View Issue Details
| ID | Project | Category | View Status | Date Submitted | Last Update |
|---|---|---|---|---|---|
| 0002527 | unreal | ircd | public | 2005-05-20 16:04 | 2005-05-20 17:43 |
| Reporter | Xpl0it3r | Assigned To | |||
| Priority | normal | Severity | crash | Reproducibility | always |
| Status | closed | Resolution | open | ||
| Product Version | 3.2.3 | ||||
| Summary | 0002527: Crash on Linking | ||||
| Description | Hi, When am linking my newly compiled and perfectly configured Unreal3.2.3 ircds, they are crashing. | ||||
| 3rd party modules | |||||
|
|
do: -- gdb src/ircd name.of.core.file bt -- and paste us the results. |
|
|
Done. naksha@c3p0:/home/naksha/Unreal3.2> ls bla bla file and one is ircd.core. Then ... naksha@c3p0:/home/naksha/Unreal3.2> gdb src/ircd ircd.core GNU gdb 4.18 (FreeBSD) This GDB was configured as "i386-unknown-freebsd"...Deprecated bfd_read called at /staff/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 2627 in elfstab_build_psymtabs Deprecated bfd_read called at /staff/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 933 in fill_symbuf "/home/naksha/Unreal3.2/ircd.core" is not a core dump: File format not recognized (gdb) bt No stack. |
|
|
Hm, that isn't very useful output indeed. Is this on both sides? Apparantly the core file is corrupt or not fully generated. 1. Check your quota (type 'quota'), if that applies. 2. Check your resource limits, type 'ulimit -a'. Also try 'ulimit -c unlimited' and see if that printed out any error. 3. Could you paste the output of 'ls -al ircd.core' and 'file ircd.core' ? 4. Could you let us know what things you enabled: ziplinks? ssl? remote includes? anything else non-default? |
|
|
naksha@c3p0:/home/naksha/Unreal3.2> quota Disk quotas for user naksha (uid 1193): none naksha@c3p0:/home/naksha/Unreal3.2> bash-2.05b$ ulimit -a core file size (blocks, -c) 2048 data seg size (kbytes, -d) 51200 file size (blocks, -f) 102400 max locked memory (kbytes, -l) 51200 max memory size (kbytes, -m) 51200 open files (-n) 512 pipe size (512 bytes, -p) 1 stack size (kbytes, -s) 8192 cpu time (seconds, -t) 21600 max user processes (-u) 32 virtual memory (kbytes, -v) unlimited bash-2.05b$ ulimit -c unlimited bash: ulimit: core file size: cannot modify limit: Operation not permitted bash-2.05b$ ls -al ircd.core -rw------- 1 naksha naksha 0 May 20 16:22 ircd.core bash-2.05b$ file ircd.core ircd.core: empty I have enabled only ziplinks and I added ... options { zip; quarantine; nodnscache; }; |
|
|
I see. Seems your shell disallows dumping corefiles. That complicates things a bit. But since it seems you are saying this crash is reproducable it should still be easy, do the following (ensure the ircd is not already running): - gdb src/ircd handle SIGPIPE nostop noprint run -F - Then when it crashes, you should see gdb saying some stuff and you get a (gdb) prompt again, then do: - bt - Paste all the results of the 'bt' and the few lines output before that. I don't know if this is a live network with users on it, but.. if you don't have "real users" then I suggest keeping the gdb prompt open (your ircd will stay completely stalled but I could then ask you to type some other things to gather more info if needed). Otherwise, just quit the gdb session and restart your ircd again. |
|
|
theres what gdb saying when Unreal crashing ... naksha@c3p0:/home/naksha/Unreal3.2> gdb src/ircd (gdb) handle SIGPIPE nostop noprint Signal Stop Print Pass to program Description SIGPIPE No No Yes Broken pipe (gdb) run -F Starting program: /home/naksha/Unreal3.2/src/ircd -F _ _ _ ___________ _____ _ using TRE 0.7.2 (GPL) using zlib 1.1.4 * Loading IRCd configuration .. * Configuration loaded without any problems .. * Loading tunefile.. * Dynamic configuration initialized .. booting IRCd. --------------------------------------------------------------------- Program received signal SIGSEGV, Segmentation fault. 0x28258a96 in nns_nickchange (ovr=0x8178120, cptr=0x8177800, sptr=0x8177800, parc=12, parv=0x81180e0) at nonickswear.c:97 97 for (chmbr = cptr->user->channel; chmbr; chmbr = chmbr->next) ----- (gdb) bt #0 0x28258a96 in nns_nickchange (ovr=0x8178120, cptr=0x8177800, sptr=0x8177800, parc=12, parv=0x81180e0) at nonickswear.c:97 #1 0x8061331 in parse (cptr=0x8177800, buffer=0x8177938 "&", bufend=0x8177986 "") at parse.c:438 #2 0x80603f8 in dopacket (cptr=0x8177800, buffer=0x8118b20 "xÚl\220ÃJÃ@\024E¡Ë@þáI@ÔÖaæM2I\a*ÕTkѶÚt!-]LÒÔNMÓ\222¤ \e7ý\024¿Ã/rÃ¥Bü\005#HA(\217\013\227Ëåq¸Ò\237«\202t\006þU¦ãtJüÕ\022\202n\037V \217*\211N\237*Ç0êÜÃ\226\016öTOÃà zsñ?\0347\205 \210\016AJ\211ë\022\206Lx\023\210óB\205\211ÎçñÔ4\016áa\235P]ð\f\030\0340\034)\235AKe7\200\224\223:#ÃŒv\tÚ\f8P¨êás{\r'ðÂ6Cï¾Ñ", length=296) at packet.c:138 #3 0x8066547 in read_packet (cptr=0x8177800, rfd=0xbfbff9e0) at s_bsd.c:1447 #4 0x8066e5d in read_message (delay=1, listp=0x81453a0) at s_bsd.c:1937 #5 0x805ca88 in main (argc=2, argv=0xbfbffb18) at ircd.c:1545 (gdb) It seems like one of unofficial modules call nonickswear.c making prob. Should I remove the module? Any other solution than it? |
|
|
Yeah, try with all 3rd party modules unloaded. |
|
|
Thanks. Let me try. |
|
|
Its been fixed. Thank you Syzop :) |
|
|
ok |
| Date Modified | Username | Field | Change |
|---|---|---|---|
| 2005-05-20 16:04 | Xpl0it3r | New Issue | |
| 2005-05-20 16:10 | syzop | Note Added: 0009963 | |
| 2005-05-20 16:20 | Xpl0it3r | Note Added: 0009964 | |
| 2005-05-20 16:34 | syzop | Note Added: 0009965 | |
| 2005-05-20 16:41 | Xpl0it3r | Note Added: 0009966 | |
| 2005-05-20 16:47 | syzop | Note Added: 0009967 | |
| 2005-05-20 16:56 | Xpl0it3r | Note Added: 0009968 | |
| 2005-05-20 17:05 | syzop | Note Added: 0009969 | |
| 2005-05-20 17:09 | Xpl0it3r | Note Added: 0009970 | |
| 2005-05-20 17:34 | Xpl0it3r | Note Added: 0009971 | |
| 2005-05-20 17:43 | syzop | Status | new => closed |
| 2005-05-20 17:43 | syzop | Note Added: 0009972 |
