Bug 1609607

steps are same with comment#0;

packages info:
qemu-kvm-rhev-2.12.0-8.el7.x86_64
kernel-3.10.0-926.el7.x86_64
virtio-win-prewhql-157
seabios-bin-1.11.0-2.el7.noarch

Can reproduce in virtio-win-prewhql-151 and virtio-win-prewhql-144.
So it's not a regression.

Created attachment 1471409 [details]
hck log

Why this bug is a clone?

According to https://bugzilla.redhat.com/show_bug.cgi?id=1248873#c9 , #1248873 was fixed.

Are you should it is not a regression, did you compare to latest released version?

>>>>
Hello Yan,

Rerun the test case NDIS6.5 - InvalidPackets passed with multiqueue on build 111.  -win8.1_32 

Rerun the test case NDIS6.5 - InvalidPackets passed with multiqueue on build 111.  -win10_64 


Best Regards!
Peixiu Hou
>>>>

Created attachment 1471419 [details]
I am confused. Attached log shows that Invalid Packets test passes...

I am confused. Attached log shows that Invalid Packets test passes...

(In reply to Yan Vugenfirer from comment #7)
> Created attachment 1471419 [details]
> I am confused. Attached log shows that Invalid Packets test passes...
> 
> I am confused. Attached log shows that Invalid Packets test passes...

That's because I run three times,two failed with mq and one pass without mq.

Created attachment 1471423 [details]
all logs.

Could you see all logs?

(In reply to xiagao from comment #9)
> Created attachment 1471423 [details]
> all logs.
> 
> Could you see all logs?

I don't have failed results in the attached package.

Created attachment 1471462 [details]
update another hck log

I update one again, pls have a check.

Test result:
build 111 with mq,hit this issue.
build 145(released version of rhel7.5) with mq,hit it.
build 157 in win8.1-32 with mq,pass
build 157 without mq,pass

(In reply to xiagao from comment #11)
> Created attachment 1471462 [details]
> update another hck log
> 
> I update one again, pls have a check.
> 
> Test result:
> build 111 with mq,hit this issue.
> build 145(released version of rhel7.5) with mq,hit it.
> build 157 in win8.1-32 with mq,pass
> build 157 without mq,pass

I disabled netfilter on the bridge already in the test host.

# sysctl -a | grep bridge
net.bridge.bridge-nf-call-arptables = 0
net.bridge.bridge-nf-call-ip6tables = 0
net.bridge.bridge-nf-call-iptables = 0
net.bridge.bridge-nf-filter-pppoe-tagged = 0
net.bridge.bridge-nf-filter-vlan-tagged = 0
net.bridge.bridge-nf-pass-vlan-input-dev = 0

Current failure:

Error 7/30/2018 11:46:41.670 AM 8 total breakpoints were hit in the protocol driver while this test was executing 
File:    Line: 0 
Error Type:   WIN32 
Error Code:   0x88888 
Error Text:   Error 0x00088888 
Message 7/30/2018 11:46:41.680 AM NDISTest Protocol driver breakpoints are usually hit when a serious error is detected. Please investigate and fix the issue causing the break to solve this failure. 
Message 7/30/2018 11:46:41.683 AM There were 2 unique breakpoints hit: 
Message 7/30/2018 11:46:41.687 AM Breakpoint Summary
NDISTest Source File	Line	Hit Count	Message	
ndistestheadermodule.cpp	696	20	CNDTNDISTestHeaderModule::OnPacketListsIn: Dropped indications, Expected 1289 but received 1483 
	
ndistestheadermodule.cpp	671	8	CNDTNDISTestHeaderModule::OnPacketListsIn: Out of order indication, Expected 29 but received 20

(In reply to xiagao from comment #11)
> Created attachment 1471462 [details]
> update another hck log
> 
> I update one again, pls have a check.
> 
> Test result:
> build 111 with mq,hit this issue.
> build 145(released version of rhel7.5) with mq,hit it.
> build 157 in win8.1-32 with mq,pass
> build 157 without mq,pass

I am very confused. According to the comment above in build 157 the test passes with no issues. When does it fail? 

What about build 159?

(In reply to Sameeh Jubran from comment #14)
> (In reply to xiagao from comment #11)
> > Created attachment 1471462 [details]
> > update another hck log
> > 
> > I update one again, pls have a check.
> > 
> > Test result:
> > build 111 with mq,hit this issue.
> > build 145(released version of rhel7.5) with mq,hit it.
> > build 157 in win8.1-32 with mq,pass
> > build 157 without mq,pass
> 
> I am very confused. According to the comment above in build 157 the test
> passes with no issues. When does it fail? 

The point is for win8-32 without mq,the result is pass. With mq,the result is fail.

> 
> What about build 159?
Hit the same issue.

But, if I set all network card's txqueuelen value  with 8196 on host, I test five times, failed once.

