device: 0.751
kernel: 0.613
semantic: 0.600
virtual: 0.545
architecture: 0.543
hypervisor: 0.518
ppc: 0.512
graphic: 0.467
performance: 0.466
VMM: 0.458
permissions: 0.434
register: 0.375
mistranslation: 0.366
socket: 0.364
vnc: 0.362
PID: 0.352
risc-v: 0.342
x86: 0.332
assembly: 0.309
i386: 0.285
files: 0.269
boot: 0.251
user-level: 0.249
debug: 0.245
network: 0.225
arm: 0.196
peripherals: 0.133
TCG: 0.133
KVM: 0.068

qemu-ga fs-freeze causes VM to
Description of problem:
I have fresh install of Debian 11 and installed MariaDB 10.7 from MariaDB's Repo. Guest is fully up to date.
When Proxmox goes to do a backup it will call fs-freeze to the VM via the agent which then causes the backup process to hang and the VM will lockup or causes kernel message such as `**task qemu-ga:370 blocked for more than 120 seconds**`. The VM from what I can tell no longer is able to write to disk, and the only fix is to force reset the VM.

The issue doesn't happen when the VM has first started or the agent has been restart from what I can tell, but if you leave it and wait for the nightly backup to run of the VM, it will then cause this issue to happen.

There are other reports of this happening on the [proxmox forums](https://forum.proxmox.com/threads/snapshot-backup-not-working-guest-agent-fs-freeze-gets-timeout.99887/) More details on this topic. Other reports with the issue with MariaDB 10.6.

My other Debian 11 VMs which were also setup recently, do not experience this problem, only difference is this VM is running the MariaDB. I have Gitlab, Docker, Mailcow, PowerDNS, OPNsense (each of these separate VM) in the other VMs and they do not experience this issue. All these VMs are running Debian 11.

Agent Info
```
qemu-guest-agent/stable,stable-security,now 1:5.2+dfsg-11+deb11u1 amd64
```
Steps to reproduce:
1. Install Proxmox (although I would assume any QEMU 6.1.1)
2. Create a Debian 11 guest with MariaDB 10.7 from MariaDB repo
3. Wait good few hours
4. Issue a backup or fs-freeze