@thomas-wood

you might have a look at this topic:
nfs sr connection problem

There we also discussed a NFS problem between XOA and TN Core. Maybe there is something in as a hint in the right direction for you.

@manilx said in Passthru of Graphics card:
... and need the HDMI port for display as I have a piKVM hooked up for remote interaction.....

Before I forget:
You could use an DP to HDMI cable for that, but given the other things we talked about, this might also only be a crutch.

@manilx

Yeah, it's a bummer that active PCI/USB passthrough blocks snapshots/backups.

I understand that it is a problem for the migration since there is no guarantee that the device is also available on the other host, but for snapshot/backup it should not be of importance

@manilx

I was going through the same with my Protectli devices and found the following, which also might apply to your situation:

• When enabling passthrough for the IrisXE graphic, the HDMI port showed no picture anymore - BUT - then DP port still shows everything, even with passthrough enabled

My Protectli uses openboot as BIOS, so this might make a difference, so better mentioning it.

@manilx

True, not a big deal, but this two host were fresh installs with 8.3 release and only joint the pool to keep everything running.

So this seems to be a thing on pool level, since on those hosts only the Legrand was ever attached to.

Hi,

at the moment I'm a bit confused abotu the handling of USB Devices. And I'm not sure that I picked the right category for it, so please move if it is wrong here.

I have 2 identical hosts (8.3 release), both have 1 USB device attached - an Legrand USV.
Since it looks identical on both hosts, I list the details only for one of them.

In XO only the one USB device is listed:

On the host only the one device is shown via lsusb :

[22:51 localhost ~]# lsusb
Bus 004 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 003 Device 002: ID 1cb0:0038  
Bus 003 Device 003: ID 8087:0032 Intel Corp. 
Bus 003 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
[22:51 localhost ~]# 

When I list PUSB device things get more interesting:

[22:51 localhost ~]# xe pusb-list
uuid ( RO)            : 295f4189-f4d0-8b6a-3bbc-c6283369c493
            path ( RO): 3-6
       vendor-id ( RO): 1cb0
     vendor-desc ( RO): 
      product-id ( RO): 0038
    product-desc ( RO): 
          serial ( RO): 
         version ( RO): 2.00
     description ( RO): 
           speed ( RO): 12.000


uuid ( RO)            : 76fce6a5-7658-0c8b-59fb-73eb32bf9a02
            path ( RO): 2-2.1
       vendor-id ( RO): 0bc2
     vendor-desc ( RO): Seagate RSS LLC
      product-id ( RO): ab28
    product-desc ( RO): 
          serial ( RO): NA7PACFG
         version ( RO): 3.00
     description ( RO): Seagate RSS LLC_NA7PACFG
           speed ( RO): 5000.000


uuid ( RO)            : 936340fa-45c2-24d3-8371-5929d352e284
            path ( RO): 1-2.1
       vendor-id ( RO): 1cb0
     vendor-desc ( RO): 
      product-id ( RO): 0038
    product-desc ( RO): 
          serial ( RO): 
         version ( RO): 2.00
     description ( RO): 
           speed ( RO): 12.000


uuid ( RO)            : bc0eb774-3163-1cf6-d6f8-433549565780
            path ( RO): 3-6
       vendor-id ( RO): 1cb0
     vendor-desc ( RO): 
      product-id ( RO): 0038
    product-desc ( RO): 
          serial ( RO): 
         version ( RO): 2.00
     description ( RO): 
           speed ( RO): 12.000


uuid ( RO)            : 1cf04699-233c-e6f1-10b7-b889b016f469
            path ( RO): 1-2.1
       vendor-id ( RO): 1cb0
     vendor-desc ( RO): 
      product-id ( RO): 0038
    product-desc ( RO): 
          serial ( RO): 
         version ( RO): 2.00
     description ( RO): 
           speed ( RO): 12.000


uuid ( RO)            : 36432077-1c2a-5ade-c931-13e2b92593e5
            path ( RO): 3-6
       vendor-id ( RO): 1cb0
     vendor-desc ( RO): 
      product-id ( RO): 0038
    product-desc ( RO): 
          serial ( RO): 
         version ( RO): 2.00
     description ( RO): 
           speed ( RO): 12.000


uuid ( RO)            : 79276bcc-5143-721d-3b37-7c1cf6dec4ad
            path ( RO): 1-2
       vendor-id ( RO): 0627
     vendor-desc ( RO): Adomax Technology Co., Ltd
      product-id ( RO): 0001
    product-desc ( RO): 
          serial ( RO): 42
         version ( RO): 2.00
     description ( RO): Adomax Technology Co., Ltd_42
           speed ( RO): 12.000