(In reply to xiagao from comment #15)
> (In reply to Sameeh Jubran from comment #14)
> > (In reply to xiagao from comment #11)
> > > Created attachment 1471462 [details]
> > > update another hck log
> > > 
> > > I update one again, pls have a check.
> > > 
> > > Test result:
> > > build 111 with mq,hit this issue.
> > > build 145(released version of rhel7.5) with mq,hit it.
> > > build 157 in win8.1-32 with mq,pass
> > > build 157 without mq,pass
> > 
> > I am very confused. According to the comment above in build 157 the test
> > passes with no issues. When does it fail? 
> 
> The point is for win8-32 without mq,the result is pass. With mq,the result
> is fail.
> 
> > 
> > What about build 159?
> Hit the same issue.
> 
> But, if I set all network card's txqueuelen value  with 8196 on host, I test
> five times, failed once.

Can you check statistics on the host: packets rx\tx\dropped and compare it to the test statistics (the test should show how much packets it sent)?

(In reply to Yan Vugenfirer from comment #16)
> (In reply to xiagao from comment #15)
> > (In reply to Sameeh Jubran from comment #14)
> > > (In reply to xiagao from comment #11)
> > > > Created attachment 1471462 [details]
> > > > update another hck log
> > > > 
> > > > I update one again, pls have a check.
> > > > 
> > > > Test result:
> > > > build 111 with mq,hit this issue.
> > > > build 145(released version of rhel7.5) with mq,hit it.
> > > > build 157 in win8.1-32 with mq,pass
> > > > build 157 without mq,pass
> > > 
> > > I am very confused. According to the comment above in build 157 the test
> > > passes with no issues. When does it fail? 
> > 
> > The point is for win8-32 without mq,the result is pass. With mq,the result
> > is fail.
> > 
> > > 
> > > What about build 159?
> > Hit the same issue.
> > 
> > But, if I set all network card's txqueuelen value  with 8196 on host, I test
> > five times, failed once.
> 
> Can you check statistics on the host: packets rx\tx\dropped and compare it
> to the test statistics (the test should show how much packets it sent)?

As beaker host is reinstalled, I installed two new guests in it to reproduce this issue with build 157/159,but didn't reproduce it. :(
Tried 5 times.

(In reply to xiagao from comment #17)
> (In reply to Yan Vugenfirer from comment #16)
> > (In reply to xiagao from comment #15)
> > > (In reply to Sameeh Jubran from comment #14)
> > > > (In reply to xiagao from comment #11)
> > > > > Created attachment 1471462 [details]
> > > > > update another hck log
> > > > > 
> > > > > I update one again, pls have a check.
> > > > > 
> > > > > Test result:
> > > > > build 111 with mq,hit this issue.
> > > > > build 145(released version of rhel7.5) with mq,hit it.
> > > > > build 157 in win8.1-32 with mq,pass
> > > > > build 157 without mq,pass
> > > > 
> > > > I am very confused. According to the comment above in build 157 the test
> > > > passes with no issues. When does it fail? 
> > > 
> > > The point is for win8-32 without mq,the result is pass. With mq,the result
> > > is fail.
> > > 
> > > > 
> > > > What about build 159?
> > > Hit the same issue.
> > > 
> > > But, if I set all network card's txqueuelen value  with 8196 on host, I test
> > > five times, failed once.
> > 
> > Can you check statistics on the host: packets rx\tx\dropped and compare it
> > to the test statistics (the test should show how much packets it sent)?
> 
> As beaker host is reinstalled, I installed two new guests in it to reproduce
> this issue with build 157/159,but didn't reproduce it. :(
> Tried 5 times.

So it doesn't reproduce anymore with the new host?
Which version is the new host?
Did you use a newer qemu version aswel?

(In reply to Sameeh Jubran from comment #18)
> (In reply to xiagao from comment #17)
> > (In reply to Yan Vugenfirer from comment #16)
> > > (In reply to xiagao from comment #15)
> > > > (In reply to Sameeh Jubran from comment #14)
> > > > > (In reply to xiagao from comment #11)
> > > > > > Created attachment 1471462 [details]
> > > > > > update another hck log
> > > > > > 
> > > > > > I update one again, pls have a check.
> > > > > > 
> > > > > > Test result:
> > > > > > build 111 with mq,hit this issue.
> > > > > > build 145(released version of rhel7.5) with mq,hit it.
> > > > > > build 157 in win8.1-32 with mq,pass
> > > > > > build 157 without mq,pass
> > > > > 
> > > > > I am very confused. According to the comment above in build 157 the test
> > > > > passes with no issues. When does it fail? 
> > > > 
> > > > The point is for win8-32 without mq,the result is pass. With mq,the result
> > > > is fail.
> > > > 
> > > > > 
> > > > > What about build 159?
> > > > Hit the same issue.
> > > > 
> > > > But, if I set all network card's txqueuelen value  with 8196 on host, I test
> > > > five times, failed once.
> > > 
> > > Can you check statistics on the host: packets rx\tx\dropped and compare it
> > > to the test statistics (the test should show how much packets it sent)?
> > 
> > As beaker host is reinstalled, I installed two new guests in it to reproduce
> > this issue with build 157/159,but didn't reproduce it. :(
> > Tried 5 times.
> 
> So it doesn't reproduce anymore with the new host?
run seven times, didn't reproduce. 

> Which version is the new host?
KERNEL: 3.10.0-931.el7.x86_64
QEMU: qemu-kvm-rhev-2.12.0-9.el7.x86_64

> Did you use a newer qemu version aswel?
yes,the latest one.

Test it in the new host with build 160.The result is pass.
So close this bug.