View Issue Details
ID | Project | Category | View Status | Date Submitted | Last Update |
---|---|---|---|---|---|
0001851 | unreal | ircd | public | 2004-06-01 12:09 | 2004-06-01 18:00 |
Reporter | Troco | Assigned To | |||
Priority | normal | Severity | crash | Reproducibility | always |
Status | resolved | Resolution | fixed | ||
Platform | win32 | OS | XP Professional | OS Version | 5.1 - 2600 |
Summary | 0001851: Another core | ||||
Description | Exception: Access Violation ----------------- Backup Buffer: psLag-137853656 ----------------- Registers: EAX=0x0140faaf EBX=0x00c645b0 ECX=0x00000000 EDX=0x00000000 ESI=0x009a41a0 EDI=0x77d13c1f EIP=0x0044477a EBP=0x0140f60c ESP=0x0140f5c4 ----------------- Stack Trace: Module: WIRCD #0 src/ircsprintf.c:289: ircvsprintf #1 src/ircsprintf.c:471: ircsprintf #2 src/send.c:1796: sendto_connectnotice #3 src/s_misc.c:451: exit_client #4 src/s_bsd.c:1921: read_message #5 src/ircd.c:1551: SocketLoop Module: MSVCRTD #6 src/ircd.c:1551: beginthread Module: kernel32 #7 src/ircd.c:1551: RegisterWaitForInputIdle | ||||
Tags | No tags attached. | ||||
3rd party modules | |||||
|
Which version? (build id.. from /info, the last number.. eg: '.29') I hate to ask that every time, could you mention it in the future in every of your bugreports? Thanks ;p. |
|
Correction... could you just include the FULL text (and NOT censored/ripping stuff out ;p) of the core file. I hate it you are hiding stuff all the time. |
|
Yeah, we output all the stuff that the file contains, because that's what we need to solve the problem. If you don't want to give it to us, that's your choice, but we will probably be unable to solve it then. |
|
Generated at Tue Jun 01 11:24:19 2004 Windows XP Service Pack 1 (5.1.2600) Unreal3.2[FinWXeOoZEm3] (1.1.1.1.2.38 2004/05/30 23:29:29) ----------------- Memory Information: Physical: (Available:29MB/Total:223MB) Virtual: (Available:2010MB/Total:2047MB) ----------------- Exception: Access Violation ----------------- Backup Buffer: psLag-137853656 ----------------- Registers: EAX=0x0140faaf EBX=0x00c645b0 ECX=0x00000000 EDX=0x00000000 ESI=0x009a41a0 EDI=0x77d13c1f EIP=0x0044477a EBP=0x0140f60c ESP=0x0140f5c4 ----------------- Stack Trace: Module: WIRCD #0 src/ircsprintf.c:289: ircvsprintf #1 src/ircsprintf.c:471: ircsprintf #2 src/send.c:1796: sendto_connectnotice #3 src/s_misc.c:451: exit_client #4 src/s_bsd.c:1921: read_message #5 src/ircd.c:1551: SocketLoop Module: MSVCRTD #6 src/ircd.c:1551: beginthread Module: kernel32 #7 src/ircd.c:1551: RegisterWaitForInputIdle My apologizes. |
|
That helps :). Try upgrading to latest CVS (.41), a couple of NICKIP crashbugs were fixed. |
|
Another more recent, Generated at Tue Jun 01 16:32:40 2004 Windows XP Service Pack 1 (5.1.2600) Unreal3.2[FinWXeOoZEm3] (1.1.1.1.2.39 2004/05/31 21:10:40) ----------------- Memory Information: Physical: (Available:29MB/Total:223MB) Virtual: (Available:2010MB/Total:2047MB) ----------------- Exception: Access Violation ----------------- Backup Buffer: NOTICE Trocotronic :PING 690802371 MIDOLAG ----------------- Registers: EAX=0x0140f681 EBX=0x00c645b0 ECX=0x00000000 EDX=0x00000000 ESI=0x00921190 EDI=0x77d13c1f EIP=0x0044477a EBP=0x0140f5a4 ESP=0x0140f55c ----------------- Stack Trace: Module: WIRCD #0 src/ircsprintf.c:289: ircvsprintf #1 src/s_extra.c:169: ircd_log #2 src/s_misc.c:432: exit_client #3 src/s_bsd.c:1921: read_message #4 src/ircd.c:1532: SocketLoop Module: MSVCRTD #5 src/ircd.c:1532: beginthread Module: kernel32 #6 src/ircd.c:1532: RegisterWaitForInputIdle It's from latest CVS. editada el: 2004-06-01 12:38 |
