Hi @olivierlambert !

Thanks for your help but we've noticed, that on all of ours CentOS-VMs we haven't start 'xe-linux-distribution.service'.
After doing an 'systemctl enable xe-linux-distribution.service' followed by an 'systemctl start xe-linux-distribution.service' it works fine in XOA.

Thanks a lot for your help.

@olivierlambert said in Get VM from <ID>.vhd from Storage-Record:

But why? What's the purpose? What do you want to achieve in the end?

I wanna know which Disk from which VM needs more space because we've an Disk-Alert on the SR in our Monitoring and I wanna know the exact VM.
But when I hover ofer the Disks from the VMs in XOA I can't find the respective VM.

@olivierlambert said in Get VM from <ID>.vhd from Storage-Record:

Hi,

It's unclear. What do you want to achieve exactly? Functionally speaking.

Hi!

I want to get the respective VM to an VHD (Disk) File. Hops that's clearer...

Hi there!

Our customer creates the Disc-Images of its VMs on the Local Storage Record under /run/sr-mount/<ID>/<ID_of_Disk_Image>.vhd. ...

This path is also mounted in XOA... Now I wanna get the specified VM for the matchig disk-Image. How can I get this with xe-commands?

I can't find the concerned VM to a spcial Disk-Image.

Thanks for your help.

@olivierlambert said in Alert: Control Domain Memory Usage:

In Xen Orchestra, restarting the toolstack is clicking on a button. This won't affect running VMs (except if you are in the middle of an hypervisor operation, like migrating/exporting). Restarting the toolstack won't fundamentally solve your issue.

Adding more memory into the dom0 is a good idea in general (depending on the total memory you have). 8GiB is a good start. This will require a host reboot in any case, to be taken into account.

Hi Olivier, thanks for your fast reply. We've identified on our systems that the Openvswitch needs a lot of memory and therefore we wanna restart it.
Is it possible to do this without restarting the hole host and without an downtime?

@dave said in Alert: Control Domain Memory Usage:

@olivierlambert said in Alert: Control Domain Memory Usage:

Netdata

Hello Oliver,
thanks for your quick Response.

Yes, htop is much prettier, but it doen`t see more the top, at least in this case:

I think, Netdata will also just see those things?

Hi there! Hope it's okay that I answer regarding this topic....

We've also this Error and in 'Notifications' in the XCP-NG Center I can see the following:

Is there any possibility to resolve this error without restarting the hole host? How can I restart the toolstack? Can I do it without an interrupt of the business?
We've also raised up the Memory for the Controller Domain from 4 --> 8GB.....

Hi @olivierlambert !

Thanks for your help but we've noticed, that on all of ours CentOS-VMs we haven't start 'xe-linux-distribution.service'.
After doing an 'systemctl enable xe-linux-distribution.service' followed by an 'systemctl start xe-linux-distribution.service' it works fine in XOA.

Thanks a lot for your help.

@olivierlambert :

Okay sorry, I've re-checked it and it seems that the Xen-Guest-Tools are installed on our CentOS-based-VMs:

[18:00:21][root@admin:~]$yum list installed | grep xen*
kexec-tools.x86_64                     2.0.15-33.el7                  @base     
libXfixes.x86_64                       5.0.3-1.el7                    @base     
xe-guest-utilities-latest.x86_64       7.17.0-2.el7                   @epel     
xe-guest-utilities-xenstore.x86_64     7.0.0-24                       installed 
[18:01:22][root@admin:~]$

It seems that we haven't installed the Xen-Tools in our CentOS-based VMs. It's like as you have discribed it here, right? --> https://xen-orchestra.com/blog/install-xenserver-tools-in-your-vm/

Is there a possibility to reinstall the Xen-Tools without an reboot of the VMs?

Hi there!

We've the problem, that the memory usage wouldn't be displayed on CentOS-VMs in XOA but on Debian and SLES11 based VM's it'll be displayed:

CentOS-VM:

SLES11-VM:

Under the 'Advanced Options' we can see that the 'Xen tools' aren't installed on the CentOS-VMs:

CentOS-VM:

SLES11-VM:

As it's discribed here (Xen-Orchestra Blog) it seems that only Debians Default-Kernel PVHVM Support.

Is it possible to enabled this also on CentOS-VMs to see the Memory-Usage in XOA?

Okay, we've done the Upgrade on every of our XCP-NG Nodes via 'yum update'. So we haven't done it via the ISO and I understood yours statement in the other topic, that the statistics should be on the XCP-NG Nodes anymore.

But is there also an possibility to get the statistics in XOA and to store it for the time 'after the reboot'?

Background is that our customer would like to have this in the XOA at the next update of the nodes, because he only has access to the XOA and not to XCP-NG Center.