@danp said in This server is a different version to the master:

5.601

You are rigth, the problem was there. I didn´t found any package related to the schema version, but are rigth, so I've look into this. Here a problem like mine, and the key to resolv it. As my host0 was not updated, it was working with schema version 5.601. just a xe-tools-restart was enough to update the schema version to 5.602 and now they match.

Thanks!

@danp Yes, I considered it, even tried a pool migration, but disks of 1TB take long hours to finally fail. What's more when I've migrate a machine, (and I've migrated several hundreds of them), time to time one of them gets bock and need a reboot, which is something I want to avoid, specially if I don't know when is going to finish, because i takes several hours (may finish at 3 a.m and no one there to check it's correct)

Hello everyone,
I have one standalone host in a pool (yes, a pool of one), let's name it host0. This was a 8.1 versión updated to 8.2 by "yum update" before any VM was running there.

Now, I have a new and identical host (host1) to add to this pool, but I can't add it to the pool because it says that version of master (host0) is different.
Host0 was updated to 8.2 and started running several VM on it, so is not possible to reboot it in order to be updated, but a yum update was done, I mean, it is waiting for a reboot to be 100% efective.
Host1 was install in the same way, firstly 8.1 and it is updated later to 8.2. This process lets host1 updated to last available versions, while host0 is not, because the reboot is not possible.

I've look into host0 version and downgrade host1 to the same version, but it continues mismatching version.

xe host-param-list uuid=7fa3f334-f878-4608-9cb5-0d2707706f86 |grep version
                     API-version-major ( RO): 2
                     API-version-minor ( RO): 16
                    API-version-vendor ( RO): XenSource
     API-version-vendor-implementation (MRO): 
                      software-version (MRO): product_version: 8.2.0; product_version_text: 8.2; product_version_text_short: 8.2; platform_name: XCP; platform_version: 3.2.0; product_brand: XCP-ng; build_number: release/stockholm/master/7; hostname: localhost; date: 2020-11-05; dbv: 0.0.1; xapi: 1.20; xen: 4.13.1-9.8.2.xcpng8; linux: 4.19.0+1; xencenter_min: 2.16; xencenter_max: 2.16; network_backend: openvswitch; db_schema: 5.601
    virtual-hardware-platform-versions (SRO): 0; 1; 2

xe host-param-list uuid=7e5f26d9-59d6-4674-ad7f-344e93c5ebb2 |grep version
                     API-version-major ( RO): 2
                     API-version-minor ( RO): 16
                    API-version-vendor ( RO): XenSource
     API-version-vendor-implementation (MRO): 
                      software-version (MRO): product_version: 8.2.0; product_version_text: 8.2; product_version_text_short: 8.2; platform_name: XCP; platform_version: 3.2.0; product_brand: XCP-ng; build_number: release/stockholm/master/7; hostname: localhost; date: 2021-05-20; dbv: 0.0.1; xapi: 1.20; xen: 4.13.1-9.8.2.xcpng8; linux: 4.19.0+1; xencenter_min: 2.16; xencenter_max: 2.16; network_backend: openvswitch; db_schema: 5.602
    virtual-hardware-platform-versions (SRO): 0; 1; 2

As you can see, according to host, they are running the same version, but something is missing. What I'm trying is add host1 to the pool ruled by host0, move every VM to host1 and reboot host0 to get it updated, repeating the process with host1.
Any idea about how to get a matching version working only on host1?
Thanks in advance

@olivierlambert It didn't work. So I had to reboot the host. Is the same under xcp-ng 8.2?

@olivierlambert Hi,
I'm using xenserver 7.2 (pending migratin to xcp-ng in this pool)
I had to reboot the host, and as espected, it worked, any way, I'd like to know it another option would had worked.

Hello everyone.
I had to restart my pool master toolstack, and no way to reconnect to it again from XCP-ng center.
I'm talking about a pool of 8 host running 7.2 xenserver.
After log into the host I realized the /var/log partition was full, so there is the problem. I take from granted that a host reboot should resolve the problem, but as always, it's complicated.
Are there any way to restart the service (or whaever ) to get my pool back again?. A "systemctl restart xapi" or just xe-toolstack-restart doesn't work. According to xenserver documentation, I must free space (done) and restart the host, but this doc was for 6.5.

I'm asking because I need to increase my pool's CPU resources, the easy way is buy the same processor, but thinking in a more powerful option, RAM and networking features are ok, but compute performance is being starvest by some VM.

I have an old pool, which I had to upgrade some years ago, but no more original model components were available, so I had to buy a new model, new cpu generation, so it became an heterogenous pool. The problem was not the lost of new cpu features, the problem is that time to time, when I performe a live migration VM, it hangs and I must reboot the VM.

Keeping in mind to avoid and heterogeneus pool, I think the safest option is 5220S, because 6226R implies more changes which I unknow in the architecture, maybe new features, that's what I unknow.

Any piece of advice?

I think no, but I'm wondering your opinions.

