Topics created by lawrencesystems

hi @lawrencesystems

You still need to enable it in config fr now. I hope we 'll be able to improve the ui to enable it on a per backup setting later

The nbd protocol use the 10809 port in TCP by default. XO only read from this port f the host.

The Nbd setting is used for the whole xo, regardeless of the pool

The lock have been replaced by a select, which is more self explanatory . I think the main control pint is to check if there is multiple NBD network for a VM, XO use the first one returned by the Xapi, and it may be , for example, a managemet network as opposed to a backup / high speed one

Regards

It is released! Kudos to the XO team (and @Danp for spotting another quirk)

@lawrencesystems We added the healthcheck for full backup (here : https://github.com/vatesfr/xen-orchestra/pull/6401 )
Also we removed our js smb implementation ( https://github.com/vatesfr/xen-orchestra/commit/b50e95802c6a876d4e6ffd0c62c209a187b1bfb2 )

to try to circumvent the strange samba comportment

fbeauchamp opened this pull request in vatesfr/xen-orchestra closed fix(backups): add healthcheck to full backup #6401 0 julien-f committed to vatesfr/xen-orchestra feat(fs): remove JS based SMB handler It's not well tested nor maintained.

@florent
Yes, I have the cifs-utils installed.

Since the update I have run full backup that have taken over an hour to complete due to their size and many smaller delta backups and the error has not come up again. I assume whatever was causing the error was fixed in the xo-server 5.84.2 update.

That's the part where are potential/future public API could be very helpful 🙂