'All' Backup Jobs runs in parallel

[xsigndll]

I've just recognized, that the '-- All --' Backup Job is running everthing in parallel. Is it possible to switch it to sequentially (except setting it up manually for every VM or Host)?

Reason for asking is: When 8 hosts are doing backups at once, the network locks up noticeably. IMHO a sequential backup option would lower the pressure on the overall system.

 

[Neobin]

For reference:
https://bugzilla.proxmox.com/show_bug.cgi?id=3086

 

[itNGO]

I've just recognized, that the '-- All --' Backup Job is running everthing in parallel. Is it possible to switch it to sequentially (except setting it up manually for every VM or Host)?

Reason for asking is: When 8 hosts are doing backups at once, the network locks up noticeably. IMHO a sequential backup option would lower the pressure on the overall system.

View attachment 41119

View attachment 41118

There is a Bugzilla for this as Neobin already wrote. But for now, create single Jobs for every Host to reduce the load. This is by far not perfect as Jobs can still overlap, but currently the only way.

 

[Lephisto]

Is tehre some solution for this already? In the current form, the PBS does not scale, unless you have all-flash Backup storages.

 

[ku4eto]

Its 2024, and there is no decision on what to do about this, let alone an actual implementation. Thats a big bummer.

 

[Tony]

I ran into this problem, too. After reading this thread and the bug report, I ended up with the solution suggested by comment #13 in the bug tracker: run ssh to each node to invoke vzdump.