In my pool I have 5 server with double intel 5217 (8C 3GHz) processors, I'detected that CPU start to be very higth , so thinking in more powefull CPU. Model 5220S is 18C cores at a similar speed. Both, 5220S and 5217 are gold series of "cascade lake" architecture. So, features must be the same, only cores and speed differ. Intel 6226R is another interesting option, quite similar speed and doubles chip cores.

According to citrix, "Combining different CPU models that have identical features" is the 4th type of heterogeneous pool, and "Type 4 represents CPUs with different marketing model names but identical model number, family, and feature flag attributes. Because they have identical attributes, these combinations have always been supported and do not require a mask, but their compatibility has not been obvious when comparing the marketing model names."

Do you think these chips (5220s and 6226R) would mean an heterogeneous pool or not?

In other words, you must empty the host before add it to the primary pool, of course, you can migrate VM from one pool to another. And if your hw is not the same, but similar, you'll have an heterogeneous pool.

5% sounds too much for me, but I'll check it in some populated pools (with host's cpu below 80%) to have a better idea.

Maybe somebody could point some metrics to predict future cpu/Network overcommit.

As always, reading you expands my knowledge. point 1, perfect, as I thought it was. Point 2, I thought dom0 talk to xen hypervisor by XAPI, thanks for your correction, I supposse dom0 is the only one that can communicate with xen and xcp-ng xencenter or XO use dom0 via XAPI, beeing dom0 who finally passes the order to xen. Point 3, that's what I want, observe parameters like wait and st from top command to detect bottleneck or overloaded resources, like CPU or SR access.

And going back to my original question, what st values should be allowed?. for instance, 20% must indicate a serious lack os resources in the VM.

Anrother wrong concept here for me or I've explained myself wrong. Xen is the hypervisor, is not dom0 a hidden vm where you connect when ssh to the host?, where xe commands are played and the one who controls the rest of the VM running on the same host? Like xencenter, xcp-ng center or XO, dom0 is a way to give orders and get states to/from xen hypervisor via XenAPI.
For instance, start or shutdown any VM is a order that dom0 gives to xen hypervisor, or at least it's what I've always understood.

When I said that I was looking in top, I mean the top command in the VM, not the dom0

Hi everyone.
As far as I'm concerned, the st parameter or steal time is the % of cpu the hypervisor (Domu0) steal from the VM, because it can't supply enough time slots due to a overcommited host. When the host is empty and has plenty of resources, it may give as much cpu time as asked.
I don't know if DomU0 works in extrict or best effort mode, giving just the resources aplied or even more when plenty of resources.

My question is, keeping in mind that hypervisors are not ideal, what % of st is allowable when the host is relaxed?, because I observe st values below 1% in a VM on am empty host.

I'm going to populate the host in next days, and a bit worried about not overcommit it, so I'll be watching this parameter.

As always, if I'm wrong in some point, please correct me.

Well, for me was not clear untill your current answer. As you can see, my conclussions were wrong, but that was just a suggestion.

Could I suggest to update https://xen-orchestra.com/blog/xen-virtualization-modes/ with your answer as the current state-of-the-art?
Do you want me to post my question to get your answer? It's clear for me now.

Another solved post

Regards

So, if I've undertood you well, the mode to get the best performance is HVM, which gets PVHVM under Linux VM when tools/management agent is installed (the drivers are currently in the kernel) and gets PVHVM after install PV drivers and management agent on Windows OS. Rigth?

Hello everyone,
After googling aboit it to refresh concepts, I've ended up confused, coul anybody correct me?

In short:
PV : The best to emulate when no hw assisted exists (really old proccesors). Need kernel modification, no problem for linux.
HVM: When HW help virtualization, present in nowadays proccessors. No Need kernel modification and allow SO act over real HW
PVHVM: The best of both worlds, PV improves it's performance helped by HW assistance
PVH: This concept is new for me, and accordding to this post of Oliver, there 2 versions, v1 and v2.
After read the post, the conclusion for xcp-ng is that when I see a VM working in PV mode , it's indeed PVHv1 and when I see HVM we are really talking about PVHv2. Am I wrong? pv, hvm and pvhvm are no longer in use under xcp-ng, right?

An finally, must Windows VM work in HVM or PV?, Because I undertand that HVM is the best way, but citrix tools refers as PV drivers, not HVM drivers.

P.S. I moved the most of my VMs to HVM when specter and meltdown vulnerabilities.

As always, thanks in advance for your time.

Everything clear.
Thanks.

Thanks for your reply, everything clear, I didn't catch it, beeing in the same website I though some kind of special license for XO over XCP-NG and didn't think it was a quote for xcp-ng hypervisor support. Just a couple of questins more, is XO flat price for one pool (any number of hosts in it) or for any number of pools (any number of hosts in each one)?, because I've already have 6 pools, 4 of them only 2 host members each and the others 2 pools are 5 and 6 host members respectively.
Finally, my hardware is 4 years old in some pools, and is not officialy supported in xcp-ng, but working anyway, for instance lenovo x240M5, is this a problem to purchase XO or XCP-NG support?, not beeing a supported HW, I suppose is not possible for XCP-NG, but XO works over XEN API, which is supported depending on the SW version, not HW, but SW is supported depending a HW, so not sure if it's possible to pay for a support in this situation.

Thanks again