Re: [Qemu-discuss] Qemu-COLO bug report

[wenzt]

Hi, Lukas

I execute { 'execute': 'nbd-server-stop' },{ "execute":
"x-colo-lost-heartbeat" } on SVM.
SVM  working well except network.

I also get this error:
address@hidden:colo_compare_main : secondary: unsupported packet in
address@hidden:colo_compare_main : primary: unsupported packet in

IPV4 Package:
0000000 0001 0000 0000 0022 0010 0000 be00 d175
0000010 fcf1 5452 1200 5634 0008 0045 2800 aa1f
0000020 0040 0680 0b93 790a 71db 1275 1ded 2fc0
0000030 5000 9ed8 8837 071e 1b7a 1450 0000 3548
0000040 0000
Maybe two issues are related?


> Hi,
> I noticed the same bug. But failover works successfully for me including
network connectivity. Did you execute "{ "execute":
> "x-colo-lost-heartbeat" }" on failover?

> On Thu, 11 Apr 2019 14:37:27 +0800
> "wenzt" <address@hidden> wrote:

> Hi, chen
>
>
>
> After I set up COLO between PVM and SVM.
>
> I get this ERROR occasionally: qemu-system-x86_64:
> net/colo-compare.c:823: colo_compare_handle_event: Assertion 
> `event_unhandled_count > 0' failed, on PVM side.
>
>
>
> I think assert(event_unhandled_count )  should be protected by 
> event_mtx  in colo_compare_handle_event, because
> bh_schedule(colo_compare_handle_event)
>
> maybe read event_unhandled_count  just before event_unhandled_count++.
>
>
>
> By the way, in my test environment, SVM lose network connections after 
> failover. (Can't ping out or in to the SVM)
>
>
>
>
>
> Thanks,
>
> Zhengtao
>