XCP-ng 8.3 betas and RCs feedback π
-
@Andrew said in XCP-ng 8.3 beta
:@ravenet I did not have any problems with my Mellanox Connect-4 Lx card during upgrades from 8.2.1 (5.0 driver), to 8.3 (5.4 driver), then update to current 8.3 (5.9 driver), and then update to Xen 4.17.3. My card has the 14.32.1010 firmware. It is not used for the management interface. It is used by VMs and continued to keep the same ethernet name and function for the VMs.
Thanks for testing
I finally got around to updating the firmware on this Mellanox connectx-4 LX adapter. Bit of jumping through hoops to install tools, find proper firmware, but went from FW version14.18.1000 to New FW version: 14.32.1010. Had to do an emergency network reset again, but it's back.
Obviously an incompatibility with the new 5.9 driver with much older mellanox firmware.
-
@ravenet This worries me because I use the Connectx-3 cards in my setup. I wonder if I am going to run into the same problem or is it only a Connectx-4 problem.
-
@Anonabhar I don't think so as the ConnectX-3 is supported only in the older 4.9 drivers and not the newer versions (5.0 or 5.9 or newer). As all of these cards are old and EOL, it would be good to have the latest/last firmware installed anyway.
-
Any reason why xcp-ng is focussing on back porting kernel drivers and fixes to an old kernel instead of using a newer kernel and porting back in what is needed for xcp-ng?
-
@john-c Yeah, tried all of that. I get the same issue with older GPU's when vGPU'ing them. So I'm wondering if it's a setting on the Dell Server BIOS.
-
@xerxist Who says we're not doing both?
-
@Andrew @Anonabhar Indeed, it's not the same driver so the old card will keep using the old driver.
-
@cunrun said in XCP-ng 8.3 beta
:@john-c Yeah, tried all of that. I get the same issue with older GPU's when vGPU'ing them. So I'm wondering if it's a setting on the Dell Server BIOS.
Have you enabled SR-IOV in Integrated Devices on the Dell Server's BIOS?
-
-
Which kernel are you looking at since 4.19 will be EOL in 9 months?
-
I applied the very last batch of XCP-ng 8.3 updates and after a smart reboot (which took forever to complete and did not resume all my VM automatically), I have lost my ISO storage: the folder "/ISO" has been deleted by the update and the fstab has been overwritten. This is becoming annoying, I never ever had to "repair" anything related to my storage in my 20 years of using Vmware Esx but this is already the second time I have to do it in my 2 years of using XCP-ng and the very first time an update deletes a mounting point. I understand 8.3 is still a beta, hence my feedback.
-
@ThierryC01 Why are you using fstab?
-
@Tristis-Oris Well, this is how local disks are mounted right?
-
@ThierryC01 possible way, but not only one. It recomended to configure nothing at dom0.
literally created 2nd local storage with 1 click:
-
@Tristis-Oris Except that your method is to create the SR, mine already exists, is full of .iso files and could be wiped doing your method!!! The SR exists, I can see the list of files that should be there but it is marked as "disconnected".
-
@ThierryC01 iso sr can be mounted same way without wipe. I admit some cases where fstab is required, but not for this.
-
Yeah... point is, the mounting point has been deleted and the fstab overwritten during the updates... as I mentioned in my post above.
-
@ThierryC01 well, such unpredictable thing shouldn't happens.
-
@ThierryC01 I don't see how an update could delete a
/ISOfolder on the system. An upgrade using the ISO, yes, because it actually reinstalls XCP-ng and migrates the configuration it knows about, but not a simpleyum update. What happened exactly? How did you update? -
An update will not overwrite
/etc/fstabeither, or there's a serious packaging bug somewhere. I will do some tests.
