@DustinB said:

@laszlobortel While I can understand "Upgrading not being an option" you're lift and shifting the workload (or at least have been attempting to do this to date).

Are you unable to build new and migrate data over to XCP-ng, while I could see this causing more work, lift and shifting is almost always a guaranteed way to cause headaches - like the ones you're experiencing.

That is why each service provider recommends building new if you can. At the same time that you're building new, you're updating which of course can cause issues - but continuing to run Rocky8 is only receiving security updates until 2029. Sure it has a few years left, but why not take the opportunity to upgrade?

They can possibly go up to Rocky 9 but Rocky 10 later may be harder, as it requires a higher baseline on the CPU. Also they may have legacy software that only works on Rocky 8.

@unreal-shizzle said:

@olivierlambert said:

Hi,

Somewhere between Q1 and Q2 next year, hard to be more precise

My company is desperately looking to get away from VMware because of the predatory pricing. I showed my boss XCP-ng because it looks a lot like ESXi. He looked into the software and asked why XO lite has been "coming soon" for 5 years now.

What should I tell him?

It may be help to actually look at Xen Orchestra as XO Lite is targeted at basic functions. Anyway network creation is present in Xen Orchestra 5 and/or 6. Maybe more extensive, but will allow for better ACL (via RBAC) capabilities.

https://docs.xen-orchestra.com/xo5/sdn_controller#network-creation

The pieces are coming together it’s a new software XO Lite, the teams at Vates are growing more of XO Lite is being developed and tested. Though Xen Orchestra is a more suitable base for a business environment, with multiple departments and teams, due to XO Lite only using and allowing the use of root account authentication. The root account of XO Lite refers to the XCP-ng host server administrator account - full access!

@tjkreidl said:

@john.c @johnnezero Well, the Github reporsitory is in place already! Access it at:
https://github.com/tobiaskreidl/Citrix-Tobias-Kreidl-Collection
Feel free to find more material to add, as will I, and thank you kindly for the suggestions and support.

Is the repository supposed to just have the readme, or are there supposed to be other files (e.g. MDs, pics or the PDFs? Or is it in the releases section - which is currently empty?

@tjkreidl said:

@john.c Wow, that was amazing -- not sure why my searches were unsuccessful, but many thanks! I think Github might be a good option for putting these on-line as a more reliable spot. And, yes, preserving images is always a challenge.
I do hope some of that information may be useful to you and thanks much again for all your efforts, John!

A bit of AI archeology combined with Internet Archive Wayback Machine helped. The AI helped to surface the exact URL of the blog posts and these were fed into Wayback Machine. Used the snapshot from 2022 anything newer and they weren’t there anymore. By the way it may be worth sending a donation Internet Archive’s way (if not done so already), as without this your blog posts would be truly gone for good!!

Oh forgot to add GitHub and GitLab follows digital wills, so you can pass this off to another in a will, or generally be made read only while being kept up, when the time comes. So this hard work will be preserved even in the face of the end.

Just please let me and the rest of us know where it’s at when done!

@tjkreidl said:

@john.c Not found with the Wayback Machine, alas. Still not finding it anywhere else, but will keep looking!
It's a crying shame Citrix didn't preserve the treasure trove of old community blogs.

I did a bit of digging with the aid of AI and I’ve managed to uncover the original three blog posts of NUMA and references to UMA. If you do some more digging you may be able to uncover the rest, so it can be rewritten and/or updated, then be hosted somewhere that won’t go down so easily, or be lost as easily. If you do an update or write etc, may I suggest switching the images used to WebP or AVIF format, will seriously help file size while maintaining their quality (or even giving room for higher quality). Consider switching to SVG for diagrams rather than raster (or as the default). I’d suggest checking out to consider using Mermaid for diagrams (https://mermaid.ai/open-source/?utm_medium=hero&utm_campaign=variant_a&utm_source=mermaid_js). Maybe if you do a rewrite use markdown and something like Hugo, to generate it from the files to host on GitHub or some other pages providing repository (e.g. GitLab Pages or Codeberg Pages).

https://web.archive.org/web/20220527221535/https://www.mycugc.org/blogs/tobias-kreidl/2019/03/07/tale-of-two-servers-bios-settings-affect-apps-gpu

https://web.archive.org/web/20220527213026/https://www.mycugc.org/blogs/tobias-kreidl/2019/04/30/a-tale-of-two-servers-part-2

https://web.archive.org/web/20220527215004/https://www.mycugc.org/blogs/tobias-kreidl/2019/04/30/a-tale-of-two-servers-part-3

https://community.citrix.com/forums/topic/235895-xenserver-vm-citrix-worker-sizing-question/

https://xcp-ng.org/forum/topic/9359/cpu-provisioning

https://community.citrix.com/forums/topic/237493-memory-and-cpus-assigning-to-vms-in-order-to-obtain-maximum-performance-according-to-numa-topology/

https://community.citrix.com/forums/topic/241553-bios-power-performance-settings/

https://community.citrix.com/forums/topic/243640-citrix-hypervisor-performance-tips/

