Wireshark-dev: Re: [Wireshark-dev] Npcap 0.03 call for test
From: Tyson Key <tyson.key@xxxxxxxxx>
Date: Sat, 1 Aug 2015 14:30:11 +0100
Hi Yang,
Thanks for looking at this. I've just enabled full memory dumps, after reading https://support.microsoft.com/en-us/kb/969028 - but I'll need to do the Right Ctrl + Scroll Lock X2 trick at a time when I can afford to lose state data.
I've got the Windows SDK installed (but not the WinDBG?), if I remember correctly - but I'll install the symbols, and WDK, when I get time. In the meantime, since I've got a %SystemRoot%\MEMORY.DMP file from some crash, but don't know how big it is (since I now have 6GB of RAM, and probably only had 4GB, when it was generated - assuming that "automatic" dumps are "full" dumps), I guess that I've got something to practice post-mortem on...
Tyson.
2015-08-01 4:18 GMT+01:00 Yang Luo <hsluoyb@xxxxxxxxx>:
Hi Tyson,I think I have reproduced the BAD_POOL_CALLER error, the step is: 1) reboot the system, 2) start Wireshark UI, 3) Open VMware Workstation. As you installed VMware Player, maybe it's the same reason. I will look into this later. And I found that a full dump file (memory.dmp) has more useful information (the error position in Npcap driver) than a minidump, so if you can provide full dumps, it will be better. A simpler way is you open the full dump file by yourself using WinDBG (with suitable symbols) and type in "!analyze -v", and then paste the output in this thread, so you don't need to upload such a big dump file.Get WinDBG:Get Windows symbols:Cheers,YangOn Tue, Jul 28, 2015 at 11:09 PM, Tyson Key <tyson.key@xxxxxxxxx> wrote:Aah, I had a look at "Programs, and Features", and it says that the AppEx thing is "AMD Quick Stream" 3.4.4.0, published by AppEx Networks, of Beijing (http://www.appexnetworks.com.cn/). I found a marketing document regarding it at http://support.amd.com/en-us/kb-articles/Pages/AMDQuickStreamTechnology.aspx.Tyson.2015-07-28 16:03 GMT+01:00 Tyson Key <tyson.key@xxxxxxxxx>:Hi Yang,Thanks for looking at these dumps.Yup, I think I enabled the verifier, a few months ago, whilst trying to debug some other issue (probably related to the AppEx thing), and I forgot that I kept it enabled.As for the dumpcap arguments, I just let Wireshark invoke it, through the GUI - so the arguments are whatever it spits out by default, to set up various pipes. I'd have to surgically remove NPCap, and replace it with regular WinPCap, and then try to trace Wireshark Qt/GTK, to learn the arguments (or see if "tasklist /V", or some other utility reveals them). I'd expect that they'd look similar to the ones issued under Linux, modulo device names, though.I'm kinda surprised that Asset is responsible for some of the crashes, to be honest. Sure, it does funny things with multicasting, as a UPnP server implementation, but it's usually pretty reliable, in general operation. Might be worth me reporting a bug to Illustrate, when I get chance; and I'll see what happens if I uninstall it, in the meantime.As for AppEx, I'm pretty sure that I removed its driver from all of my interfaces, but I wouldn't be surprised if there's not something vestigial. Going to see if I can fully cleanse it from my system, since it was an OEM-supplied product, and not something that I opted to install. (And I've had BSoDs from it before, whilst trying to diagnose some WLAN problems). I think it's supposed to be some sort of "game/multimedia quality-of-service optimisation" tool.Take care,Tyson.2015-07-28 12:41 GMT+01:00 Yang Luo <hsluoyb@xxxxxxxxx>:Hi Tyson,I have analyzed the five dumps you provided:1) 072715-32078-01.dmpThis dump is caused by nt!VerifierBugCheckIfAppropriate+0x3c code from process svchost.exe, and it seems to be that you switched on Verifier function for your system. I think there's