Posts made by lawrencesystems

@olivierlambert

I work with enterprise companies and they NEED those reminders to let them know they should be using the supported version. I don't really see them as annoying at all.

@damjank
I have a video on dynamic memory here:
https://youtu.be/Lsi2-hAoKSE

@olivierlambert

This was a full backup not a delta so that makes sense. I think I need more of my lab servers to be on 8.3 so I can do some more testing. I did not find any issues and nothing broke using it so I need to try harder.

I did some testing with the 8.3 and it all went smooth except for one thing. We did a backup of a VM on 8.3 and if we try and restore it to version 8.2 I get the error

"IMPORT_INCOMPATIBLE_VERSION"

which makes sense and it's just a lab system so nothing critical. Is there a way to bring it back to 8.2 or should I update the rest of my lab to 8.3?

I am running xo-server 5.109.0 & xo-web 5.111.0 and I have enabled NBD from the UI and verified that the parameter is showing on that network. This is the network on which XO is talking to XCP-NG .

xe network-param-list uuid=12eaf1ae-9596-723e-a534-4d5a101ec4b4

uuid ( RO)                    : 12eaf1ae-9596-723e-a534-4d5a101ec4b4
              name-label ( RW): Dot3
        name-description ( RW): 
               VIF-uuids (SRO): 66e895eb-b52a-a370-37a4-4bd74f299be0; 345ff1c0-0ea2-44dd-6308-07eabd75f770; 36072018-1e8b-03ea-a675-94c7b540e0e2
               PIF-uuids (SRO): cc5f027f-cdf4-1e17-f393-51bf450c6855
                     MTU ( RW): 1500
                  bridge ( RO): xenbr3
                 managed ( RO): true
            other-config (MRW): 
                   blobs ( RO): 
                    tags (SRW): 
    default-locking-mode ( RW): unlocked
                 purpose (SRW): nbd

I know I also have to be using the multiple data blocks on for the backup repository but I do have a few other questions:

• Because the NBD option is now in the XO UI do I still need to edit the XO config.toml file & restart XO to enable or is that automatic?
• Are there any firewall changes on the XCP-NG host or does this work over 443?
• Does it matter if XO is running in another pool?
• In the Xen Orchestra 5.76 Blog Post it was mentioned the "Lock" in the new "NBD" row which I see but is there some change I need to make?

@bullerwins

It's a wonderful feature, you can have customized filters for VM, storage, Templates, and Pools. Combine that with using lots of tags and it makes larger deployments very manageable.

@isener

I think the XO-CLI has to be registered with XO to get it working and seeing as you are locked out that is not an option. Maybe someone here will offer a way to edit the database if not I would just delete and rebuild XO.

@AlexanderK

Clonezilla is my go to for doing that. I have a tutorial where I should how to migrate from Virtualbox to XCP-NG here:
https://youtu.be/wSTk9BLwF5k

@mauzilla
Just an FYI, I have not tested the latest Scale 22.12 RC1 for performance but the previous versions of scale under some conditions was quite a bit slower. I have a video covering this topic here with the benchmarks and methodology used for testing.

https://youtu.be/BoiHHnBDg0E

@MichaelCropper
I recently did a video tutorial on how XCP-NG virtual machine dynamic memory management and over allocation works.

https://youtu.be/Lsi2-hAoKSE

@Gheppy

The current version of TrueNAS-13.0-U3.1 has been stable for me, but this issue has persisted when doing Delta backups with SMB
& TrueNAS, even older versions.
https://xcp-ng.org/forum/topic/6148/delta-backup-showing-success-no-delta-saved

My resolution to the issue was to move the backups over to NFS and they work well. Also, since I am using the newer "Store backup as multiple data blocks instead of a whole VHD file" option NFS is generally a bit faster with handling small files.

@julien-f

Thanks, that fixed it!

@julien-f
I rebooted and I have tested both tcp:// and udp:// and it's not sending anything. Also checked using ss -t to and I don't see any TCP connection going to the logging server. Support # is 44423

@julien-f
Tunnel support # is 44423

@julien-f Yes, should I DM you the support number?

I really like this feature but I am having some issues getting it to work. I assume this goes in the config.toml file located in /opt/xo/xo-server for XO Community or /usr/local/lib/node_modules/xo-server for the XO Appliance.

[log.transport.syslog]
target = 'tcp://syslog.example.org:514'

I have tested on both XO community and XOA and nether worked and I don't have any error messages in the log files stating the config is incorrect. I have rsyslog setup and working so I know it's able to reach and send to the log server and send data. Also, I assume this can this be done via UDP as well but changing the prefix to udp://?

@kevdog

Not really sure what your goal is, but for simplicity I recommend keeping the main interface on the native VLAN. If you are doing this for security, then you should use separate physical interfaces to keep the network segmented.

I really think there is some bug in the TrueNAS Core SMB stack that I am fighting here because this issue
https://xcp-ng.org/forum/topic/6148/delta-backup-showing-success-no-delta-saved?_=1666089968101
I had reported before was only happening on TrueNAS Core via SMB. Also this is solved sine XO that extra integrity checking feature was added to the delta backups.

@florent

No encryption on this remote. And if I have time before I leave I will test importing the XVA with CLI and see what error I get.

I am going to be out of town until next week, but I can keep the backups so we can arrange a time to test it.

I have done a few more tests, I setup a Synology using SMB and it backs up and restores just fine. I also moved the existing backups over to an NFS share and that previously created backup gives the same Stream_vdi.Invalid_checksum error with the same block number on restore.