Freenas 8.2 (final) Serial shown as "Unknown" after reboot, always need to reimport..

Status
Not open for further replies.

kaspa

Cadet
Joined
Jul 20, 2012
Messages
4
hello,
after upgrading my nas to 8.2 (final), i always need to export the volume and auto import it back (both from GUI) after rebooting my nas.
with version 8.04 i used before, i didn't had this problem. so after rebooting, my hdd was "healthy" and not unknown and my files accessable.

http://support.freenas.org/ticket/860 helped me with the "unknown"-problem...

how do i get my nas to save this settings to survive a reboot, without an export and import of my volumes?

thanks for answers.

kaspa
 

William Grzybowski

Wizard
iXsystems
Joined
May 27, 2011
Messages
1,754
Its a bug :(

Expect a 8.2.0-p1 pretty soon

Workaround:

# mount -uw /
# grep -v "# BEFORE" /etc/rc.d/ix-sysctl > /conf/base/etc/rc.d/ix-sysctl
# mount -ur /
 
J

jpaetzel

Guest
I've replaced the 8.2.0-RELEASE files on SourceForge with 8.2.0-RELEASE-p1 which fixes this issue.
 

trevize

Cadet
Joined
Jul 6, 2012
Messages
5
I had the same issue. SO I did what William Grzybowski suggested.

Then I saw jpaetzel RELEASE-p1. when I try and update 8.2.0-RELEASE to 8.2.0-RELEASE-P1 I get:

Jul 20 19:51:29 freenas manage.py: [middleware.exceptions:38] [MiddlewareError: The firmware is does not meet the pre-install criteria: Command '['bin/install_worker.sh', 'pre-install']' returned non-zero exit status 1]
 

freshfeesh

Explorer
Joined
Oct 10, 2011
Messages
72
'Unrecoverable Error' also part of the standard symptoms?

I'm now having this problem on a 1 week old 3 disc RaidZ array of new discs, on a fresh install of 8.2.0-RELEASE-p1 (not an upgrade). Is everyone else getting this full message?:

The volume Z1 (ZFS) status is UNKNOWN: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected.Determine if the device needs to be replaced, and clear the errors using 'zpool clear' or replace the device with 'zpool replace'.​

photo.jpg
(Meanwhile, you can see that the status of volumes in the main volumes page background shows the array as healthy. All the individual discs show as healthy in "Volume Status" as well.)

It's that "unrecoverable error" part that has me most worried. Getting this box together and running has been a very long background project, and I was quite excited to get the RaidZ array up and shared and my data transfered over the network onto what I thought would be a sorely needed, high reliability solution. So this is frustrating. Anyway, how can I locate the file(s) in which the error occured? I don't want the "clear" the error in case that would mean that I won't be able to locate the corrupted file(s), and I'm not going to trust anymore files to this system until I can at least sort that out.

If this helps, here is some detail of the events leading up to the error: Before adding the new discs to the array, I ran WD's "extened" test on each individual, unformatted disc under Windows 7 x64. After running the new install for about a week, in which I put about 2TB onto a (3) 3TB disc RaidZ array, I powered it down and added an additional Intel (on forum recommendation) PCIe NIC alongside my existing DLink PCI NIC (Marvell chipset). I left the Dlink in for possible future link aggregation, but for now I only have garden variety ethernet, so plugged that cable into the Intel NIC and booted back up. I was watching the boot sequence on a monitor, and after it got near the end, some error messages came up too quick to read, and the machine rebooted. The second reboot was stable, but I couldn't get Freenas to recognize a link on the new card, working on a physical keyboard and monitor on the machine. I moved the ethernet cable into the Dlink card, and got back online on that. This is on an old AMD S939 CPU and mobo with 1GB of RAM. Low RAM, I know, but I don't care about the performance nearly as much as the touted reliability of ZFS, and my research on the forums turned up a number of people running small arrays on low RAM.

I have 2GB of brand new RAM that just arrived that I ordered right after the new Intel NIC. I'm going to want to fix the NIC hardware problem before installing new RAM, and I need to fix the ZFS problem before dealing with the NIC problem.
 

cyberjock

Inactive Account
Joined
Mar 25, 2012
Messages
19,525
If you are running RAIDZ you need a minimum of 6GB of RAM. This is clearly mentioned in the manual as well as the first sticky in the FreeNAS 4 Noobs forum. It is also mentioned all over the forums by other people that have tried to setup a server without doing the proper research. 1GB is the absolute minimum to boot FreeNAS with no ZFS support.
 

freshfeesh

Explorer
Joined
Oct 10, 2011
Messages
72
If you are running RAIDZ you need a minimum of 6GB of RAM

Understood. I've read that. Right below the 6GB min mention, the manual states simply that prefetching is disabled with less than 4GB of usable RAM, i.e. there is recognition that some systems have less. I've also read real examples of people having success with much less. I understand that this is a venture into unsupported waters, but I want to make us of the hardware that I have, which supports 4GB max. The facts in my case are that my install was successful without tweaking, and my RaidZ share worked on my network with modest speeds under greater load than it is likely to experience in normal home use. Then I shut down and got an error that a lot of other people are experiencing. Maybe in my case it's because of my RAM and no one can help me, but I suspect not.

I don't want to hijack this thread with talk of my RAM, but if you think that my low RAM may actually be the cause of this error in my case, for specific reasons, I'd love to know that, and I'd love to know whether other people are getting an "unrecoverable error" alongside the "status is UNKNOWN" message.

Thanks.
 

paleoN

Wizard
Joined
Apr 22, 2012
Messages
1,402

freshfeesh

Explorer
Joined
Oct 10, 2011
Messages
72
My point was, I don't want there to be the appearance that I'm soliciting advice on the suitability of my installed amount of RAM for running ZFS in Freenas. I have started such a thread, and I'm not interested in revisiting the topic.

I'm experiencing the problem that is the topic of this thread. I listed a lot of detail about my system, including amount of RAM, to aid in diagnosis of the problem. I also had a couple specific questions, the most important being wanting to know whether others have had "unrecoverable error"s in conjunction with the "Unknown" after reboot error. I'm still quite interested in the answer to that, as it will help me determine whether I have one problem only, or one problem on top of another.
 

paleoN

Wizard
Joined
Apr 22, 2012
Messages
1,402
I'm not interested in revisiting the topic.
OK.

I'm experiencing the problem that is the topic of this thread.
The volume Z1 (ZFS) status is UNKNOWN: One or more devices has experienced an unrecoverable error. An attempt was made to correct the error. Applications are unaffected.Determine if the device needs to be replaced, and clear the errors using 'zpool clear' or replace the device with 'zpool replace'.​
Actually you're not. You have a different error. From an SSH session as root.

Code:
zpool status -v
 

freshfeesh

Explorer
Joined
Oct 10, 2011
Messages
72
OK, thanks. Now my box isn't accessible using either DHCP or a manual address, so I'll have to try "zpool status -v" later. Looks like I have a few new posts in my future.
 
Status
Not open for further replies.
Top