no relationship with Npcap.2) 072715-31968-01.dmp and 072715-32468-01.dmpthis dump provides BSoD about SYSTEM_SERVICE_EXCEPTION. It is caused by ndis!NdisFOidRequest+62 code from process dumpcap.exe. As Npcap uses NdisFOidRequest calls, I think it's possibly a bug. I'd like to know how you used dumpcap.exe, like parameters?3) 072715-33859-01.dmp and 072715-48062-01.dmpIt is caused by Asset-uPNP.exe, from Asset audio server software provided by illustrate. I think maybe you would like to disable or uninstall it first, to see if the fault still happens. WinDbg also reports that OVERLAPPED_MODULE: Address regions for 'nwifi' and 'appexDrv.sys' overlap. 'appexDrv.sys''s description is " "AppEx Accelerator LWF/WFP Driver L.E."". nwifi.sys seems to be a Microsoft built-in component, and AppEx Networks Accelerator seems to be a VPN software, unfortunately, I didn't find a download link. But this is maybe not the main cause, whatever you can try to shutdown it to see if there's any change.072715-48062-01.dmp's report is pasted here:******************************************************************************** ** Bugcheck Analysis ** ********************************************************************************Use !analyze -v to get detailed debugging information.BugCheck C2, {7, 1200, 0, ffffe0008d01cbf8}fffff80059152240: Unable to get special pool infofffff80059152240: Unable to get special pool infounable to get nt!MmPoolCodeStartunable to get nt!MmPoolCodeEndProbably caused by : NETIO.SYS ( NETIO!NetioCompleteCloneNetBufferListChain+1508d )Followup: MachineOwner---------0: kd> !analyze -v******************************************************************************** ** Bugcheck Analysis ** ********************************************************************************BAD_POOL_CALLER (c2)The current thread is making a bad pool request. Typically this is at a bad IRQL level or double freeing the same allocation, etc.Arguments:Arg1: 0000000000000007, Attempt to free pool which was already freedArg2: 0000000000001200, (reserved)Arg3: 0000000000000000, Memory contents of the pool blockArg4: ffffe0008d01cbf8, Address of the block of pool being deallocatedDebugging Details:------------------OVERLAPPED_MODULE: Address regions for 'nwifi' and 'appexDrv.sys' overlapPOOL_ADDRESS: ffffe0008d01cbf8FREED_POOL_TAG: NDndBUGCHECK_STR: 0xc2_7_NDndCUSTOMER_CRASH_COUNT: 1DEFAULT_BUCKET_ID: VISTA_DRIVER_FAULTPROCESS_NAME: Asset-uPNP.exeCURRENT_IRQL: 2LAST_CONTROL_TRANSFER: from fffff8005912fff2 to fffff80058fdbca0STACK_TEXT:ffffd000`27118f88 fffff800`5912fff2 : 00000000`000000c2 00000000`00000007 00000000`00001200 00000000`00000000 : nt!KeBugCheckExffffd000`27118f90 fffff800`3763083d : 00000000`00000000 ffffe000`8d596040 000008fe`00000010 00000014`00000000 : nt!ExAllocatePoolWithTag+0x1102ffffd000`27119080 fffff800`376023f1 : 00000000`00000000 ffffe000`8ceb3740 00000000`00000000 00000000`00000000 : NETIO!NetioCompleteCloneNetBufferListChain+0x1508dffffd000`271190f0 00000000`00000000 : 00000000`00000000 00000000`00000000 00000000`00000000 00000000`00000000 : NETIO!NetioDereferenceNetBufferListChain+0x2d1STACK_COMMAND: kbFOLLOWUP_IP:NETIO!NetioCompleteCloneNetBufferListChain+1508dfffff800`3763083d 90 nopSYMBOL_STACK_INDEX: 2SYMBOL_NAME: NETIO!NetioCompleteCloneNetBufferListChain+1508dFOLLOWUP_NAME: MachineOwnerMODULE_NAME: NETIOIMAGE_NAME: NETIO.SYSDEBUG_FLR_IMAGE_TIMESTAMP: 540ebbe6FAILURE_BUCKET_ID: X64_0xc2_7_NDnd_NETIO!NetioCompleteCloneNetBufferListChain+1508dBUCKET_ID: X64_0xc2_7_NDnd_NETIO!NetioCompleteCloneNetBufferListChain+1508dFollowup: MachineOwner---------On Tue, Jul 28, 2015 at 3:12 PM, Tyson Key <tyson.key@xxxxxxxxx> wrote:I just uploaded my MiniDumps to https://dl.dropboxusercontent.com/u/670345/MiniDump.rar, if it makes debugging this