TrueNAS 12.0-U7 Released

[Miloman]

Works here, so you may have resolved to the old location. Try an ipconfig /flushdns.

Thanks, works again (without flushing).

 

[Constantin]

Curiously enough, my pool got nuked as part of the upgrade. It's showing now with a white X in a red ball and my only option is to disconnect it.

Time to brush up on the pool re-importation process, and thank goodness there is a auto-save function for the config file as part of the upgrade process.

[EDIT]
Turns out all three sVDEV drives were disconnected per the UI. As if I had pulled their sleds from the backplane. The CLI suggested I destroy the pool and rebuild from backup.

Once I pulled and reinserted the drives from their SATA backplane, all three re-appeared per the UI, and now the pool could be successfully imported. The pool is now 100% functional. Weird!

 

[winnielinnie]

Curiously enough, my pool got nuked as part of the upgrade. It's showing now with a white X in a red ball and my only option is to disconnect it.

Time to brush up on the pool re-importation process, and thank goodness there is a auto-save function for the config file as part of the upgrade process.

[EDIT]
Turns out all three sVDEV drives were disconnected per the UI. As if I had pulled their sleds from the backplane. The CLI suggested I destroy the pool and rebuild from backup.

Once I pulled and reinserted the drives from their SATA backplane, all three re-appeared per the UI, and now the pool could be successfully imported. The pool is now 100% functional. Weird!

This was triggered by the update?

So physically unplugging and re-plugging resolved it? Not even a reboot was enough?

That's scary, and you seem pretty calm and chill for someone who at first believed their pool was nuked.

 

[Constantin]

This was triggered by the update?

So physically unplugging and re-plugging resolved it? Not even a reboot was enough?

As best as I can tell, it was the update. That in turn seems to have borked all three Intel 1.6TB DC S3610 that I use in a 3-way sVDEV mirror at once. Per the UI, they were gone. Restarting didn't help, reverting to 12u6.1, etc. did nothing. See my report here.

So I decided to pull the drives from the hard drive enclosure and re-insert them in other SATA backplane positions. As soon as I did that, the console lit up with its usual torrent of data associated with devices getting added. After that, importing the pool worked just fine. It's now online, no issues.

My ask to iXsystems as a result of this experience is giving users better feedback when a pool import fails. Such as "Hey, your pool import failed.

• The following drives are present.
  • Drive Type/ Role | VDEV | SATA PORT | Capacity | Serial Number
  • xxxxx
• The following drives are missing:
  • Drive Type/ Role | VDEV | SATA PORT | Capacity | Serial Number
  • xxxxx
• The following drives are reporting they have failed (using SMART data)
  • Drive Type/ Role | VDEV | SATA PORT | Capacity | Serial Number
  • xxxxx"

Should not be insurmountable and a better feedback than "your pool needs to be destroyed and rebuilt from backups"

As @jgreco noted, sometimes a hard power cycle is necessary where a restart does not suffice. My guess it has to do with the firmware in the SSDs since all other SSDs in my system restarted fine, ditto the HDDs.

 

[SKova]

Before updating (with U6) the graphs worked correctly with chrome 96, now with U7 are broken.

I noticed the same thing, immedidately after the update, they stopped working in Brave where they were working before.

 

[Constantin]

Same issue here. Curiously, the dashboard still works. Different kind of graphs, I guess?

 

[winnielinnie]

Same issue here. Curiously, the dashboard still works. Different kind of graphs, I guess?

TrueNAS12.0-U7 GUI reporting broken in Chrome Again

TrueNAS12.0-U6.1 Reporting GUI page works fine in my Chrome Browser. However, it shows blank after I upgrade to 12.0-U7. 12.0-U7 Reporting GUI page only works in Safari. I knew you recently change the reporting GUI in 12.0-U7. That may cause this issue. I have attached the picture with all...

www.truenas.com

 

[SKova]

I noticed the same thing, immedidately after the update, they stopped working in Brave where they were working before.

Add that Brave on iOS is still working correctly, just not in the Windows version.

 

[Ericloewe]

Add that Brave on iOS is still working correctly, just not in the Windows version.

Brave, like all browsers on iOS, is just a skin on top of Safari. On other platforms, it's glorified Chromium.

 

[Constantin]

Brave, like all browsers on iOS, is just a skin on top of Safari. On other platforms, it's glorified Chromium.

As I understand it, Braves business model is not as intrusive as Google’s? It’s the primary reason I tend to use Brave for casual browsing.

 

[SeaFox]

As I understand it, Braves business model is not as intrusive as Google’s? It’s the primary reason I tend to use Brave for casual browsing.

Chromium != Google Chrome

Chrome vs. Chromium: What's the Difference?

Chrome and Chromium are similar web browsers, but they have some important differences. So which one should you use: Chrome or Chromium?

www.lifewire.com

 

[Constantin]

Thank you for that. Fwiw, brave seems to be built on chromium and seems to feature auto-updates. To me, that’s a pretty good combination since flash support was never my thing, ditto tracking by google.

