wrong Pool space reported in WEB UI

[Davide Zanon]

Hi everybody,
I have a RAIDZ1 with 12x3tb (2.73tb reported) disks, for a total of 32.5tb of pool. I don't use snapshots.
Today I noticed that the space reported in the WEB UI is totally wrong, here is what I see:

59,75TiB used, just physically impossible.

This is the "zpool list" output from the console:

I think the problem happened upgrading from FreeNAS 11.3-U5 to TrueNAS Core 12.0U5.1 but I'm not sure 100%,
it just caught my eye today because the Dashboard reported 93% of used space and it didn't look right because
I always keep it checked to stay below 80%.
Now I'm just wildly guessing, but could it be that I'm seeing the space saved by the deduplication summed up there?
Is there a way to know how much space the deduplication is saving right now?
Please let me know what could be the cause, if I did something wrong or went wrong during the upgrade.
I didn't reboot the machine yet because I need to schedule it, if needed I'll try next week.
If you need more informations I'll post them as soon as possible, I'm not used to posting so please forgive me if my
writing is a bit confusing, I searched the forum for the same problem before posting but I couldn't find anything.

Thanks

 

[NugentS]

System Spec?

 

[Davide Zanon]

System Spec?

Hi,
it's a Netgear ReadyData 5200, these are the specs:
MOBO: Supermicro X8SI6-F-NI015
RAM: 32gb non-ecc
CPU: Intel Xeon X3450
SAS CONTROLLER: SASII LSI2008
DISKS: 12x3tb Seagate ST33000650NS (RAIDZ1)

Tell me if I'm missing some information.

Thans

 

[NugentS]

Has the controller been flashed to IT Mode?

Other than that I have no idea I am afraid although something does seem off

 

[c77dk]

Seing the output of 'zpool list' I think your guess about it being caused by deduplication might be spot on - haven't got a system with DD myself, so can't verify. Hopefully someone using it can be of more help.

Just out of curiosity - How performant is that system? Remembering a lot of threads about performance with DD when 12.0 was released, and the consensus seemed to be that performance was _bad_ unless you had a lot of RAM (which you don't) and/or used sVDEV for DDT.

 

[Davide Zanon]

Has the controller been flashed to IT Mode?

Other than that I have no idea I am afraid although something does seem off

No, the controller has not been flashed to IT Mode, but everything was working fine until recently
so I don't think that's the problem (I've been using Freenas on this system for about ~2yrs).

what seems off is the "BackupNew" dataset space usage reported both on WebUI and console,
this is the output of "df -h":

I didn't notice this before but that's the only consistent data between the two, but on WebUI the space
report is incorrect and from the console is right, so I think the former is summing that up and the latter
is not.
Do you think is better to report a bug for this?

Thanks

 

[Davide Zanon]

Seing the output of 'zpool list' I think your guess about it being caused by deduplication might be spot on - haven't got a system with DD myself, so can't verify. Hopefully someone using it can be of more help.

Just out of curiosity - How performant is that system? Remembering a lot of threads about performance with DD when 12.0 was released, and the consensus seemed to be that performance was _bad_ unless you had a lot of RAM (which you don't) and/or used sVDEV for DDT.

Hi,
this is just a backup system with no particular services running, for what it does the performances are ok I think
but I there is always room for improvement, I'd like to add some SSD for L2ARC, SLOG and so on but I'm still
really really new to zfs so I'll have to study a little bit more before making any changes.