easier.Tyson.2015-07-28 8:08 GMT+01:00 Tyson Key <tyson.key@xxxxxxxxx>:Hi Yang,Thanks for looking into this.I can't remember when/how I installed Win10PCap (guessing that I briefly had a look, but couldn't get it to do anything on my machine, and just removed it), but I'm using VMware Player 6.0.7 build-2844087 (haven't got Workstation/Server installed); and I tried a dance of upgrading/downgrading/upgrading my AR9485WB-EG WLAN driver (first by downloading the package from http://support.lenovo.com/us/en/downloads/ds032333, to take me from 10.0.0.242, to 10.0.0.75; and then using Device Manager's driver update function, to take me to 3.0.1.155 (which I'm guessing is probably older than 242 - I'm just guessing from the sketchy build dates) - which gave me a different type of BSoD, initially, after starting Wireshark, but let me capture traffic for a little while, after rebooting.Here's all of the MiniDump summaries that I could find:==================================================Dump File : 072715-31968-01.dmpCrash Time : 27/07/2015 07:02:32 pmBug Check String : SYSTEM_SERVICE_EXCEPTIONBug Check Code : 0x0000003bParameter 1 : 00000000`c0000005Parameter 2 : fffff801`1be5d485Parameter 3 : ffffd000`2324e980Parameter 4 : 00000000`00000000Caused By Driver : ntoskrnl.exeCaused By Address : ntoskrnl.exe+150ca0File Description : NT Kernel & SystemProduct Name : Microsoft® Windows® Operating SystemCompany : Microsoft CorporationFile Version : 6.3.9600.17736 (winblue_r9.150322-1500)Processor : x64Crash Address : ntoskrnl.exe+150ca0Stack Address 1 :Stack Address 2 :Stack Address 3 :Computer Name :Full Path : C:\WINDOWS\Minidump\072715-31968-01.dmpProcessors Count : 4Major Version : 15Minor Version : 9600Dump File Size : 281,520Dump File Time : 27/07/2015 07:03:33 pm====================================================================================================Dump File : 072715-32078-01.dmpCrash Time : 27/07/2015 06:47:01 pmBug Check String : BAD_POOL_CALLERBug Check Code : 0x000000c2Parameter 1 : 00000000`00000099Parameter 2 : ffffe000`7d4b31b8Parameter 3 : 00000000`00000000Parameter 4 : 00000000`00000000Caused By Driver : tcpip.sysCaused By Address : tcpip.sys+42856File Description : TCP/IP DriverProduct Name : Microsoft® Windows® Operating SystemCompany : Microsoft CorporationFile Version : 6.3.9600.16384 (winblue_rtm.130821-1623)Processor : x64Crash Address : ntoskrnl.exe+150ca0Stack Address 1 :Stack Address 2 :Stack Address 3 :Computer Name :Full Path : C:\WINDOWS\Minidump\072715-32078-01.dmpProcessors Count : 4Major Version : 15Minor Version : 9600Dump File Size : 281,520Dump File Time : 27/07/2015 06:48:04 pm====================================================================================================Dump File : 072715-32468-01.dmpCrash Time : 27/07/2015 06:34:37 pmBug Check String : SYSTEM_SERVICE_EXCEPTIONBug Check Code : 0x0000003bParameter 1 : 00000000`c0000005Parameter 2 : fffff801`962a446eParameter 3 : ffffd001`1bd0f980Parameter 4 : 00000000`00000000Caused By Driver : ndis.sysCaused By Address : ndis.sys+546eFile Description : Network Driver Interface Specification (NDIS)Product Name : Microsoft® Windows® Operating SystemCompany : Microsoft CorporationFile Version : 6.3.9600.16384 (winblue_rtm.130821-1623)Processor : x64Crash Address : ntoskrnl.exe+150ca0Stack Address 1 :Stack Address 2 :Stack Address 3 :Computer Name :Full Path : C:\WINDOWS\Minidump\072715-32468-01.dmpProcessors Count : 4Major Version : 15Minor Version : 9600Dump File Size : 281,520Dump File Time : 27/07/2015 06:35:48 pm====================================================================================================Dump File : 072715-33859-01.dmpCrash Time : 27/07/2015 05:11:25 pmBug Check String : BAD_POOL_CALLERBug Check Code : 0x000000c2Parameter 1 : 