JM.Beaubourg Posted March 10, 2009 Report Share Posted March 10, 2009 I noticed that for about two or three weeks my PC keeps crashing (well, as I should... dang!).I run my crash dumps through WinDbg:Microsoft (R) Windows Debugger Version 6.11.0001.402 X86Copyright (c) Microsoft Corporation. All rights reserved.Loading Dump File [C:\WINDOWS\MEMORY.DMP]Kernel Summary Dump File: Only kernel address space is availableSymbol search path is: SRV*C:\Programme\Debugging Tools for Windows (x86)\symbols*http://msdl.microsoft.com/download/symbolsExecutable search path is: Windows XP Kernel Version 2600 (Service Pack 3) MP (2 procs) Free x86 compatibleProduct: WinNt, suite: TerminalServer SingleUserTS PersonalBuilt by: 2600.xpsp_sp3_gdr.080814-1236Machine Name:Kernel base = 0x804d7000 PsLoadedModuleList = 0x8055d720Debug session time: Tue Mar 10 22:37:01.718 2009 (GMT+1)System Uptime: 0 days 0:56:53.312Loading Kernel Symbols............................................................................................................................Loading User SymbolsPEB is paged out (Peb.Ldr = 7ffdd00c). Type ".hh dbgerr001" for detailsLoading unloaded module list..............******************************************************************************** ** Bugcheck Analysis ** ********************************************************************************Use !analyze -v to get detailed debugging information.BugCheck D1, {2c, 2, 1, b6e7d782}Page 23254 not present in the dump file. Type ".hh dbgerr004" for detailsPEB is paged out (Peb.Ldr = 7ffdd00c). Type ".hh dbgerr001" for detailsPEB is paged out (Peb.Ldr = 7ffdd00c). Type ".hh dbgerr001" for detailsProbably caused by : tcpip6.sys ( tcpip6!SendACK+1af )Followup: MachineOwner---------0: kd> !analyze -v******************************************************************************** ** Bugcheck Analysis ** ********************************************************************************DRIVER_IRQL_NOT_LESS_OR_EQUAL (d1)An attempt was made to access a pageable (or completely invalid) address at aninterrupt request level (IRQL) that is too high. This is usuallycaused by drivers using improper addresses.If kernel debugger is available get stack backtrace.Arguments:Arg1: 0000002c, memory referencedArg2: 00000002, IRQLArg3: 00000001, value 0 = read operation, 1 = write operationArg4: b6e7d782, address which referenced memoryDebugging Details:------------------Page 23254 not present in the dump file. Type ".hh dbgerr004" for detailsPEB is paged out (Peb.Ldr = 7ffdd00c). Type ".hh dbgerr001" for detailsPEB is paged out (Peb.Ldr = 7ffdd00c). Type ".hh dbgerr001" for detailsWRITE_ADDRESS: 0000002c CURRENT_IRQL: 2FAULTING_IP: tcpip6!SendACK+1afb6e7d782 f00fc10f lock xadd dword ptr [edi],ecxDEFAULT_BUCKET_ID: DRIVER_FAULTBUGCHECK_STR: 0xD1PROCESS_NAME: utorrent.exeTRAP_FRAME: b663e970 -- (.trap 0xffffffffb663e970)ErrCode = 00000002eax=89795388 ebx=8976e918 ecx=00000001 edx=00000002 esi=89743fe8 edi=0000002ceip=b6e7d782 esp=b663e9e4 ebp=b663ea0c iopl=0 nv up ei pl nz na po nccs=0008 ss=0010 ds=0023 es=0023 fs=0030 gs=0000 efl=00010202tcpip6!SendACK+0x1af:b6e7d782 f00fc10f lock xadd dword ptr [edi],ecx ds:0023:0000002c=????????Resetting default scopeLAST_CONTROL_TRANSFER: from b6e7d782 to 805446f0STACK_TEXT: b663e970 b6e7d782 badb0d00 00000002 096009e3 nt!KiTrap0E+0x238b663ea0c b6e7efe3 8976e900 00000000 00000c1b tcpip6!SendACK+0x1afb663ea28 b6e7bca0 893409b8 897e1aa0 893409f0 tcpip6!ProcessTCBDelayQ+0xedb663ea44 b6e5d6be b663ea68 00000020 89340a74 tcpip6!TdiReceive+0x227b663ea7c b6e5de46 89340900 89340a70 89340980 tcpip6!TCPReceiveData+0x72b663ea98 804ef19f 89f1d030 893409b8 897a1db0 tcpip6!TCPDispatchInternalDeviceControl+0x4cb663eaa8 b6e483a4 00000000 00000008 b663eb1c nt!IopfCallDriver+0x318a087e00 89e9f4f8 00000012 b6e5a000 00037240 afd!AfdFastConnectionReceive+0x25fWARNING: Frame IP not in any known module. Following frames may be wrong.8a087e14 8a087ea8 001c001c e1668298 8067d260 0x89e9f4f88a22dab0 8a4b87e0 b6e85480 00000001 00000000 0x8a087ea88a4bee48 8a22dab0 b6e39fa8 00000001 00000000 0x8a4b87e08a4bee4c b6e39fa8 00000001 00000000 00000000 0x8a22dab08a4bee50 00000000 00000000 00000000 b6e38000 afd!_load_config_used+0x48STACK_COMMAND: kbFOLLOWUP_IP: tcpip6!SendACK+1afb6e7d782 f00fc10f lock xadd dword ptr [edi],ecxSYMBOL_STACK_INDEX: 1SYMBOL_NAME: tcpip6!SendACK+1afFOLLOWUP_NAME: MachineOwnerMODULE_NAME: tcpip6IMAGE_NAME: tcpip6.sysDEBUG_FLR_IMAGE_TIMESTAMP: 485b8fa4FAILURE_BUCKET_ID: 0xD1_tcpip6!SendACK+1afBUCKET_ID: 0xD1_tcpip6!SendACK+1afFollowup: MachineOwner---------Using the minidumps gives the pretty much the same result.I understand (and that is confirmed by my experience that my PC crashed only when µTorrent v1.8.2.14458 is running) that there is some problem between µTorrent and tcpip.sys that keeps it crashing.My PC seams to be clean (I have AVG Free Edition, occasionally run Autoruns and RootkitRevealer from Sysinternals/Microsoft and HijackThis), also ran LSPFix and have not additional Firewall installed, though my router