Bug 1088815

Created attachment 887121 [details]
SUT guest hang when guest installed windbg at this stage

Hi Mike,

Can you reply comment9?

Best Regards,
Junyi

job failed with net transport.
steps:
1.boot intel-max guest with one virtio-net-pci and one e1000:
/usr/libexec/qemu-kvm --nodefaults --nodefconfig -m 2G -smp 2 -cpu Nehalem,+kvm_pv_unhalt,hv_spinlocks=0x1fff,hv_relaxed,hv_vapic,hv_time -M pc-i440fx-rhel7.0.0 -usb -device usb-tablet,id=tablet0 -drive file=win2012R2-intel-max.raw-new,if=none,id=drive-virtio0-0-0,format=raw,werror=stop,rerror=stop,cache=none,serial=number -device virtio-blk-pci,drive=drive-virtio0-0-0,id=virti0-0-0,bootindex=1 -uuid 80f123e0-6380-48f6-b9cb-049d1a110d2d -monitor stdio -vnc :20 -vga cirrus -name win2012R2-INTEL-MAX-new -global PIIX4_PM.disable_s3=1 -global PIIX4_PM.disable_s4=1 -cdrom /usr/share/virtio-win/virtio-win.iso -fda /usr/share/virtio-win/virtio-win-1.7.0_amd64.vfd -boot menu=on -device usb-ehci,id=ehci0 -drive file=usb-storage-intel-max.raw,if=none,id=drive-usb-2-0,media=disk,format=raw,cache=none,werror=stop,rerror=stop,aio=threads -device usb-storage,bus=ehci0.0,drive=drive-usb-2-0,id=usb-2-0,removable=on -rtc base=localtime,clock=host,driftfix=slew -chardev socket,id=b111a,path=/tmp/monitor-win2012R2-intel-max,server,nowait -mon chardev=b111a,mode=readline -netdev tap,id=hostnet1,vhost=on,script=/etc/qemu-ifup1 -device virtio-net-pci,netdev=hostnet1,id=net1,mac=00:52:52:2a:28:5f,addr=0x6 -netdev tap,id=hostnet0,vhost=on,script=/etc/qemu-ifup1 -device e1000,netdev=hostnet0,id=net0,mac=00:52:52:2a:22:3f,addr=0x7

2.boot debug machine with one e1000:
/usr/libexec/qemu-kvm --nodefaults --nodefconfig -m 4G -smp 4 -cpu Opteron_G3,+kvm_pv_unhalt,hv_relaxed,hv_spinlocks=0x1fff,hv_vapic,hv_time -M pc-i440fx-rhel7.0.0 -usb -device usb-tablet,id=tablet0 -drive file=debug-amd.raw,if=none,id=drive-virtio0-0-0,format=raw,werror=stop,rerror=stop,cache=none,serial=number -device virtio-blk-pci,drive=drive-virtio0-0-0,id=virti0-0-0,bootindex=1 -uuid d04887dd-6a55-4c5f-b9d3-dc76759edd37 -monitor stdio -vnc :6 -vga cirrus -name amd-debug -global PIIX4_PM.disable_s3=1 -global PIIX4_PM.disable_s4=1 -boot menu=on -rtc base=localtime,clock=host,driftfix=slew -chardev socket,id=b111a,path=/tmp/monitor-win2012R2-amd-max,server,nowait -mon chardev=b111a,mode=readline -netdev tap,id=hostnet1,script=/etc/qemu-ifup1 -device e1000,netdev=hostnet1,id=net1,mac=00:52:56:2a:42:2e,addr=0x07

3.submit the job with:
BusParams  0.7.0  
NetHostIP  192.168.1.195  ---> debug machine ip
NetKey  this.key.isnt.secure  
NetPort  50000
Transport  net

Actual result:
Job failed at child job "vefify debug configuration";
error message:System.Exception: Kernel debugging over the network fails due to the non-supported NIC 0.For a list of supported NICs please visit http://go.microsoft.com/fwlink/?LinkID=230754 or run the test from another debug connection. at KernelDebugLogoTestSetup.DebugSetupVerifier.DebugSetupVerify()

Expected result:
job can pass,no error

Note:
boot intel-max guest with two e1000,job also failed.

mount these two images on rhel6 host,then run the job,job can pass easily.

(In reply to lijin from comment #13)
> mount these two images on rhel6 host,then run the job,job can pass easily.

also with net protocol,steps same as comment #12

Yan ,according to comment #13 ,it seems another bug that HCK can not detect emulated e1000 card device ID as 100E ,Do we need to open a new bug to track it ?

rerun this job on rhel7 host with -M rhel6.5.0,job still failed,but has different error message:
System.Exception: Kernel debugging over the network fails due to InitializeNetwork failed to get the ethernet address of the host debugger. error code -1073741643.Please address the problem or run the test from another debug connection. at KernelDebugLogoTestSetup.DebugSetupVerifier.DebugSetupVerify()

e1000 of SUT ***CANNOT*** get ip after the first reboot during job running,after the second reboot,everything get normal,both virtio-net-pci and e1000 can get ip.

I will upload the log file ASAP

Created attachment 888028 [details]
intel-max-debug-capacity-test hck file

Created attachment 888029 [details]
screenshot for comment #16

(In reply to lijin from comment #16)
> rerun this job on rhel7 host with -M rhel6.5.0,job still failed,but has
> different error message:
> System.Exception: Kernel debugging over the network fails due to
> InitializeNetwork failed to get the ethernet address of the host debugger.
> error code -1073741643.Please address the problem or run the test from
> another debug connection. at
> KernelDebugLogoTestSetup.DebugSetupVerifier.DebugSetupVerify()
> 
> e1000 of SUT ***CANNOT*** get ip after the first reboot during job
> running,after the second reboot,everything get normal,both virtio-net-pci
> and e1000 can get ip.

the detailed network info:
on RHEL7 host:
ipconfig of e1000:192.168.1.29/255.255.255.0/192.168.1.1(gw)
after first reboot during job running
ipconfig of e1000:169.254.241.190/255.255.0.0/null
job failed

on RHEL6 host:
ipconfig of e1000:192.168.1.29/255.255.255.0/192.168.1.1(gw)
after first reboot during job running
ipconfig of e1000:192.168.1.29/255.255.255.0/192.168.1.1
job passed

> I will upload the log file ASAP

(In reply to Mike Cao from comment #15)
> Yan ,according to comment #13 ,it seems another bug that HCK can not detect
> emulated e1000 card device ID as 100E ,Do we need to open a new bug to track
> it ?

Mike, Please open a new bug for this, and assign to akong. Thanks.

(In reply to Mike Cao from comment #0)
> 2.It passed w/ -smp 2 & -m 2G ,but not sure whether MSFT will accept ths
> results ,I would like to add testblocker flag first

Mike,
Did you contact MSFT?

After further research the root cause is the hosts in china lab does not permit external access. the workaround is to add _NT_SYMBOL_PROXY and _NT_SYMBOL_PATH to envir variable. it can pass although not 100% 

Close this bug based on above