|
It ocurs always that I use /disconnect (with /quit bla no problems). |
|
Yes, well.. let us know if you still have it with .41 ;P. both nickip and socket error reporting bugs have been fixed in that. *edit* > It's from latest CVS. But... it says .39, latest is .41. Be sure to do a clean compile? I dunno. edited on: 2004-06-01 13:04 |
|
Yes, but if your guide is releaseid I have to tell that I use my own CVS system and it changes id exp (cacthed from changes). In changes file there is: * $Id: Changes,v 1.1.1.1.2.39 2004/05/31 21:10:40 Trocotronic Exp $ ... - And another one, should be fixed now. (last change made). My own CVS system it's built in order to translate entire unreal to spanish. |
|
Right, well.. we can't support that. Please use official unmodified cvs if you report crashbugs (shouldn't be hard to just compile). In this particular case you are "lucky" because my win cvs build from yesterday is .41 -> http://www.vulnscan.org/UnrealIrcd/win_cvs_download.php?f=Unreal3.2-cvs20040531.exe And also check if you still crash without using any 3rd party modules. [Even if it still crashes, debugoutput would be more trustfull...] I cannot reproduce it here, tried /disconnect various times, with and without cloaked host, with +sF and with&without PROTOCTL HCN. I got logging enabled etc. Even linked to another server and services to see if it then crashes, but still nothing. |
|
Well, I've installed your cvs build and I get the "same" core: Generated at Tue Jun 01 18:24:29 2004 Windows XP Service Pack 1 (5.1.2600) Unreal3.2cvs20040531[FinWXeOoZE] (1.1.1.1.2.1.2.1.2.2234.2.41 2004/05/31 18:49:24) ----------------- Memory Information: Physical: (Available:39MB/Total:223MB) Virtual: (Available:2010MB/Total:2047MB) ----------------- Exception: Access Violation ----------------- Backup Buffer: MODE #rallados nt ----------------- Registers: EAX=0x011af68d EBX=0x00f4b500 ECX=0x00000000 EDX=0x00000000 ESI=0x009de960 EDI=0x77d13c1f EIP=0x0045640a EBP=0x011af5f4 ESP=0x011af5ac ----------------- Stack Trace: Module: wircd #0 c:\dev\unreal3.2\src\ircsprintf.c:289: ircvsprintf #1 c:\dev\unreal3.2\src\ircsprintf.c:471: ircsprintf #2 c:\dev\unreal3.2\src\send.c:1796: sendto_connectnotice #3 c:\dev\unreal3.2\src\s_misc.c:451: exit_client #4 c:\dev\unreal3.2\src\s_bsd.c:1921: read_message #5 c:\dev\unreal3.2\src\ircd.c:1551: SocketLoop Module: MSVCR70D #6 c:\dev\unreal3.2\src\ircd.c:1551: beginthread Module: kernel32 #7 c:\dev\unreal3.2\src\ircd.c:1551: RegisterWaitForInputIdle There aren't any services or servers, I'm completly alone. Furthermore, I have two more files (*.mdmp) about 215K and 46M. I'm connecting through SSL. editada el: 2004-06-01 14:27 editada el: 2004-06-01 14:28 |
|
> Well, I've installed your cvs build and I get the "same" core: Yeah, except the line numbers now make more sense ;). > I'm connecting through SSL. That's it :p. I'm now able to crash too... will take a look. edited on: 2004-06-01 15:01 |
|
