Hi All,
Recently got me 4 HGST 12TB drives and configured them in Raid-Z1 on a HP Proliant Microserver Gen8 (I’ll call this system C), with mirrored USB thumbdrives for freenas-boot (version 11.2-u7).
Have been using FreeNAS since 9.2 on two other Gen8 Microservers (called systems A and B, with Raid-Z1 pools of 4x3TB and 4x4TB) but just a ‘user’, definitely no expert.
I wanted to concentrate all data on A and B to C, some 18TB in total, and replicated snapshots of A and B to C, did a rollback of those replicated snapshots and set up necessary users, permissions and shares.
So far all good and all data available on system C.
Then I did something a bit careless. I swapped disks and USB thumbdrives between systems C and B, but forgot to export the pools first. I know, rtfm! But systems C and B are identical hardware so I thought this swap could still turn out ok and the shares did work and all data was available, now on system B.
However when I checked the boot-process something made me nervous.
After the boot-loader, just before the Welcome to FreeNAS screen with the usual boot-options appears the following message came up:
(my datapool is indeed called tank.)
Next the ‘FreeBSD/x86 ZFS enabled bootstrap loader’ continues with the Welcome to FreeNAS boot-screen, after which boot seems to finish normally.
Having found threads that this may be due to failure of the USB thumbdrives I installed a fresh 11.2-U7 copy on two fresh USB thumbdrives.
The same error message however persists and did so even before I imported the ‘tank’ pool (pool tank was exported before the install of fresh bootdrives this time).
(This struck me as odd, at such an early stage the boot process apparently already assessed there is such a pool ‘tank’ and there is something wrong with is, the MOS for whatever that may be?!?!)
Zpool Status however indicates normal operation for both pools freenas-boot and tank, with no known errors.
System dataset is on tank also.
Smart output shows no errors for the new 12TB drives.
From a ‘usage’ standpoint, everything appears to be working normally.
And now (sorry for the somewhat long introduction but wanted to give some background) my question(s):
-Is this error message a threat to current of future data-integrity?
-What caused this error and can I fix it without copying some 18TB now residing on B back to A and C systems and starting from scratch?
(The Microservers all have 16GB ECC memory, Celeron 1610T processors, HP specific motherboards, Broadcom NIC’s and HDD controller … don’t know, in ACHI mode off course.)
Any help is well appreciated;-)
Regards and thanks,
Richard
Recently got me 4 HGST 12TB drives and configured them in Raid-Z1 on a HP Proliant Microserver Gen8 (I’ll call this system C), with mirrored USB thumbdrives for freenas-boot (version 11.2-u7).
Have been using FreeNAS since 9.2 on two other Gen8 Microservers (called systems A and B, with Raid-Z1 pools of 4x3TB and 4x4TB) but just a ‘user’, definitely no expert.
I wanted to concentrate all data on A and B to C, some 18TB in total, and replicated snapshots of A and B to C, did a rollback of those replicated snapshots and set up necessary users, permissions and shares.
So far all good and all data available on system C.
Then I did something a bit careless. I swapped disks and USB thumbdrives between systems C and B, but forgot to export the pools first. I know, rtfm! But systems C and B are identical hardware so I thought this swap could still turn out ok and the shares did work and all data was available, now on system B.
However when I checked the boot-process something made me nervous.
After the boot-loader, just before the Welcome to FreeNAS screen with the usual boot-options appears the following message came up:
Code:
ZFS: i/o error- all block copies unavailable ZFS: can’t read MOS of pool tank
(my datapool is indeed called tank.)
Next the ‘FreeBSD/x86 ZFS enabled bootstrap loader’ continues with the Welcome to FreeNAS boot-screen, after which boot seems to finish normally.
Having found threads that this may be due to failure of the USB thumbdrives I installed a fresh 11.2-U7 copy on two fresh USB thumbdrives.
The same error message however persists and did so even before I imported the ‘tank’ pool (pool tank was exported before the install of fresh bootdrives this time).
(This struck me as odd, at such an early stage the boot process apparently already assessed there is such a pool ‘tank’ and there is something wrong with is, the MOS for whatever that may be?!?!)
Zpool Status however indicates normal operation for both pools freenas-boot and tank, with no known errors.
System dataset is on tank also.
Smart output shows no errors for the new 12TB drives.
From a ‘usage’ standpoint, everything appears to be working normally.
And now (sorry for the somewhat long introduction but wanted to give some background) my question(s):
-Is this error message a threat to current of future data-integrity?
-What caused this error and can I fix it without copying some 18TB now residing on B back to A and C systems and starting from scratch?
(The Microservers all have 16GB ECC memory, Celeron 1610T processors, HP specific motherboards, Broadcom NIC’s and HDD controller … don’t know, in ACHI mode off course.)
Any help is well appreciated;-)
Regards and thanks,
Richard