@tjkreidl said:

@johnnezero It would be also interesting to take UMA/NUMA into account as VMs -- in particular, VMs with vGPUS -- can run much more efficiently if they do not cross memory bank boundaries that span more than one CPU instance. On some Linux systems -- not sure about the one hosting XCP-ng -- you can even disable NUMA. Just an additional thought here. I published a number of years ago a three-part series "A Tale of Two Servers" discussing a number of related optimizations but alas, the Citrix blogs were eliminated and I'm snot sure where copies of these articles exist anymore. But there are plenty of articles on this, in particular by Frank Denneman, and also ones like the following;
https://indico.cern.ch/event/304944/contributions/1672535/attachments/578723/796898/numa.pdf
https://docs.xenserver.com/en-us/xenserver/9/numa.html

If you remember the URL and date maybe try the wayback machine of Internet Archive. They’re known to archive sites and articles wherever they can, may hold a copy that’s accessible.

@johnnezero said:

@john.c Thanks much, looking into it.
"Open-Source for the Win!"

Your welcome!

@johnnezero said:

@Pilow Sounds like an awesome idea. Send any details you may have on how to make plugins (if you know how that is).
Adding to the ToDo list - Thanks!

@johnnezero said:

@Pilow Sounds like an awesome idea. Send any details you may have on how to make plugins (if you know how that is).
Adding to the ToDo list - Thanks!

Looks like I can help out again just tracked down this past thread on the forums.

https://xcp-ng.org/forum/topic/7202/how-do-i-create-a-new-plugin

How’s your JavaScript (Typescript), JSON etc? These are the languages needed to develop plugins for Xen Orchestra.

@johnnezero Also there’s shortcut directories including one called cron.hourly, you can place a symlink (or hard link) to the script there. Cron will then execute the script without you needing to alter the crontab file.

Just drop .sh for using the shortcut directory as it will not run there otherwise.

@johnnezero said:

@john.c Great ideas (especially if Vates decides to bake something similar into XO someday) but may be getting too far into the weeds for now...

That’s okay. Just putting it out there - no rush! Any way to maintain Linux good practice relocate the conf file to /usr/local/etc/ (or /etc) then keep script in /usr/local/bin.

@johnnezero said:

@john.c Great idea, and done! ("Keep 'em coming")

Thanks for the change. By the way I meant when doing Vendor config and customer workflow config. To implement that requires nested config file. In other words the set-performance.conf, then a file for instance in a set-performance.conf.d/custom.conf (or similar). The custom.conf in the conf.d directory overrides the same property as well as section within set-performance.conf.

@johnnezero said:

WHAT: Automatically assigns CPU weights and I/O priorities based on assigned VM tag (i.e. replicating what vcenter did via resource pools etc.).

It would be even better if you could split the configuration section off, so that it’s in its own conf file. Would make it easier to manage, also if this ends up being used, by Vates in the Vates VMS software. There can then be a vendor recommended configuration with the option of customer’s own workflow based, configuration.

@tsukraw said:

Hey guys,

Looking for some help at understanding auto power on as the documentation possibly looks to need some refreshing.
https://docs.xcp-ng.org/guides/autostart-vm/

From the Xen Orchestra, the guide does not mention the pool level it is only under the CLI that it mentions the pool level.

Just looking for some clarification.

In order for auto power on to function, we need to enable it on the pool level, and then per VM correct?

If you have auto power on enabled at the VM level and the pool level disabled, the VM level is ignored, correct?

What is actually triggering the power on?
Is it XOA or the host itself?
I am assuming the host itself, just looking for clarification if XOA VM was part of the systems that went offline, should we expect to see it auto power back on if it is set to auto power on.

What determines the start order for auto power on when you have multiple VMs on a host?

Thank you!!

Yes, it needs to be enabled at pool level and at VM level. I tried once at only enabling at the VM level and it didn’t auto power on, as it wasn’t enabled at the pool level, it started auto powering on once it’s enabled at both levels.

It’s configuring a XAPI value as this occurs at the Xen level (virtualisation server). Xen Orchestra is just making this value visible with a nice UI interface toggle.

@gcpeters4 said:

@john.c

This is a good idea as well. If the previous suggestion of trying to do a DNS round robin approach doesn't work, this may be my best option.

Thanks for taking the time to provide your suggestion!