(void)exit_client(cptr, cptr, &me, ((sockerr = get_sockerr(cptr)) ? STRERROR(sockerr) : "Client exited")); -> sockerr becomes 5 STRERROR() [alias sock_strerror] doesn't recognize this error and returns NULL. I think we should change this to something like "Unknown error %d". Note that '5' is access denied, so there might be even a bug 'below this', but in any case.. returning NULL doesn't feel good :p. So let's dig some deeper... on why it's actually returning 5 (!?)... in get_sockerr(), WSAGetLastError() returns 5... then the SOL_SOCKET/SO_ERROR stuff is done, which says the error is 0... So the value keeps at 5... I just checked with cvs from a few days ago (before our new strerror stuff) and there it doesn't crash and the exit_client comment is 'Input/output error'. Fun eh ;). edited on: 2004-06-01 15:05 |
|
/me hands the bug over to codemastr ;P. |
|
if (error < WSABASEERR) return strerror(error); ??? That would tell it to use the system error, which would be "Input/Output Error" |
|
Yeah, but the errornr we get from get_sockerr() is '5'... which is AFAIK not 'Input/Output Error' but 'Access denied' which is incorrect (see the whole story @ previous bugnote ;p)... Win32 ownz~ |
|
Well, if you use FormatMessage with a 5, you'll get "access denied", if you use strerror(5), you get "Input/Output Error". So, are you sure that by 5 they mean "Access denied" and not "Input/Output Error"? Or are you just guessing? |
|
I was using my memory (see those codes all the time) + m$ visual studio error lookup. I didn't expect error numbers to have different meanings depending on the weather... Anyway, if you think that the strerror() thing fixes it, then go ahead ;) [cannot test right now]. |
|
Ah, that's the problem then, you're not thinking like MS... things don't have to make sense :P Anyway, I added the strerror() in .47, it works for me, but I guess I might as well wait till Troco can confirm it. |
|
Yeah, it seems that works fine :P |
|
Ok, if it happens again, let us know! |
Date Modified | Username | Field | Change |
---|---|---|---|
2004-06-01 12:09 | Troco | New Issue | |
2004-06-01 12:13 | syzop | Note Added: 0006494 | |
2004-06-01 12:17 | syzop | Note Added: 0006495 | |
2004-06-01 12:21 |
|
Note Added: 0006496 | |
2004-06-01 12:29 | Troco | Note Added: 0006499 | |
2004-06-01 12:33 | syzop | Note Added: 0006500 | |
2004-06-01 12:37 | Troco | Note Added: 0006501 | |
2004-06-01 12:38 | Troco | Note Edited: 0006501 | |
2004-06-01 12:53 | Troco | Note Added: 0006502 | |
2004-06-01 13:01 | syzop | Note Added: 0006503 | |
2004-06-01 13:04 | syzop | Note Edited: 0006503 | |
2004-06-01 13:13 | Troco | Note Added: 0006504 | |
2004-06-01 14:14 | syzop | Note Added: 0006506 | |
2004-06-01 14:25 | Troco | Note Added: 0006507 | |
2004-06-01 14:27 | Troco | Note Edited: 0006507 | |
2004-06-01 14:28 | Troco | Note Edited: 0006507 | |
2004-06-01 14:51 | syzop | Note Added: 0006508 | |
2004-06-01 14:53 | syzop | Note Edited: 0006508 | |
2004-06-01 14:58 | syzop | Note Added: 0006509 | |
2004-06-01 15:00 | syzop | Note Edited: 0006509 | |
2004-06-01 15:01 | syzop | Note Edited: 0006508 | |
2004-06-01 15:05 | syzop | Note Edited: 0006509 | |
2004-06-01 15:06 | syzop | Note Added: 0006510 | |
2004-06-01 15:37 |
|
Note Added: 0006511 | |
2004-06-01 16:27 | syzop | Note Added: 0006519 | |
2004-06-01 16:31 |
|
Note Added: 0006520 | |
2004-06-01 17:03 | syzop | Note Added: 0006523 | |
2004-06-01 17:26 |
|
Note Added: 0006524 | |
2004-06-01 17:40 | Troco | Note Added: 0006525 | |
2004-06-01 18:00 |
|
Status | new => resolved |
2004-06-01 18:00 |
|
Resolution | open => fixed |
2004-06-01 18:00 |
|
Assigned To | => codemastr |
2004-06-01 18:00 |
|
Note Added: 0006526 |