Re: 4.0.5RC1 crash...

On Wed, 21 Mar 2001, Dan Kalowsky wrote: >[Fri Mar 16 09:42:37 2001] [notice] SIGHUP received. Attempting to Ehem. Did you try to 'apachectl stop ; apachectl start' ?? I've seen odd things to happen if 'apachectl restart' is used. --Jani >restart >[Fri Mar 16 09:42:38 2001] [notice] Apache/1.3.17 (Unix) mod_perl/1.24 >PHP/4.0.5RC1 configured -- resuming normal operations >zend_hash.c(202) : Freeing 0x083066E4 (20 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(219) : Freeing 0x08173124 (48 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(208) : Actual location (location was relayed) >zend_hash.c(202) : Freeing 0x0835E6E4 (20 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(219) : Freeing 0x081E37A4 (48 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(208) : Actual location (location was relayed) >[Fri Mar 16 17:13:05 2001] [error] [client 172.16.64.5] File does not >exist: /usr/local/www/data/interactive/internet/index.html > > >Yet again even more..... from Friday March 19th.... >zend_hash.c(202) : Freeing 0x081B9E24 (20 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(219) : Freeing 0x081B86A4 (48 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(208) : Actual location (location was relayed) >zend_hash.c(202) : Freeing 0x081A8E24 (20 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(219) : Freeing 0x0819F5A4 (48 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(208) : Actual location (location was relayed) >zend_hash.c(202) : Freeing 0x081BAE24 (20 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(219) : Freeing 0x081A96A4 (48 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(208) : Actual location (location was relayed) >zend_hash.c(202) : Freeing 0x081B8E24 (20 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 212 times >zend_opcode.c(304) : Freeing 0x08205024 (2100 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 1 time >./zend_language_scanner.l(1059) : Freeing 0x0820D824 (39 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 8 times >zend_hash.c(461) : Freeing 0x081DCBA4 (76 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 13 times >./zend_execute.c(1629) : Freeing 0x0824B564 (24 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_variables.c(117) : Actual location (location was relayed) >Last leak repeated 41 times >Last leak repeated 6 times >var.c(475) : Freeing 0x0833AA24 (2 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 15 times >var.c(562) : Freeing 0x0833A9A4 (12 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 16 times >session.c(501) : Freeing 0x082529A4 (48 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_API.c(188) : Actual location (location was relayed) >session.c(500) : Freeing 0x0833BFE4 (12 bytes), >script=/usr/local/www/data/Login/users_login.php >mod_files.c(134) : Freeing 0x0822DFA4 (33 bytes), >script=/usr/local/www/data/Login/users_login.php >mod_files.c(209) : Freeing 0x081BBC64 (5 bytes), >script=/usr/local/www/data/Login/users_login.php >mod_files.c(201) : Freeing 0x08335E24 (16 bytes), >script=/usr/local/www/data/Login/users_login.php >session.c(873) : Freeing 0x081F60A4 (33 bytes), >script=/usr/local/www/data/Login/users_login.php >zend_compile.c(1605) : Freeing 0x083186E4 (12 bytes), >script=/usr/local/www/data/Login/users_login.php >Last leak repeated 3 times >main.c(1059) : Freeing 0x081B9C24 (48 bytes), >script=/usr/local/www/data/home/menu_walled_garden.php > > > >This goes on for some time, so if you want to see it all, you can find >the full version at >http://www.deadmime.org/~dank/sitics-error.log > >But it also leaves an interesting spot in the middle: > >--------------------------------------- >zend_opcode.c(168) : Block 0x081ACAC0 status: >zend_variables.c(62) : Actual location (location was relayed) >Beginning: OK (allocated on ./zend_language_scanner.l:739, 89 >bytes) > End: Overflown (magic=0x283F791C instead of 0x2A8FCC84) > At least 4 bytes overflown >--------------------------------------- > ><..... SOME TIME/BAD-FILE-REQUESTS PASSING .... > > >--------------------------------------- >zend_opcode.c(168) : Block 0x08AE4B80 status: >zend_variables.c(62) : Actual location (location was relayed) >Beginning: OK (allocated on ./zend_language_scanner.l:1152, 217 >bytes) > End: Overflown (magic=0x283F791C instead of 0x2A8FCC84) > At least 4 bytes overflown >--------------------------------------- >[Tue Mar 20 23:50:21 2001] [notice] child pid 257 exit signal >Segmentation fault (11), possible coredump in /usr/local >[Tue Mar 20 23:58:11 2001] [error] [client 172.16.64.5] File does not >exist: /usr/local/www/data/index.html >httpd in free(): warning: chunk is already free. >httpd in free(): warning: chunk is already free. >httpd in free(): warning: page is already free. >httpd in free(): warning: chunk is already free. > > >Now the more interesting part... we were able to actually capture state >in gdb when some of these happened, and have two core dumps for your >parusal now.... > >(gdb) bt >#0 0x282f525c in shutdown_memory_manager (silent=0, clean_cache=0) at >zend_alloc.c:477 >#1 0x28329fc0 in php_request_shutdown (dummy=0x0) at main.c:665 >#2 0x28326c57 in php_apache_request_shutdown (dummy=0x0) at >mod_php4.c:290 >#3 0x805144a in run_cleanups () >#4 0x80503aa in ap_clear_pool () >#5 0x805040c in ap_destroy_pool () >#6 0x805037f in ap_clear_pool () >#7 0x805bc36 in child_main () >#8 0x805c239 in make_child () >#9 0x805c4a8 in perform_idle_server_maintenance () >#10 0x805c905 in standalone_main () >#11 0x805ce3b in main () >#12 0x804fbf1 in _start () > > >The above seems to be the common backtrace for them all, only change the >zend_alloc.c line from 477 to 460 (it seems to be split... 2 on 477 and >2 on 460) > >There is one other lone core dump that I'm not sure what to make of from >this though... >(gdb) bt >#0 0x8050533 in ap_palloc () >#1 0x8050614 in ap_pstrdup () >#2 0x8050d06 in ap_table_set () >#3 0x38080537 in ?? () >Cannot access memory at address 0x2abfbffb. >(gdb) > > >I can more than likely post the users_login.php (which seems to cause >these errors), but it's a bit lengthy... I'm working on trying to >pinpoint an area which causes this series of events (no luck yet). > > >System FreeBSD kalowsky-bsd.eng.tvol.net 4.2-STABLE FreeBSD >4.2-STABLE #2: Fri > Mar 2 09:03:58 EST 2001 > >kalowsky@kalowsky-bsd.eng.tvol.net:/usr/obj/usr/src/sys/DEVEL i386 > >Configure Command >./configure --with-imap --with-gettext --with-apxs=/usr/local/sbin/apxs > --prefix=/usr/local --with-pgsql --enable-debug >--enable-snmp > --enable-track-vars --enable-trans-sid >--disable-magic-quotes > --enable-short-tags --without-mysql --without-tsrm-pthreads > --enable-wddx > > >Apache Version 1.3.17 >Loaded Modules > mod_perl, mod_php4, mod_setenvif, mod_so, mod_unique_id, >mod_usertrack, > mod_headers, mod_expires, mod_cern_meta, mod_proxy, mod_digest, > mod_auth_db, mod_auth_anon, mod_auth, mod_access, mod_rewrite, >mod_alias, > mod_speling, mod_actions, mod_imap, mod_cgi, mod_dir, >mod_include, mod_info, > mod_status, mod_negotiation, mod_mime, mod_mime_magic, >mod_log_config, > mod_env, mod_vhost_alias, http_core > >