Recent content by TheGrandWazoo

Hi, Just fixed this with commit 0b760eefc3ae8787e4d83235ca20e2ae886cf166. In TrueNAS-Scale the slashes (/) needed to be converted to dashes (-) which it was doing but only the first slash (/). Adding the 'g' parameters to the regex substitution fixed this. You may need to 'edit' your extents...

I found that the pvestatd service was not running after a 6.4-13 to 7.0-11 upgrade. Just did a 'pvestatd restart' and the grey questions marks turned to the green icons. Not sure why this happened but it seems fine now.

Ok. Thanks for that. I will give Version 3 a try. Lot of hard work and info in that article but was not sure if it was fixed in any of the kernels Proxmox is using. Thanks for posting it.

James, Are you saying to use version 3 instead of version 4? Both of my nodes are mounted as version 4 via the `mount -v` output.

This machine does have the CVE microcode fixes... [ 22.919602] ------------[ cut here ]------------ [ 22.919604] General protection fault in user access. Non-canonical address? [ 22.919611] WARNING: CPU: 13 PID: 2990 at arch/x86/mm/extable.c:126 ex_handler_uaccess+0x52/0x60 [ 22.919612]...

This is the dmesg output after the server rebooted with another kworker tainted issue. [ 17.464801] ------------[ cut here ]------------ [ 17.464802] General protection fault in user access. Non-canonical address? [ 17.464812] WARNING: CPU: 16 PID: 1582 at arch/x86/mm/extable.c:126...

I just realized that the CPU's in these machines do NOT have the new Microcode applied to them (For the CVE's) . I will attempt to do that update in the near future if this could cause an issue.

Since upgrading from 6.0 to 6.1 I have had two different host machine generate the following similar error. It always had kworker as the issue PID. First system became unresponsive and unrecoverable over time. Tried to reboot the host safely but had to power reset the physical host. This one is...

Just posting another 'Oops' dump. I am hoping it is the same issue. [1406350.728555] BUG: kernel NULL pointer dereference, address: 0000000000000014 [1406350.728598] #PF: supervisor read access in kernel mode [1406350.728624] #PF: error_code(0x0000) - not-present page [1406350.728638] PGD 0...

Thomas, In your post you used... apt update apt full-upgrade I always thought we had to do... apt update apt dist-upgrade When should we use the 'full-upgrade'? Thx, -Waz

I have a few systems that noVNC seems to behave differently between them. If I open a noVNC to a non-running VM of course I get a "Failed to connect to server" but if I use the noVNC commands from the GUI to start it, it will sit and wait for a connection and present me with the logo and...

Thomas, Thank you for the info. Thankfully it was my lab machine (home) and not production/customers machines...always do it in the lab ;). I will mark this post because I have gone through many upgrades from previous versions that might have caused the issue. I think the only thing I install...

Reinstalled from the ISO 5.3-5 and recovered my ZFS and ZPOOL's. Restored the configs and all was fine in the world again. Then, just because, I did the upgrade again to 5.3-11 with that kernel 4.15.18-12 that just jack me and rebooted with no issues. Go figure. Hopefully it was just a fluke...

I wouldn't unless you have a backup plan...right now, I don't except reinstall. Not even the 'rescue' on the ISO helping.

Just did an upgrade to 4.15.18-12-pve which looked like ZFS changes only and now my system boots into a read-only state just after 'cgmanager' starts and that's it. Just a '(none):~#' prompt. I am running ZFS on the server. Scientific Progress Goes Boink. -Waz