00000000`00000007Parameter 2 : 00000000`00001200Parameter 3 : 00000000`00000000Parameter 4 : ffffe000`8d01cbf8Caused By Driver : ntoskrnl.exeCaused By Address : ntoskrnl.exe+150ca0File Description : NT Kernel & SystemProduct Name : Microsoft® Windows® Operating SystemCompany : Microsoft CorporationFile Version : 6.3.9600.17736 (winblue_r9.150322-1500)Processor : x64Crash Address : ntoskrnl.exe+150ca0Stack Address 1 :Stack Address 2 :Stack Address 3 :Computer Name :Full Path : C:\WINDOWS\Minidump\072715-33859-01.dmpProcessors Count : 4Major Version : 15Minor Version : 9600Dump File Size : 281,520Dump File Time : 27/07/2015 05:12:34 pm====================================================================================================Dump File : 072715-48062-01.dmpCrash Time : 27/07/2015 05:00:25 pmBug Check String : BAD_POOL_CALLERBug Check Code : 0x000000c2Parameter 1 : 00000000`00000007Parameter 2 : 00000000`00001200Parameter 3 : 00000000`00000000Parameter 4 : ffffe000`4bc1b4c8Caused By Driver : ntoskrnl.exeCaused By Address : ntoskrnl.exe+150ca0File Description : NT Kernel & SystemProduct Name : Microsoft® Windows® Operating SystemCompany : Microsoft CorporationFile Version : 6.3.9600.17736 (winblue_r9.150322-1500)Processor : x64Crash Address : ntoskrnl.exe+150ca0Stack Address 1 :Stack Address 2 :Stack Address 3 :Computer Name :Full Path : C:\WINDOWS\Minidump\072715-48062-01.dmpProcessors Count : 4Major Version : 15Minor Version : 9600Dump File Size : 281,520Dump File Time : 27/07/2015 05:01:58 pm==================================================Frustratingly, since there are so many variables involved (unscientific method!), it seems like I'm playing a Jenga game with trying to make this work, since if I remove, or change something, it works for a little while, and then crashes in a creative, new way. (And I don't want to reinstall everything, since I don't have a disk big enough to back everything up). :(I've uploaded a copy of the Nurago Web Meter to https://dl.dropboxusercontent.com/u/670345/nurago%20web%20meter.exe, and I seem to also have an older installer for it in my "Downloads" directory, which may exercise the LSP architecture of WinSock differently.The SYSTEM_SERVICE_EXCEPTION error is interesting, as it is one of the few that reveals a problem in WinSock/NDIS...I would try it in a virtual machine - but it wouldn't get us any closer to diagnosing why it fails to work, with my not-so-unique configuration.Tyson.2015-07-28 7:27 GMT+01:00 Yang Luo <hsluoyb@xxxxxxxxx>:___________________________________________________________________________On Mon, Jul 27, 2015 at 10:42 PM, Tyson Key <tyson.key@xxxxxxxxx> wrote:After rebooting from uninstalling MS NetMon, I restarted Wireshark, and got the usual "NPF service not running; no interfaces available" note. This persists, even if I try "NPFInstall -r", and Wireshark still claims that no interfaces are available."NPFInstall -r" isn't used in Npcap. "NPF service not running; no interfaces available" is a common problem for Npcap previous versions. And I think it should disappear if you have uninstalled previous versions totally.Eventually, after uninstalling NPCap, removing all of the loopback interfaces, and running CCleaner to remove any residual registry data, and then rebooting yet again, I could start Wireshark, and list the installed interfaces - but unsurprisingly, a few moments later, I received another BSoD.If it helps, my Wireshark version is:Version 1.99.8-492-g3f0f49d (v1.99.8rc0-492-g3f0f49d from master)Copyright 1998-2015 Gerald Combs <gerald@xxxxxxxxxxxxx> and contributors.License GPLv2+: GNU GPL version 2 or later <http://www.gnu.org/licenses/old-licenses/gpl-2.0.html>This is free software; see the source for copying conditions. There is NOwarranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.Compiled (64-bit) with GTK+ 2.24.23, with Cairo 1.12.16, with Pango 1.36.8, withWinPcap (unknown), with libz 1.2.8, with GLib 2.42.0, with SMI 0.4.8, withc-ares 1.9.1, with Lua 5.2, with GnuTLS 3.2.15, with Gcrypt 1.6.2, with MITKerberos, with GeoIP, with PortAudio V19-devel (built Jul 22 2015), withAirPcap.Running on 64-bit Windows 8.1, build 9600, with locale English_UnitedKingdom.1252, with Npcap version 0.01 (packet.dll version 0.03), based onWinPcap version 4.1.3 (packet.dll version 4.1.0.3001), based on libpcap version1.0 branch 1_0_rel0b (20091008), with GnuTLS 3.2.15, with Gcrypt 1.6.2, withoutAirPcap.AMD A6-5200 APU with Radeon(TM) HD Graphics (with SSE4.2), with 5577MB ofphysical memory.Built using Microsoft Visual C++ 12.0 build 31101Wireshark is Open Source Software released under the GNU General Public License.Check the man page and http://www.wireshark.org for more information.I used Wireshark latest stable version: Version 1.12.6 (v1.12.6-0-gee1fce6 from master-1.12). But I don't think it makes a difference by using stable version or development version, as its WinPcap related low-level code rarely changed between these two versions.Other than NetMon (which I've removed), the only other things that I think could be causing a conflict are either the VMware host-only networking filters; the networking components included with whatever Bluetooth stack Lenovo shipped; the massive pile of hacks installed by the Gacela component of "Nurago Web Meter", or my Atheros WLAN drivers (which caused Acrylic Wi-Fi's NDIS filters to crash, when I briefly had that installed, a while ago).What version VMware are you using? Workstation or just Player? I used VMware Workstation 11.1.2 build-2780323 on my host, but I didn't install it on my test VM yet.Cheers,Yang
Sent via: Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives: https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe--http://vmlemon.wordpress.com | Twitter/FriendFeed/Skype: vmlemon | 00447934365844Fight Internet Censorship! http://www.eff.org--Fight Internet Censorship! http://www.eff.org
http://vmlemon.wordpress.com | Twitter/FriendFeed/Skype: vmlemon | 00447934365844
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives: https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives: https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe--Fight Internet Censorship! http://www.eff.org
http://vmlemon.wordpress.com | Twitter/FriendFeed/Skype: vmlemon | 00447934365844--Fight Internet Censorship! http://www.eff.org
http://vmlemon.wordpress.com | Twitter/FriendFeed/Skype: vmlemon | 00447934365844
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives: https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe
___________________________________________________________________________
Sent via: Wireshark-dev mailing list <wireshark-dev@xxxxxxxxxxxxx>
Archives: https://www.wireshark.org/lists/wireshark-dev
Unsubscribe: https://wireshark.org/mailman/options/wireshark-dev
mailto:wireshark-dev-request@xxxxxxxxxxxxx?subject=unsubscribe
Fight Internet Censorship! http://www.eff.org
http://vmlemon.wordpress.com | Twitter/FriendFeed/Skype: vmlemon | 00447934365844
http://vmlemon.wordpress.com | Twitter/FriendFeed/Skype: vmlemon | 00447934365844
- Follow-Ups:
- Re: [Wireshark-dev] Npcap 0.03 call for test
- From: Tyson Key
- Re: [Wireshark-dev] Npcap 0.03 call for test
- References:
- Re: [Wireshark-dev] Npcap 0.03 call for test
- From: Yang Luo
- Re: [Wireshark-dev] Npcap 0.03 call for test
- Prev by Date: Re: [Wireshark-dev] Npcap 0.03 call for test
- Next by Date: Re: [Wireshark-dev] Npcap 0.03 call for test
- Previous by thread: Re: [Wireshark-dev] Npcap 0.03 call for test
- Next by thread: Re: [Wireshark-dev] Npcap 0.03 call for test
- Index(es):