Bug 2143840

The problem here is not the destroy command but how it's invoked, because 'virsh destroy' does a bit more than just destroying the domain. It also needs to open a connection. And opening a new connection consists of multiple RPC calls. Most of them are marked as high priority (meaning the priority worker can process them), except for one: REMOTE_PROC_CONNECT_REGISTER_CLOSE_CALLBACK. And this is the point where virsh gets stuck (long before it's even able to call virDomainDestroy()). Therefore, if you'd start virsh interactively, then destroy would work just fine. I need to investigate whether aforementioned RPC call can be marked as high priority and what would be the implications.

Patch posted onto the list:

https://listman.redhat.com/archives/libvir-list/2022-November/235887.html

Merged upstream as:

commit a6d3717e7f28bfb574f3122d9b99703a7d722033
Author:     Michal Prívozník <mprivozn>
AuthorDate: Wed Nov 23 09:50:29 2022 +0100
Commit:     Michal Prívozník <mprivozn>
CommitDate: Wed Nov 23 12:13:10 2022 +0100

    rpc: Mark close callback (un-)register as high priority
    
    Our RPC calls can be divided into two groups: regular and high
    priority. The latter can be then processed by so called high
    priority worker threads. This is our way of defeating a
    'deadlock' and allowing some RPCs to be processed even when all
    (regular) worker threads are stuck. For instance: if all regular
    worker threads get stuck when talking to QEMU on monitor, the
    virDomainDestroy() can be processed by a high priority worker
    thread(s) and thus unstuck those threads.
    
    Now, this is all fine, except if users want to use virsh
    non interactively:
    
      virsh destroy $dom
    
    This does a bit more - it needs to open a connection. And that
    consists of multiple RPC calls: AUTH_LIST,
    CONNECT_SUPPORTS_FEATURE, CONNECT_OPEN, and finally
    CONNECT_REGISTER_CLOSE_CALLBACK. All of them are marked as high
    priority except the last one. Therefore, virsh just sits there
    with a partially open connection.
    
    There's one requirement for high priority calls though: they can
    not get stuck. Hopefully, the reason is obvious by now. And
    looking into the server side implementation the
    CONNECT_REGISTER_CLOSE_CALLBACK processing can't ever get stuck.
    The only driver that implements the callback for public API is
    Parallels (vz). And that can't block really.
    
    And for virConnectUnregisterCloseCallback() it's the same story.
    
    Therefore, both can be marked as high priority.
    
    Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2143840
    Signed-off-by: Michal Privoznik <mprivozn>
    Reviewed-by: Daniel P. Berrangé <berrange>

v8.9.0-281-ga6d3717e7f

Tested pass with libvirt-8.10.0-2.el9.x86_64.

Verified with libvirt-8.10.0-2.el9.x86_64.

Test steps:
1.Edit virtqemud.conf with following setting and restart virtqemud:
#cat /etc/libvirt/virtqemud.conf
max_clients = 500
min_workers = 2
max_workers = 4
prio_workers = 5

#systemctl restart virtqemud

2.Start guest:
#virsh start avocado-vt-vm1

3.Open 4 terminals and switch non-root user to connect 'qemu:///system' in every terminal:
#su - test
$ virsh -c qemu:///system
==== AUTHENTICATING FOR org.libvirt.unix.manage ====
System policy prevents management of local virtualized systems
Authenticating as: root
Password: 

4.Destroy guest in another terminal:
#virsh destroy avocado-vt-vm1
Domain 'avocado-vt-vm1' destroyed

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (libvirt bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2023:2171