5.2.0 (Oracle as php-pear-module)
and also on many older versions, but this installations has already been
upgraded.

the error message is "PHP Warning: ocilogon() [function.ocilogon]:
OCIEnvNlsCreate() failed. There is something wrong with your system - please
check that LD_LIBRARY_PATH includes the directory with Oracle Instant Client
libraries"

when you take a look at google, you find a lot of reports of this issue.

e.g.
http://www.eckford.ca/OpenSourceDev/subjects/UbuntuServer/php-oci8.php
https://bugs.php.net/bug.php?id=36355

but if you find a solution on the sites on google, everyone is telling
something different... I would say there is no solution for this issue, just
after changing something and restarting the mashine the issue is gone for a
long time, but it is still there.

Since i've seen this issue on many mashines i tested a lot of things, about
setting the "LD_LIBRARY_PATH" in all possible files; but this issue is happing
about 1 time / year.

Now i've got one mashine where it is happening about once a day.
i don't see any difference to the other mashines, just that it is a virtual
mashine.

When this problem is happening, it is not happening for every client; just
"randomly". because this is happening that often i now know it it just happing
because of ONE (or maybe also 2 at the same time) Apache childs.

when i do a phpinfo() for this childs where is NO difference.

but the processes itself are different:

USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
root     20847  0.0  0.5  55496 11572 ?        Ss   09:16   0:00
/usr/sbin/apach
www-data 20855  0.0  0.6 128600 13888 ?        Sl   09:16   0:00
/usr/sbin/apach
www-data 20856  0.0  0.6 127764 13748 ?        Sl   09:16   0:00
/usr/sbin/apach
www-data 20857  0.0  0.6 128520 14640 ?        Sl   09:16   0:00
/usr/sbin/apach
www-data 20859  0.0  0.5  58228 11128 ?        S    09:16   0:00
/usr/sbin/apach
www-data 20875  0.1  0.7 129832 16216 ?        Sl   09:17   0:00
/usr/sbin/apach

here the issue was with PID 20859.
as you see this PID is using less memory (VSZ + RSS) and also is not
multi-threaded (STAT is "S" instead of "Sl").

so i believe this i an apache issue.

the main question is: how to find this issue; what to test when it's happening
again etc.

BR johannes

-- 
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


Mime
View raw message