kernel 6.8.4-2-pve and many backups failed

[ednt]

Hi,

we were up to date with PBS 3.2-2.
After the last update with the new 6.8 kernel many backups failed.

2024-05-01T08:33:02+02:00: created new fixed index 6 ("ns/backup-41/vm/1025701/2024-05-01T06:32:54Z/drive-virtio5.img.fidx")
2024-05-01T08:33:02+02:00: add blob "/mnt/datastore/pk-pbs-01/ns/backup-41/vm/1025701/2024-05-01T06:32:54Z/qemu-server.conf.blob" (491 bytes, comp: 491)
2024-05-01T08:33:04+02:00: add blob "/mnt/datastore/pk-pbs-01/ns/backup-41/vm/1025701/2024-05-01T06:32:54Z/fw.conf.blob" (103 bytes, comp: 103)
2024-05-01T10:13:02+02:00: POST /fixed_chunk: 400 Bad Request: error reading a body from connection: connection reset
2024-05-01T10:13:02+02:00: POST /fixed_chunk: 400 Bad Request: error reading a body from connection: connection reset
2024-05-01T10:13:02+02:00: POST /fixed_chunk: 400 Bad Request: error reading a body from connection: connection reset
2024-05-01T10:13:02+02:00: POST /fixed_chunk: 400 Bad Request: error reading a body from connection: connection reset
2024-05-01T10:13:02+02:00: backup failed: connection error: connection reset

This happened 2 days and so I took a deeper look on it.
The only big thing changed was the kernel. So I pinned the 'old' 6.5.13-5-pve rebooted and the next backup was working again 100% correct.

Server: HP ProLiant DL380 Gen9
CPU: Intel(R) Xeon(R) CPU E5-2680 v3 @ 2.50GHz
Networkcard for the backup: HP Ethernet 10Gb 2-port 530FLR-SFP+ ports are bonded via LinuxBond
Networkcard for management: HPE Ethernet 1Gb 4-port 331i

Since it is kernel related I think it has something to do with the network stuff.

Btw. on an older SuperMicro board this kernel does not boot in general with a cpu ... failed message.
CPU is XEON X3320 @ 2.50GHz