@karlisi Could you please elaborate why you cant use open source in production? do you think the software is any better with a payed support agreement? do you think the vendor you pay actual is liable for any damage if its not working? do you pay for the expertise from someone other than you to service your infrastructure, so you sit there running critical stuff without knowing what it is or how to fix it yourself just waiting for the vendor to EOL your shit and practically force you on to something new because you dont know how it works? you seems to be a microsoft guy, trained in the spirit that you need to dance by the book of your vendor supplying your hardware and software!

stop doing this, you are the customer, if the vendor cant comply and follow your pracises, find something else to use, they are SUPPORT it means they should SUPPORT your way of doing stuff, not placed in a ticket line where you beg to get them fixing your shit you pay for because documentation on in depth product knowledge is not know to you.

so you are practically downgraded to a support agreement contract portefolio manager?

Stop supporting this crap and dont think open source is any problem to run in production!!

when it comes to pay for the development and contribution from Vates into the XCP-ng project for us to use is something else then support, and that i fully agree with!

So i finaly bite in the apple and fired up a XO instance to perform some backups, my plan was to investigate the possibility for retiring my scripts.

what i found reading about backups in XO made me want to jump in a river head first, its a complete mess!

so we have backup jobs, starts out nice, each job has a planer for running this job called schedule, we can have many schedules, still amazing what!?
so to decide what kind of backups you want, here we have a naming issue, old names, new names, incomplete names, self contradictory names, huh, but not to hard to dig in with the docs. so we further move over to try defining a plan, allready concluded that a backup strategi needs multiple backup jobs, metadata, delta backup, some full ones and XO itself. starting out with the deltas, nice with long term retention policies, needing some daily, more weekly and maybe some monthly, great this gone be nice so to accomplish this we need a schedule to run this not to complicated, but what is backup retention doing there? and how will that override the long term retention and what about multiple schedules? oooh my, this gona be nice i see! lets look at the docs! nada!

so after fidling around a little i see i need some kind of sequence to run all this jobs in, whaat, you are supposed to add schedules to this sequence? i have to admit i starting to get a little warm here, scheduling a sequence to run at intervals, with backup jobs schedules in it at different hours, what the actual f... what is going to run when, and what is going to be backed up with what retention policy?

so here you have it, a complete mess! i think iam leaving this Xen Orchestra as fast as i got it up and running, my dear scripts and xcp center, seems we are going to live a long life together....

Xen 4.0 did some changes to the tick tack machinery for Dom0/U

I see that a default installation of Bookworm default to a clocksource set to xen, this is not good for some or actually most of the applications run today, latency increase and increased load.

Changing clocksource to TSC might be a better option or is it? i have been using TSC for some time with much better latency and performance results

I can agree with you that my meaning of sneaky this time was not accounted for, at first glance my thought was this function was removed by intention by vates when the xsconsole was revamped to further push backup/restore options towards XO usage. I see now that i was wrong in that assumption!

@stormi

How do you restore this metadata backup in xcp-ng 8.3 without the use of XO?

This is absolutely something to write about in the release note of 8.3. when is it expected back?

@Danp Was working perfectly fine in 8.2.1 LTS

In xcp 8.3 xsconsole has got a revamp, i suspect this function was removed in this operation.

What happen to the XSCONSOLE backup restore and update function?
where is the pool metadata backup functions and schedules menu?

why is this not mentioned in the deprecation notices here in the docs?
i am happy we did catch this before upgrading, anyway to get this function back?

Why isn't the removed functionality available in XO lite?

Snicky this one from vates

Iam considering trying out the testing branche of debian, currently bullseye. Are there anything wrong in using the debian 10 Buster template in XCP-NG 8.1 to run bullseye?

@stormi as long as you dont have any untrusted vms running on this cpus there is no problem with security issues.

Thanks you guys for testing the l5630 cpu, this gave me confidence to upgrade my dell r610 with 2x x5675 cpus with hypervisor 8.0, upgrade was done with iso and cd and the upgrade process went very smooth, server boot up and everything seems almost nice.

i did notice it load cpu microcode rev 1f on boot, i also notices systemd is throwing a error on boot: systemd failed to load kernel modules, this has no impact and the host is running fine with no error other than that. i suspect the error might be related to upgrade fra 7.6, i will try to reinstall 8.0 as a fresh install and see if this clears the kernel modules stuff.

Can anybody that has the means try to run Citrix Hypervisor 8.0 on those legacy 56xx series CPUs?

Have anybody tried xcp 8.0 with the intel x56xx series cpu?

why is support for this cpus dropped, are there any technical behind it or just that they are old and considerd Legacy?

i would like to know if this cpus have been tried with this veriosn of xcp.