Apache/Cpanel Problem

Portal Home > Knowledgebase > Articles Database > Apache/Cpanel Problem

Posted by techwebhosting, 04-11-2009, 10:15 PM
Alright, so I am stumped on this one. I have come to the best place that I know for help. Seems like everyday, at least once or twice a day Apache will begin to use up all of the system memory and then kill itself. 95% of the time, it will get killed and restarted automatically. The other 5 percent the server locks up and I have to fix it. I have some free time so now I am going to chase down the problem. I am attaching an output from the error log before the last crash which was just a few seconds ago. Please give me any input. Thanks in advance. *** glibc detected *** /usr/local/apache/bin/httpd: double free or corruption (fasttop): 0x098f6658 *** ======= Backtrace: ========= /lib/libc.so.6[0x5330f1] /lib/libc.so.6(cfree+0x90)[0x536bc0] /usr/local/apache/libexec/libphp4.so(shutdown_memory_manager+0x6e)[0xc4f91e] /usr/local/apache/libexec/libphp4.so(php_request_shutdown+0x2f7)[0xc31ec7] /usr/local/apache/libexec/libphp4.so(apache_php_module_main+0x144)[0xc775f4] /usr/local/apache/libexec/libphp4.so[0xc783cf] /usr/local/apache/bin/httpd(ap_invoke_handler+0x121)[0x808d42f] /usr/local/apache/bin/httpd[0x80a4c31] /usr/local/apache/bin/httpd(ap_process_request+0x33)[0x80a4c8e] /usr/local/apache/bin/httpd[0x809adaf] /usr/local/apache/bin/httpd[0x809b0e0] /usr/local/apache/bin/httpd[0x809b4ce] /usr/local/apache/bin/httpd[0x809bb5d] /usr/local/apache/bin/httpd(main+0x3d2)[0x809c208] /lib/libc.so.6(__libc_start_main+0xdc)[0x4dfe8c] /usr/local/apache/bin/httpd[0x8054ef1] ======= Memory map: ======== 00110000-0013c000 r-xp 00000000 fd:00 30318802 /usr/lib/libgssapi_krb5.so.2.2.#prelink#.icPFPE (deleted) 0013c000-0013d000 rwxp 0002c000 fd:00 30318802 /usr/lib/libgssapi_krb5.so.2.2.#prelink#.icPFPE (deleted) 0013d000-001d0000 r-xp 00000000 fd:00 30316265 /usr/lib/libkrb5.so.3.3 001d0000-001d3000 rwxp 00092000 fd:00 30316265 /usr/lib/libkrb5.so.3.3 001d3000-001e2000 r-xp 00000000 fd:00 11895021 /lib/libresolv-2.5.so 001e2000-001e3000 r-xp 0000e000 fd:00 11895021 /lib/libresolv-2.5.so 001e3000-001e4000 rwxp 0000f000 fd:00 11895021 /lib/libresolv-2.5.so 001e4000-001e6000 rwxp 001e4000 00:00 0 001e6000-001e8000 r-xp 00000000 fd:00 11895036 /lib/libkeyutils-1.2.so 001e8000-001e9000 rwxp 00001000 fd:00 11895036 /lib/libkeyutils-1.2.so 001e9000-0021b000 r-xp 00000000 fd:00 30321378 /usr/lib/libcurl.so.3.0.0 0021b000-0021c000 rwxp 00032000 fd:00 30321378 /usr/lib/libcurl.so.3.0.0 0021c000-00227000 r-xp 00000000 fd:00 11896374 /lib/libgcc_s-4.1.2-20080825.so.1 00227000-00228000 rwxp 0000a000 fd:00 11896374 /lib/libgcc_s-4.1.2-20080825.so.1 Last edited by techwebhosting; 04-11-2009 at 10:22 PM. Reason: Incorrect Output
Posted by Deem3n®, 04-12-2009, 03:07 PM
Check AddModule lines in httpd.conf for conflicting extensions (php4/php5)
Posted by ServerManagement, 04-12-2009, 03:49 PM
That certain error can be numerous different causes. I've seen that with everything from corrupt files, to bad config, to bad hardware, etc. Unless you are familiar with debugging apache, you may want to try reinstalling it from scratch and see how it goes. If you don't want to try that or if that doesn't help, you may want to have someone check it out for you because it's hard to tell you specifically what it is just by that error.
Posted by techwebhosting, 04-12-2009, 05:56 PM
Ok, The first thing I am going to try to do is rebuild apache and double check all of my modules. I will post an update shortly.
Posted by hexahost, 04-13-2009, 02:13 AM
Are you using MPM of any kind ? I tried MPM once, but it created more issues when all our clients were using php and database driven pages. MPM will be good only when you are using static pages. For dynamic pages MPM will be a problem. Gurus here in wht... correct me if I am wrong.
Posted by Quel, 04-13-2009, 03:59 AM
After recompiling, get a firewall (csf maybe?) and mod_secure and check that it is not due to external causes, like some bot browsing your sites Quel
Posted by techwebhosting, 04-13-2009, 08:06 PM
Alright, after the rebuild everything seems to be working ok. I am going to keep my fingers crossed. On a different note now, I get this error when trying to do things with Apache inside the WHM. Any thoughts? Acme: : Spork version v0.0.8 required--this is only version v0.0.7 at /var/cpanel/perl/easy/Cpanel/Easy/Utils/BackGround.pm line 14. BEGIN failed--compilation aborted at /var/cpanel/perl/easy/Cpanel/Easy/Utils/BackGround.pm line 14. Compilation failed in require at (eval 28) line 3. ...propagated at /usr/lib/perl5/5.8.8/base.pm line 85. at (eval 26) line 3 at /var/cpanel/perl/easy/Cpanel/Easy.pm line 15 BEGIN failed--compilation aborted at /var/cpanel/perl/easy/Cpanel/Easy.pm line 15. Compilation failed in require at (eval 16) line 3. ...propagated at /usr/lib/perl5/5.8.8/base.pm line 85. BEGIN failed--compilation aborted at /var/cpanel/perl/easy/Cpanel/Easy/Apache.pm line 8. Compilation failed in require at /usr/local/cpanel/whostmgr/docroot/cgi/easyapache.pl line 67. Premature end of script headers: /usr/local/cpanel/whostmgr/docroot/cgi/easyapache.pl: Please check / usr / local / cpanel / logs / error_log$
Posted by diligent, 04-13-2009, 08:52 PM
Which OS are you running on? Have you tried "/scripts/upcp --force"? When you say you're trying to do things with Apache inside WHM, what specifically are you trying to do?
Posted by techwebhosting, 04-14-2009, 10:58 AM
Running Cent OS 5. Specifically when I try to update apache via WHM, and view its status it reports errors. I am going to try the update first.
Posted by Servosupport, 04-14-2009, 11:22 AM
Hello, Try to set WHM in release version, update it first and then try to run updated apache again. If you do get any errors please do paste the output
Posted by mattchariot, 04-14-2009, 12:33 PM
yes.. servosupport is right run upcp using the release version and try easyapache also check the cpanel error log if you still get the error.
Posted by techwebhosting, 04-14-2009, 04:22 PM
I did that, and I am still having that one issue. The other thing is Apache is still doing the same thing it did before too. It begins to eat up all the memory and then crashes. The other error I was talking about is when you try to click on Apache status in the WHM panel it just says "Failed to receive status information from Apache." That's under Main>>Server Status>> Apache Status. It also does not seem to report any errors in the log which is weird. I know for sure though it is Apache related, everything else is working great and has been.