When I bulk migrate VMs from host to host, I always set Parallel Jobs to 1, due to only having 2.5gbit networking for the migration network.
Sometimes I will catch Proxmox scheduling two migration jobs at once from the master bulk task. I think it might be due to having some VMs having HA and some VMs not using HA.
It's annoying as the migration pause time when the memory is transferred then gets extended so the VM is frozen for a longer period of time than it should be. Which can causes some connections to timeout if the pause time ends up being too long. Usually I only see a maximum of two at once, I have seen three in the past, but it's supposed to be one. To ensure minimal freeze time.
Now really sure how I report back the problem, other than what my configuration and actions are. As the task outputs say nothing about parallel jobs or anything like that.
Bulk migrate task output
```
Check VM 101: precondition check passed
Migrating VM 101
Check VM 102: precondition check passed
Migrating VM 102
Check VM 103: precondition check passed
Migrating VM 103
Check VM 104: precondition check passed
Migrating VM 104
Check VM 105: precondition check passed
Migrating VM 105
Check VM 106: precondition check passed
Migrating VM 106
Check VM 107: precondition check passed
Migrating VM 107
Check VM 108: precondition check passed
Migrating VM 108
Check VM 109: precondition check passed
Migrating VM 109
Check VM 110: precondition check passed
Migrating VM 110
Check VM 111: precondition check passed
Migrating VM 111
Check VM 114: precondition check passed
Migrating VM 114
Check VM 116: precondition check passed
Migrating VM 116
```
Proxmox: 8.4.12
All VM storage is ZFS, with some VMs having replication enabled as well, between the hosts.
Sometimes I will catch Proxmox scheduling two migration jobs at once from the master bulk task. I think it might be due to having some VMs having HA and some VMs not using HA.
It's annoying as the migration pause time when the memory is transferred then gets extended so the VM is frozen for a longer period of time than it should be. Which can causes some connections to timeout if the pause time ends up being too long. Usually I only see a maximum of two at once, I have seen three in the past, but it's supposed to be one. To ensure minimal freeze time.
Now really sure how I report back the problem, other than what my configuration and actions are. As the task outputs say nothing about parallel jobs or anything like that.
Bulk migrate task output
```
Check VM 101: precondition check passed
Migrating VM 101
Check VM 102: precondition check passed
Migrating VM 102
Check VM 103: precondition check passed
Migrating VM 103
Check VM 104: precondition check passed
Migrating VM 104
Check VM 105: precondition check passed
Migrating VM 105
Check VM 106: precondition check passed
Migrating VM 106
Check VM 107: precondition check passed
Migrating VM 107
Check VM 108: precondition check passed
Migrating VM 108
Check VM 109: precondition check passed
Migrating VM 109
Check VM 110: precondition check passed
Migrating VM 110
Check VM 111: precondition check passed
Migrating VM 111
Check VM 114: precondition check passed
Migrating VM 114
Check VM 116: precondition check passed
Migrating VM 116
```
Proxmox: 8.4.12
All VM storage is ZFS, with some VMs having replication enabled as well, between the hosts.