@acebmxer Yeah a ~15 minute VM migration isn't unacceptable either right? Like any DR plan can simply put that into effect.

But maybe someone else has some ideas.

The 21.46MiB is per second, which is approximately 1/5 of a GB per second.

So that really all kind of tracks...

@acebmxer From the looks of it, I would assume you have a 1GB network interface somewhere in the mix. A switch or something...

But I don't think you should expect to get a full 1GB wire speed in this case either...

@acebmxer So you're getting about 1/5 of the expected throughput... how are the CPU's behaving on XO?

Are they pegged?

@acebmxer That's a loaded question.

What type of connection do you have between the locations. From your edit - "(1gb dedicated Verizon fiber)"

What kind of networking components are in each pool and what type of NIC is the "Management" NIC on your pool master in each location?

200ish mbps is probably really good, but it could be horrible if you have 10gb fiber between the locations..

Is the underlying storage a NAS or some other shared storage that might have contention for IOPS?

I provided some feedback.

@olivierlambert said in XO 6 Official Release (for community):

Hi,

Have you read the release blog post? It contains many information, including a link to this:

https://feedback.vates.tech/?tags=xo-6&view=most-wanted

Feel free to upvote stuff in there to help sorting the priorities

Was in process of it, I plan to vote for it, just gotta swap computers.

I'm exicted that XO6 has dropped for the community edition and there are obviously things that are still pending.

I love that it's polished and cleaned up, however what I must be missing or isn't currently implemented is the ability to actually control the state of a VM. While I get that it's in the plan for next month's release that is a bit of big deal.

That said the screen space usage and the ability to flip back to V5 is easy enough to address this.

@rustylh We really need more information to even begin.

What version of XCP-ng are you using?
What hardware are you using to run XCP-ng?
What is the underlying storage where the VM resides on?
What OS is the VM running?
How are you taking snapshots, XO?
How old was the snapshot?
Were the guest utilities installed into the VM?
You do know snapshots are not backups, right?

@olivierlambert Fun fact -lah is short for lazy ass human

AI is coming...

@edsilber Yeah I'm not really sure, I also assumed that cron (and XO would use the cron pattern for months). Clearly it doesn't.

Using AT on the CLI of your host/pool should work (I imagine). But yeah having a "Run Now" function in XO would be killer.

@edsilber said in Schedule a single non repeating task:

I'm trying to schedule a snapshot for 8:00 AM later this week. I've got the task created but when I go to the schedules I only get repeating cron results, I can't get it down to a single event. How can I make the equivalent of an at versus a cron.

And another possibly dumb question, Cron syntax seems to use months 1 through 12 but the web GUI seems to generate 0-11

You would do this like this,

Granted this is technically every year on the same day. But you can just disable the job after the job ran.

@Biz365 Warm migrations from ESXi are ESXi license dependent on what you have/had with ESXi. Otherwise you can replicate the VM using the migration tool and then manually shutdown the VM on ESXi and then power it on XCP-ng.

@Tristis-Oris So you're on

Correct? Maybe the dev team can help find the issue.

@Tristis-Oris said in VM migration loop:

@DustinB as i say, XO.

Okay so we'll assume XO from Source. What commit are you on?

@Danp Right..

Hrm maybe he has to many concurrent jobs that are running that are beating up dom0.

@Tristis-Oris said in VM migration loop:

@olivierlambert
XO been updated few times during this week. xen 8.3.

i'm dunno what to search at logs. See nothing interesting.

Are you using XOA or XOCE/XO from Source?

@McHenry The issue is host hst150 in pool HST150 is using more than 95% of the available memory based on your screenshot over 5 minutes.

Either migrate some VMs to another host in the pool or shutdown/scale back some of the resources of your VMs.

@samuelolavo Stop the VMs ahead of your host shutdown; gracefully and have "auto power on" enabled on these so when your pool master (and slaves) come back online the VMs start on their own.

@jshiells How much free space is on your SRs?