11.2-RELEASE-U1 won't install

[melloa]

Specs

Motherboard: X9DRi-LN4F+
CPU: 2 x Intel(R) Xeon(R) CPU E5-2650 v2 @ 2.60GHz
Memory = 191126044672 (182272 MB)
Backplane: BPN-SAS-846A
cxgbc0: <Chelsio T320, 2 ports>

Install from virtual ipmi storage

- Install will boot and stop after displaying for all disks:

Code:

GEOM: daXX: the primary GPT table is corrupt or invalid.
GEOM: daXX: using the secondary instead -- recovery strongly advised.

- Boot with gparted live, deleted all partitions, recreated -> Same behaviour

- Downloaded and installed FreeNAS 11.1-U6.3 -> No issues during installation

- dmesg still shows the GPT error above

- Pool doesn't show any errors

Code:

root@mellonas:~ # zpool status
  pool: freenas-boot
 state: ONLINE
  scan: none requested
config:

    NAME        STATE     READ WRITE CKSUM
    freenas-boot  ONLINE       0     0     0
      ada0p2    ONLINE       0     0     0

errors: No known data errors

  pool: raid
 state: ONLINE
  scan: none requested
config:

    NAME                                            STATE     READ WRITE CKSUM
    raid                                            ONLINE       0     0     0
      raidz3-0                                      ONLINE       0     0     0
        gptid/2ba49b8a-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2c7b3306-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2d4f6812-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2e382afe-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2f0c7ec9-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2fe94dc7-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/30bb606e-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
      raidz3-1                                      ONLINE       0     0     0
        gptid/31aaf28b-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/32875852-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/336ed487-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/344b98c8-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/35225ea8-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/35fc0c8e-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/36e8d960-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0

errors: No known data errors

- gparted shows:

Code:

root@mellonas:~ # gpart show ada0
=>        40  1875384928  ada0  GPT  (894G)
          40        1024     1  bios-boot  (512K)
        1064  1875383896     2  freebsd-zfs  (894G)
  1875384960           8        - free -  (4.0K)

Any suggestions?

 

[Spearfoot]

Specs

Motherboard: X9DRi-LN4F+
CPU: 2 x Intel(R) Xeon(R) CPU E5-2650 v2 @ 2.60GHz
Memory = 191126044672 (182272 MB)
Backplane: BPN-SAS-846A
cxgbc0: <Chelsio T320, 2 ports>

Install from virtual ipmi storage

- Install will boot and stop after displaying for all disks:

Code:

GEOM: daXX: the primary GPT table is corrupt or invalid.
GEOM: daXX: using the secondary instead -- recovery strongly advised.

- Boot with gparted live, deleted all partitions, recreated -> Same behaviour

- Downloaded and installed FreeNAS 11.1-U6.3 -> No issues during installation

- dmesg still shows the GPT error above

- Pool doesn't show any errors

Code:

root@mellonas:~ # zpool status
  pool: freenas-boot
state: ONLINE
  scan: none requested
config:

    NAME        STATE     READ WRITE CKSUM
    freenas-boot  ONLINE       0     0     0
      ada0p2    ONLINE       0     0     0

errors: No known data errors

  pool: raid
state: ONLINE
  scan: none requested
config:

    NAME                                            STATE     READ WRITE CKSUM
    raid                                            ONLINE       0     0     0
      raidz3-0                                      ONLINE       0     0     0
        gptid/2ba49b8a-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2c7b3306-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2d4f6812-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2e382afe-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2f0c7ec9-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/2fe94dc7-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/30bb606e-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
      raidz3-1                                      ONLINE       0     0     0
        gptid/31aaf28b-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/32875852-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/336ed487-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/344b98c8-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/35225ea8-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/35fc0c8e-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0
        gptid/36e8d960-1182-11e9-b9c3-00074307cb17  ONLINE       0     0     0

errors: No known data errors

- gparted shows:

Code:

root@mellonas:~ # gpart show ada0
=>        40  1875384928  ada0  GPT  (894G)
          40        1024     1  bios-boot  (512K)
        1064  1875383896     2  freebsd-zfs  (894G)
  1875384960           8        - free -  (4.0K)

Any suggestions?

