@panzersrmm I tried to snapshot and revert on a VM with Resource set + ACLs and they did not disappear... ?

@olivierlambert when planning our infrastructure, Vates engineer told us to avoid Broadcom nics but to go Intel or nvidia/Mellanox.

we bought Mellanox

This is the way.

@tc-atwork plus +++ on this one.
Tags in XO5 are so useful for smart backups, to filter everything.

in XO6 for now they seems to be just informational. we can't even filter the treeview by tag actually !

hope the benefits of XO5 tags will be ported to XO6

@acebmxer if possible for you, restart toolstack of the host being attached to this VM.
and check to see if this vdi still attached to dom0

sometimes a full reboot of the host is needed, and you get an orphaned snapshot instead of vdi attached to dom0

delete the orphaned snapshot and. backup will get in place like before.

@MajorP93 stress of VM migration XD

good luck with your V2Vs !

@MajorP93 in your screenshot, CA and CA_1 seems to be two disks on the original VM... ?

They're even different in size.

each are suffixed with -000001-delta, indicating one snapshot, on each disk

two vmdk on source gives you two vdis on destination, as intended

@olivierlambert I get you.

Same in veeam and synthetic fulls, it's only between the server/proxy and the repository.
but we get a progressing percentage while the job sits while doing the synthetic full.

would be cool to "see" the progress of merging.
why ? bad habits from previous technologies used I guess.

@florent i always check it, to avoid a merging process to be disrupted by another job.

so when the backup tranfer is done (we have a progress bar), the job just sits there.. .merging in background, without progress bar or percentage.
5 minutes later, the backup job finishes.

so your stats are flawed (used data over time) if time of merging is taken in account

see what I mean ?

@farokh one backup job should leave one snapshot.

this is the way. to differentiate from previous backup the data to be put in the delta, it has to refer to this snapshot.

delete it and you get a full backup ongoing.

if you have two snaps on your VM, either you have a manual one, or you have multiple backup jobs backuping this VM (check the backup tab of the VM ?)

@flakpyro thank you

@florent , would be nice to add this to XCP backup ? merge progress

@florent yep, bandwidth computation on Veeam has always been big numbers because of that method.

another thing is MERGING at each remote transfer ending that make feel the backup is longer with XCP BACKUPS
would be cool to have some insights (other than looking for the lock file in the remote) to see the progress of merge.

finally understood the network problem...

If you have a match of MAC ADDRESS in netplan

Veeam restores the VM with a NEW MAC ADDRESS

so either you delete the match, or get the good mac on the VIF...

@olivierlambert yup for sure

we will mix xcpbackup/veeam server backups (we used to mix xcpbackup/veeam agent backups) to get app aware backups (AD, SQL, ...)

on the restore side, i'll note two things :

• veeam doesnt like spaces in the VM names when you restore (but XOA is UUID based and don't care...) so we have to rename the VM without spaces
• no progress bar

restored VM is booting, but network seems broken (no ip)

just to #brag

Deploy of worker is okay (one big rocky linux with default settings 6vcpu, 6Gb RAM, 100Gb vdi)

first backup is fine, with decent speed ! (to a xcp hosted S3 minio)

will keep testing

@acebmxer did you test latest Xen Tools ?

https://www.xenserver.com/downloads#lightbox-20160

7 days old though

@olivierlambert said in little UI bug, proxy filtering remotes:

We'll have those views in XO6 very soon. Could you test there to see if it works OK when it's available there?

Reply

Hi Olivier,

tested latest 5.111 XO6 and can't find the related REMOTES page ?
every link to a remote sends me back to XO5 web ui

Proxies web configuration page seems not to be implemented in XO6 at the moment

@DustinB with Easyvirt integrated in the stack

you're right

I'm okay with error reporting, but it breaks the sync (all subsequent sync of other items to netbox is stopped)

Expected behavior would be an alarm ? and not to stop netbox plugin functionnality

Hi,

If you have a DR Job, the DR vms created have the same IP address as the source one
and you have the netbox plugins that halts on error, conflicting existing IP.

Could you change that it checks mac+IP address ? as the two DR vms seems to have different mac addresses when created by the DR Job.

Same problem if you make a COPY of a VM ! the copied VM has the same IP address and conflicts when syncing (I bypassed this one by changing a bit in the mac address and no problem anymore ? so it checks the couple MAC+IP ??)

Any advice to get a smooth netbox sync appreciated.

@acebmxer indeed, same problem here, no smart mode in MIRROR backups.

it's either MANUAL selection, or ALL vms.