From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:52366)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <den@openvz.org>) id 1a98Rf-0007ed-9b
	for qemu-devel@nongnu.org; Wed, 16 Dec 2015 04:32:28 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <den@openvz.org>) id 1a98Rb-0000uv-4T
	for qemu-devel@nongnu.org; Wed, 16 Dec 2015 04:32:27 -0500
Received: from mx2.parallels.com ([199.115.105.18]:50071)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <den@openvz.org>) id 1a98Ra-0000up-Up
	for qemu-devel@nongnu.org; Wed, 16 Dec 2015 04:32:23 -0500
References: <1450256449-23779-1-git-send-email-den@openvz.org>
	<56712AF4.9030105@redhat.com>
From: "Denis V. Lunev" <den@openvz.org>
Message-ID: <56712F9D.4000406@openvz.org>
Date: Wed, 16 Dec 2015 12:32:13 +0300
MIME-Version: 1.0
In-Reply-To: <56712AF4.9030105@redhat.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH 1/1] qmp: process system-reset event in
	paused state
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-devel@nongnu.org, Dmitry Andreev <dandreev@virtuozzo.com>, Markus Armbruster <armbru@redhat.com>

On 12/16/2015 12:12 PM, Paolo Bonzini wrote:
>
> On 16/12/2015 10:00, Denis V. Lunev wrote:
>> With pvpanic or HyperV panic devices could be moved into the paused state
>> with ' <on_crash>preserve</on_crash>'. In this state VM reacts only to
>> 'virsh destroy' or 'continue'.
>>
>> 'virsh reset' command is usually used to force guest reset. The expectation
>> of the behavior of this command is that the guest will be force restarted.
>> This is not true at the moment.
> Does "virsh reset" + "virsh continue" work, and if not why?
as far as I can see there is no such command in virsh at all :(

hades ~/src/qemu $ dpkg -l libvirt-bin
Desired=Unknown/Install/Remove/Purge/Hold
| 
Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name           Version      Architecture Description
+++-==============-============-============-=================================
ii  libvirt-bin    1.2.16-2ubun amd64        programs for the libvirt 
library
hades ~/src/qemu $

hades ~/src/qemu $ virsh continue fedora20
error: unknown command: 'continue'
hades ~/src/qemu $

hades ~/src/libvirt $ cfind . | xargs fgrep \"continue\"
./src/conf/nwfilter_conf.c:              "continue");
hades ~/src/libvirt $ cfind . | xargs fgrep \"reset\"
./src/util/virpci.c:            STREQ(ent->d_name, "reset")) {
./src/conf/domain_conf.c:              "reset",
./src/conf/domain_conf.c:            if (STREQ(tmp, "reset")) {
./src/qemu/qemu_monitor_json.c:              "none", "pause", "reset", 
"poweroff", "shutdown", "debug", "inject-nmi");
./src/vmware/vmware_driver.c:        "reset", PROGRAM_SENTINEL, "soft", NULL
./src/access/viraccessperm.c:              "start", "stop", "reset",
./tools/virsh-domain.c:    {.name = "reset",
./tools/virsh-domain.c:    if (vshCommandOptBool(cmd, "reset"))
./tools/virsh-domain.c: * "reset" command
./tools/virsh-domain.c:              N_("reset"),
./tools/virsh-domain.c:    {.name = "reset",
hades ~/src/libvirt $

Do you propose to kludge libvirt and send 'continue' unconditionally
before a 'reset'? We can but there is not much sense not to implement
this is QEMU.

Sending command in 'crashed' state only is a bit racy. The event
moving domain into 'crashed' state could be not processed at
the moment.

Den