 

[Samuel Tai]

Looking at the iX Jira, 12.0-U7.1 seems to be mostly feature complete, and will likely drop after the New Year.
These are the patches currently slated to be shipped:

• NAS-113985 - Merge OpenZFS 2.0.7
• NAS-113925 - Provide correct file generation number
• NAS-113863 - Samba Kerberos authentication fails in MIT realms since 12.0-U6.1
• NAS-113814 - Graphs are empty
• NAS-113813 - Update plugin artifact before executing pre-upgrade script
• NAS-113811 - TrueNAS can offer to offer to use HDDs of one zpool in creating another!
• NAS-113799 - 12.0U6 SMBD.CORE File Found
• NAS-113751 - httpd.core after upgrade to 12.U7
• NAS-113744 - regression in hook_setup_ha on CORE
• NAS-113621 - smbd assertion - failure to chdir() to share connectpath when session has multiple tcons with different credentials
• NAS-113545 - Issues with SSL & Certificates
• NAS-113393 - crash during snapshot enumeration in zfs_fsrvp - regression in port to samba 4.13
• NAS-113368 - Jail stopped working after upgrade to 12.0-U6.1 with utf-8 decoding error
• NAS-113356 - intermittent smbd crash during session logoff
• NAS-113240 - smbd crashes while freeing tree connection if user can't chdir() into connectpath
• NAS-106633 - Cron tasks are run on wrong time zone after initial setup

 

[winnielinnie]

NAS-113811 - TrueNAS can offer to offer to use HDDs of one zpool in creating another!

Now THAT is a scary bug.

 

[rvassar]

Now THAT is a scary bug.

"Hardware eventually fails. Software eventually works." - Michael Hartung

 

[Forza]

Looking at the iX Jira, 12.0-U7.1 seems to be mostly feature complete, and will likely drop after the New Year.
These are the patches currently slated to be shipped:

• NAS-113985 - Merge OpenZFS 2.0.7
• NAS-113925 - Provide correct file generation number
• NAS-113863 - Samba Kerberos authentication fails in MIT realms since 12.0-U6.1
• NAS-113814 - Graphs are empty
• NAS-113813 - Update plugin artifact before executing pre-upgrade script
• NAS-113811 - TrueNAS can offer to offer to use HDDs of one zpool in creating another!
• NAS-113799 - 12.0U6 SMBD.CORE File Found
• NAS-113751 - httpd.core after upgrade to 12.U7
• NAS-113744 - regression in hook_setup_ha on CORE
• NAS-113621 - smbd assertion - failure to chdir() to share connectpath when session has multiple tcons with different credentials
• NAS-113545 - Issues with SSL & Certificates
• NAS-113393 - crash during snapshot enumeration in zfs_fsrvp - regression in port to samba 4.13
• NAS-113368 - Jail stopped working after upgrade to 12.0-U6.1 with utf-8 decoding error
• NAS-113356 - intermittent smbd crash during session logoff
• NAS-113240 - smbd crashes while freeing tree connection if user can't chdir() into connectpath
• NAS-106633 - Cron tasks are run on wrong time zone after initial setup

Thanks for the list. Shame the FTP timout issue isn't fixed. Saw it is planned for TrueNAS 13. Any idea when that might be out?

 

[NASbox]

I see U7.1 seems to have disappeared from Jira and a U8 has appeared with no release date. Are they going to release a U8, or are we waiting for v13?

 

[Patrick M. Hausen]

Pending 12.0-U7.1 bumped to 12.0-U8 it seems

Hi all, I had a quick look at the state of U7.1 and found that the release has been deleted and all the issues moved to U8. With one open issue, release might happen soon(ish). Breaks the tradition :wink: https://jira.ixsystems.com/projects/NAS/versions/13801

www.truenas.com

 

[NASbox]

Pending 12.0-U7.1 bumped to 12.0-U8 it seems

Hi all, I had a quick look at the state of U7.1 and found that the release has been deleted and all the issues moved to U8. With one open issue, release might happen soon(ish). Breaks the tradition :wink: https://jira.ixsystems.com/projects/NAS/versions/13801

www.truenas.com

Thanks!

 

[morganL]

As best as I can tell, it was the update. That in turn seems to have borked all three Intel 1.6TB DC S3610 that I use in a 3-way sVDEV mirror at once. Per the UI, they were gone. Restarting didn't help, reverting to 12u6.1, etc. did nothing. See my report here.

So I decided to pull the drives from the hard drive enclosure and re-insert them in other SATA backplane positions. As soon as I did that, the console lit up with its usual torrent of data associated with devices getting added. After that, importing the pool worked just fine. It's now online, no issues.

My ask to iXsystems as a result of this experience is giving users better feedback when a pool import fails. Such as "Hey, your pool import failed.

• The following drives are present.
  • Drive Type/ Role | VDEV | SATA PORT | Capacity | Serial Number
  • xxxxx
• The following drives are missing:
  • Drive Type/ Role | VDEV | SATA PORT | Capacity | Serial Number
  • xxxxx
• The following drives are reporting they have failed (using SMART data)
  • Drive Type/ Role | VDEV | SATA PORT | Capacity | Serial Number
  • xxxxx"

Should not be insurmountable and a better feedback than "your pool needs to be destroyed and rebuilt from backups"

As @jgreco noted, sometimes a hard power cycle is necessary where a restart does not suffice. My guess it has to do with the firmware in the SSDs since all other SSDs in my system restarted fine, ditto the HDDs.

Thanks Constantin,

Could you write this up as a "suggestion" in jira (Report-a-bug).
My guess is that the error message comes from the ZFS layer... and doesn't understand "missing devices".