By the way my suggestion is recommended best practice from Microsoft for their Active Directory software and/or technology. My suggestion will be more likely to work if your DNS servers and requests have issues with Round Robin configuration. Plus with my solution you can easily add more servers as domain controllers to the cluster, as required and use replication to keep them consistent and up to date.

@gcpeters4 said:

Is there a way to specify a primary and secondary Active Directory server when configuring auth-ldap?

I've encountered situations where the specified URI (ldap://activedirectory1.domain.x.y) server isn't reachable so I'd like to be able to specify a secondary AD server using something like this:

ldap://activedirectory1.domain.x.y, ldap://activedirectory2.domain.x.y

Basically multiple AD servers; separated by a comma.

Unlike the other forum post regarding Multiple AD Sources, both of these control the same domain.

Thanks.

Another option is to setup AD clustering (clusters) with both controllers part of the same cluster. Alternatively setup an AD Failover which has a single FQDN endpoint but two members.

Either or both of these likely to result in a single FQDN that can be fed into the plugin.

@CyaVMware said:

Have a simple question for someone.

Looking at migrating a client to XCP. Currently they have a data disk that is exactly 2TB in VMware.

My question is the limitation anything OVER 2TB, or is it 2TB AND above?
I'd prefer not to use the VMware converter tool to shrink the disk to 1.99TB if I don't have to.

VHD format has a strict limit of roughly 2TB (around 2040 GB), so if your client’s VM disks are larger than this then you’re going to have trouble with VHD format. However QCOW2 support is almost here (production ready) about to reach RC2 state, once this reaches stable you’ll be able to safely reach a max of around 16TB. That is if your using ext4 based SR, other Filesystems have different limits. If using a networked file sharing protocol service then, the limit of the host server’s filesystem applies. Though QCOW2 VDI has a limit of 16 TiB.

@DustinB said:

@john.c This is already viewable under each hosts in the pool.

Then in that case the pools based view restructure can go ahead. Was just concerned about ensuring that it’s still possible to access individual hosts.

@DustinB said:

@john.c said:

@DustinB said:

@olivierlambert https://feedback.vates.tech/posts/49/pool-view-rather-than-host-view

So you’re interested in having it a bit more like when using VMware vSphere while, regarding VMs, hosts and pools? That could work, but still need a hosts view for each member of the pool. In the host’s view it needs to be able to see, what VMs are running on it.

Also needs to be compatible with https://github.com/vatesfr/xen-orchestra/issues/9430. If not already synced with NetBox, the pool and hosts should be linked by UUID when synced. So in other words the structure needs to be pool is linked by UUID to host by its UUID and then the same for VMs. So the VMs are linked by UIID to the host UUID.

Nothing in my post would restrict the view to only seeing VMs within a specific Pool, the view should be flexible enough to shows VMs in a pool.

As it is now, you see all VMs that are running under a given host and then pool.

This in and of itself could become tedious to manage/dig through.

The potential remedy here would be to show all VMs in a pool, and on the given VM show the host where it's running.

Even in VMware vSphere there’s a hosts view, that shows things specific to a particular host. This includes VMs running on it, or are currently halted or suspended on it. As well as patches missing from the host, etc. Though it more roundabout to get to it under VMware vSphere.

matiasst created this issue in vatesfr/xen-orchestra

open Link VMs to Host Devices via UUID in NetBox Plugin #9430

@DustinB said:

@olivierlambert https://feedback.vates.tech/posts/49/pool-view-rather-than-host-view

So you’re interested in having it a bit more like when using VMware vSphere while, regarding VMs, hosts and pools? That could work, but still need a hosts view for each member of the pool. In the host’s view it needs to be able to see, what VMs are running on it.

Also needs to be compatible with https://github.com/vatesfr/xen-orchestra/issues/9430. If not already synced with NetBox, the pool and hosts should be linked by UUID when synced. So in other words the structure needs to be pool is linked by UUID to host by its UUID and then the same for VMs. So the VMs are linked by UIID to the host UUID.

matiasst created this issue in vatesfr/xen-orchestra

open Link VMs to Host Devices via UUID in NetBox Plugin #9430

@acebmxer I’m sorry to say that if those Dells are at your workplace, the wrong edition of iDRAC was purchased. You see with at the very least iDRAC 9 access to the full granular IPMI sensor data, was placed behind an edition paywall by Dell Technologies. Outlet temperature is just one of the feeds, missing from the Enterprise or lower edition of iDRAC 9. You’ll get the temperature readings from the Dell iDRAC web browser based interface, but not IPMI with iDRAC 9 Enterprise.

To obtain the full IPMI sensor data you need the Datacenter edition.