seems to have one (but it's the same one for about six months now).I used LvlLord's Intelligent TCPIP.SYS patcher / EventID 4226 patch to set the maximum concurrent half-open connections to 40. In µTorrent, I set net.max_halfopen to 25 - after several crashes, and I feel it has helped in reducing the number of crashes. The global maximum number of connections is set to 250 (after the last crash to 240), maximum number of connected peers per torrent was 65 (now 60). The maximum number of active torrents is set to 22.I feel that the last crash may be the result of my limiting up- and download to 25 and 5 kB/s, respectively, but I'd like your advice on that. Earlier (i.e. a couple of days ago) I felt that my running eMule and occasionally the FreeDownloadManager at the same time was the reason. It was then that I changed my settings, and to this day it has helped. But this night (GMT+1) it crashed twice - once with multiple programs running, the other time with only µTorrent (and BOINC) running, but both time with limited up/download rates.If anyone has any idea, it's a bit warysome Link to comment Share on other sites More sharing options...
Switeck Posted March 11, 2009 Report Share Posted March 11, 2009 DRIVER_IRQL_NOT_LESS_OR_EQUAL...is OFTEN the result of bad networking drivers/bad network cards crashing.If newer drivers don't help, getting a good replacement card might. Link to comment Share on other sites More sharing options...
JM.Beaubourg Posted March 13, 2009 Author Report Share Posted March 13, 2009 I have an Atheros L1 Gigabit Ethernet 10/100/1000Base-T Controller (if that's what you mean) and it's driver is dated 12 november 2008 (v2.3.7.12). I can't tell if that's a good one or not. I have downloaded the L1WinSetup_v2.4.7.15_WHQL.rar package (Atheros L1 Windows Driver x86/x64) and hope this is the one. If anyone thinks that's not the right one, please let me know. I might install that this weekend to see. Anyway, as long as I follow my following rules, it seems to work for the moment. (Remember, they are my rules, they may or may not work for you in the same situation.)Rules:* Don't run µTorrent with any P2P application together.* Don't run µTorrent with a download manager with multiple connections for faster download.* maximum concurrent half-open connections > net.max_halfopen > maximum number of active torrents* No upload/download limiting (to use the internet connection for other reasons), stop torrents or µTorrent instead.I'll come back to this later, either when I tried the new driver or someone wrote a message in this topic. Link to comment Share on other sites More sharing options...
Switeck Posted March 13, 2009 Report Share Posted March 13, 2009 uTorrent should have upload speed max limits. Download matters far less and is usually unnecessary for all but the slowest symmetric connections. 2nd link in my signature to configure uTorrent...though you may want to limit global and per-torrent connections to 60 or less for testing. Link to comment Share on other sites More sharing options...
JM.Beaubourg Posted March 13, 2009 Author Report Share Posted March 13, 2009 To be honest, I have no idea of our ADSL connection speed. Our ISP doesn't really communicate that. "Up to 24M" or something like that. (And I have no idea, either whether it's 10 or 18 or 24 or something, nor whether it's Mbit or MByte. So I have no idea of the upload maximum. My guess would be that 1 or 1.5 mbit/s would be close enough. Which means, using your second link I'd have reduce the maximum number of active torrents, which would have been with 22 way too high... And I'll set and upload limit.You still think that I should try this driver update? Link to comment Share on other sites More sharing options...
Switeck Posted March 14, 2009 Report Share Posted March 14, 2009 If you're sure it's the right driver and newer than you have, yes.Do speed tests of your line...your upload may be only a fraction of 1 megabit/second.Download max for an ADSL 2+ line is 24 megabit/second. Link to comment Share on other sites More sharing options...
JM.Beaubourg Posted March 14, 2009 Author Report Share Posted March 14, 2009 You were right. I made several tests.So it's only ~320 kb/s. :mad:Also tried a server in Los Angeles, California which resulted in 4594 kb/s for download and 249 kb/s for upload.Will have to adjust my settings.As for the driver: I guess it's the right one (Atheros L1 Gigabit Ethernet vs. Atheros L1) and newer (2.3.7.12 vs. 2.4.7.15). And inside the archive the readme text file says "The driver installation package contains the Atheros L1 Gigabit Ethernet Controller Driver" - so I guess I'm good... Link to comment Share on other sites More sharing options...
Switeck Posted March 14, 2009 Report Share Posted March 14, 2009 My speed chart for 320 kilobit/second upload seems best suited then.If you want, you can set active downloading torrents to 2 or 3 -- since that doesn't change how fast you're uploading per upload slot. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.