[22:53 localhost ~]#

Yes, in the past I had a local Seagate HDD attached, but it was removed as stated in the documentation. So idea why it is still listed.

And when I try to create the VUSB for a VM I get this whole list of device instead of only the one attached to the host.

Where is my brain going the wrong way?

@stormi said in XCP-ng 8.3 betas and RCs feedback :

At some point, the installer needs to contact the pool master during the upgrade.

Good to know that - never had to reinstall yet. Thanks for that info stormi

Just to give some short feedback about upgrade from 8.3 beta to 8.3rc2 via ISO:

1. Pool master upgraded fine after selecting "Upgrade existing installation"
2. 2nd host in pool asked for management network configuration after selecting "Upgrade existing installation", but after that everything upgraded fine
3. Both host show "cloud" instead a build number after the upgrade
   
4. Guest tools version number of/reseted to .2, but this is already discussed earlier

All in all very smooth upgrade and everything is running as before - great work from the Vates team !

@steff22

Yes, that sounds like generally faulty unit not only the CPU related stuff.

@manilx

To be fair the MS-01 is also available with 12th gen Intel CPU.

@bloodyskullz

If you still see the old ISO SR, the easiest way to migrate is simply by creating a new one and migrate the ISOs through XO to the new one.
In regards to deletion of the old SR you need to check if it really is mapped to another drive or if the mapping was not working and it filled / if so, you might not be able to delete it

I suggest in general to use an external share for ISO storage. You don't need them that often and the speed over network if enough for that.

@bloodyskullz
Your root (/) is full, so yum is right saying no space.

Did you install extra stuff? You need to check if there is anything in / you can safely delete

@rporter

Are there really no disks or just not under the expected name /dev/sd....

You might check if there are /dev/xvd.... devices. This change can happen because XCP-NG and VMware handle the "hardware" differently. Had the same problem with some VMs and it also changed

This effect also can change the network interface name, so you might check that too.

I had the same problem with the FreePBX appliance VM - in the end I had to build it new, because I found no way to fix it.

I hope you have a recent backup of the configuration and settings of FreePBX, with that is is not to hard to be be back up and running.

@planedrop
I usually don't use SMB for remotes - prefer for stability reasons NFS.
But in the past on customer systems we had, from time to time, the problem that the SMB Client/Server (both had the problem, so it might be the case that they stalled each other) processes stalled for a moment, which might cause that file locks are not set or ended correctly.

Maybe it's something like this in your case too.
Catching that is a bit of a pain in the *** , because you need to do process memory tracking on both sides to see if and when they stall for a short moment.

Not sure if there is a way under TrueNAS to check if a file has a lock set or not. If possible it might be already enough to remove the lock of the file (if possible via cli) to make it visible to the XCP-NG host again.

@manilx

Could this be the grub menu and it's wait timer? I had it once that I did no saw grub but the system bootet fine. In the end grub was the black screen and depending on the wait interval it stayed black.

This might be a thing with how the console handles image grabbing during boot

@olivierlambert

Installed a new 22.04.2 Server in default installation and removed cloud-init. Hitting enter in grub to login screen 6 sec. on the same host as noted above

@olivierlambert said in VM’s Slow booting:

Ubuntu server or desktop?

Sorry, I should have added that. Ubuntu Server it is. A short stop is always on Cloud Init - 2-4 sec.

@olivierlambert

To give one datapoint here.

I booted a Ubuntu 22.04.2 LTS VM using bios.

From grub hitting enter to boot to the login 11 sec.

Host: AMD Ryzen 7 7840HS 64GB RAM
VM VDI is on an NFS share connected through 2,5Gb Ethernet