User @SMnasMAN had developed a pretty good case that the combination of E5-26xx v2 CPUs running in SM X9DRi/X9DR3 motherboards is incompatible with FreeBSD/FreeNAS:

https://forums.freenas.org/index.ph...r3-ln4f-board-or-similar-x9-supermicro.72608/

I'm watching developments with avid interest, because my main system contains a pair of E5-2660 v1 CPUs installed in an X9DRi-LN4F board. FWIW, @SMnasMAN has found that the older v1 CPU seems to work okay. Maybe I'll be lucky down the road.

For now, I'm planning on staying w/ the 11.1 train. It seems to be stable and reliable.

Good luck!

 

[SMnasMAN]

- Install will boot and stop after displaying for all disks:

hi melloa, (this is interesting) -- when you say "will boot and STOP after displaying for all disks" , what do you mean by STOP? does the console (ie vga output from board, or ipmi vKVM) show a kernel panic? curious as spearfoot stated, im having the same issue, with pretty much the exact same HW (only with FN 11.2) when using v2 cpus.

if you need to get FN up and running asap, i suggest you use 11.1 (current release). i have not had any issues with 11.1 and v2 CPUs. (or if you want 11.2, then you need to use v1 CPUs for now).
see my bug report (or thread spearfoot linked to):
https://redmine.ixsystems.com/issues/66130

 

[melloa]

Guys,

User @SMnasMAN had developed a pretty good case that the combination of E5-26xx v2 CPUs running in SM X9DRi/X9DR3 motherboards is incompatible with FreeBSD/FreeNAS:

That server was on FreeBSD and the version was upgraded from 11.2 to 12.0.

hi melloa, (this is interesting) -- when you say "will boot and STOP after displaying for all disks" , what do you mean by STOP? does the console (ie vga output from board, or ipmi vKVM) show a kernel panic? curious as spearfoot stated, im having the same issue, with pretty much the exact same HW (only with FN 11.2) when using v2 cpus.

From ipmi I see it booting and after shows the last disk, it seats there, waited for a good 30 minutes, no error messages displayed. Tried twice ...

For now, I'm planning on staying w/ the 11.1 train. It seems to be stable and reliable.

I'm just testing the new version and took the opportunity to rearrange the pool. It is running 11.1-U6.3, but as it is the back-up server, I can try to install for testings again on a USB stick, and capture the screens via ipmi, but will need to wait rsync finishes.

 

[melloa]

Something funny just happened. It installed.

What have changed?

- Installed 11.1-U6, created a new pool. Maybe there was a problem with the disks that was impacting the installer to finish booting (?).

Anyway, it is installed on a E5-2650 v2, so can't duplicate the v2 issue.

 

[SMnasMAN]

can you test IO to your disks? i can get 11.2 installed and up but the issue is always related to IO to a pool (ie an issue with ZFS is what ix bug devs think).

also can you say what supermicro BIOS version you are running? (will be either 3.1 3.2 or 3.3 , you can get this from IPMIview/web or during boot up ofcourse)
tks

 

[melloa]

can you test IO to your disks? i can get 11.2 installed and up but the issue is always related to IO to a pool (ie an issue with ZFS is what ix bug devs think)

The server is in prod now, so I wont be able to do any tests. I do think is HBA or Disk related as the issue disappeared after the pool was created under 11.1

also can you say what supermicro BIOS version you are running? (will be either 3.1 3.2 or 3.3 , you can get this from IPMIview/web or during boot up ofcourse)

 

[SMnasMAN]

ok thanks for the info. So to be clear you are running 11.2 in production/ going forward, right?
(im pretty sure answer is yes, just want to be sure).

also can you tell which HBA you are using to connect to your disks?
(i cant find that info in your posts nor sig)

btw, (w 2x v2 cpus) i did also try bios 3.2 , and am still having same panic at same spot (as bios 3.3), when booting 11.2.

No problems at all with 11.1 with either bios 3.3 nor 3.2.

thanks!

 

[melloa]

ok thanks for the info. So to be clear you are running 11.2 in production/ going forward, right?

Yes, on the back-up server.