Fault tolerance testing continues - Weird Symptoms Prevail

[ATCIS]

Thanks Skywalker!

Awesome reply! :worship:

[ATCIS]

Skywalker wrote:
Diagnostic programs don't pay attention to ATA errors. ATA errors can be caused on some drives by simply sending the drive an invalid command. But we don't expect that to be happening on the ReadyNAS, so ATA error are out of the ordinary. If you have SSH, you can check your SMART data and ATA errors by running these commands: CODE: SELECT ALL smartctl -a smartctl -a /dev/disk/internal/0:5*

I just tried the above and I kept getting an error. I'm pretty sure it was just a typo. Did you mean:

smartctl -a /dev/disk/internal/0:5*

Anyway, it was STILL an awesome reply! Thank you again for stopping by to enlighten us wannabe hackers. . .

[mdgm-ntgr]

Yes it is a typo. Try that.

[Skywalker]

ATCIS wrote:
Did you mean: smartctl -a /dev/disk/internal/0:5*

Yep, that's what I meant. Sorry, copy/paste error on my part. BTW, that command will check SMART data on disk 6. If you want to check disk 1, it would be 0:0:* (the channel numbers are 0-based).

[ATCIS]

Dear Skywalker,

Allow me to thank you one more time for your guidance. . .

However. . . This command:

Skywalker wrote:
From SSH, run this command to reset the error history log: rm -f /var/readynasd/smart.sq3

Is not working for me. I have literally deleted /var/readynasd/smart.sq3 four different times, four different ways, and my ATA Error history still prevails. These are the four ways I've done it:
1.) Simply the way you stated above. . .
2.) Ditto + a reboot
3.) With a SSH-FTP Client
4.) Same as number 1 above, then shutdown, then shuffle all the drives around, and power back up. What I have found is that the ATA Error count follows the physical hard drive not the Bay number. . .

I have confirmed that the file does actually delete. I really think you were on to something, so any other recommendations would also be appreciated.
Perhaps the smart.sq3 does not actually contain the ATA Error history? Just asking. . . :?

Best wishes,

[mdgm-ntgr]

Is there anything in /var/readynasd/db.sq3 (don't delete this database. If there are offending lines they would need to be edited or removed - whichever is appropriate - manually).

[ATCIS]

mdgm wrote:
Is there anything in /var/readynasd/db.sq3 (don't delete this database. If there are offending lines they would need to be edited or removed - whichever is appropriate - manually).

I don't know what the best application would be to do that.

Thank you,

[mdgm-ntgr]

# sqlite3 /var/readynas/db.sq3
# .dump

[ATCIS]

mdgm wrote:
# sqlite3 /var/readynas/db.sq3 # .dump

Thank you.

Results:

SQLite version 3.7.13 2012-06-11 02:05:22
Enter ".help" for instructions
Enter SQL statements terminated with a ";"
sqlite> .dump
Error: unable to open database "/var/readynas/db.sq3": unable to open database file

[mdgm-ntgr]

Oops. Typo.

The file is /var/readynasd/db.sq3

[Skywalker]

The SMART / ATA error history database will get re-created automatically if it doesn't exist. But it will get rid of the previous history, so the disk health checks will operate from a clean slate from that point onward.

[ATCIS]

Skywalker wrote:
The SMART / ATA error history database will get re-created automatically if it doesn't exist. But it will get rid of the previous history, so the disk health checks will operate from a clean slate from that point onward.

Dear Skywalker,

I totally understand your intent with this command:

rm -f /var/readynasd/smart.sq3

. . .and I concur that if that's where the ATA Error Data resides, deleting that file should force the statistics into a "Clean Slate" sort of mode, but in my case it doesn't do squat.
In case you missed my earlier post, it's the one with the four item list. . .

Please do not misunderstand my tone as I am mearly a Padawan and mean no disrespect

Next, I will see if I can figure out what mdgm is asking me to try out. . .

Thanks as always,

[mdgm-ntgr]

Hmmm maybe after removing the smart.sq3 file you need to restart readynasd?

[ATCIS]

mdgm wrote:
The file is /var/readynasd/db.sq3

Dear mdgm,

Okay, I found the data in the .dump that you instructed me about. It reads:

INSERT INTO "$log" VALUES(7,1374018379,'notice','disk','LOGMSG_SMART_ATA_ERR_30DAYS_WARN','<args Count="24638" Channel="6" Model="WDC WD2002FYPS-02W3B0" Serial="WD-WCAVY5956816" Events="3"/>');

So now that I've found it, how do I get rid of it in the db.sq3 file?

Thanks in advance,

[ATCIS]

mdgm wrote:
Hmmm maybe after removing the smart.sq3 file you need to restart readynasd?

Wouldn't restarting the entire NAS basically accomplish the same thing?

Just asking. . .

[Skywalker]

How are you concluding that it didn't clear the history? We'll still read the current information from the drive and re-populate the database. If your ATA error count was high before, it will still be high, because that count is stored in the drive and we can't reset it. However, any alert messages about your ATA error count increasing should go away, unless it continues to increase more than expected.

[ATCIS]

So, I just lost about 15 minutes worth of typing due to this error:

Sorry but the board is temporarily unavailable, please try again in a few minutes.

Anyway, what I think I was saying was:

Skywalker wrote:
How are you concluding that it didn't clear the history?

Because that is what the GUI is telling me. . .

Skywalker wrote:
We'll still read the current information from the drive and re-populate the database. If your ATA error count was high before, it will still be high, because that count is stored in the drive and we can't reset it. However, any alert messages about your ATA error count increasing should go away, unless it continues to increase more than expected.

Okay now I'm confused. . . Earlier you said:

Skywalker wrote:
Diagnostic programs don't pay attention to ATA errors. ATA errors can be caused on some drives by simply sending the drive an invalid command. But we don't expect that to be happening on the ReadyNAS, so ATA error are out of the ordinary.

and. . .

Skywalker wrote:
If you have SSH, you can check your SMART data and ATA errors by running these commands: smartctl -a /dev/disk/internal/0:5*

. . .which I did, and the drive reported that it was perfectly heathy. No errors whatsoever. Hence my confusion. . . :?
If you would like to see the output of the command above, this is what it looks like:
https://dl.dropboxusercontent.com/u/4603388/SmartData.txt

Is there some other stored data on the drive that "smartctl" isn't reporting?
Please clarify, and. . .
Thanks again!

[Iridabike]

I added an additional disk to my Pro6 and the rebuild is taking an age. Anyone got any ideas how I can speed it up?

[mdgm-ntgr]

SSH in and run "top". Are any services hogging resources?

[Iridabike]

Heres what I have running:

[mdgm-ntgr]

Try disabling Plex Media Server and the iTunes (forked-daapd) server and see if the rebuild speeds up.

[Iridabike]

That's reduced it to around 100 hours -Thanks. Still seems a long time though.

[mdgm-ntgr]

Hopefully it will speed up further.

[alltet]

So my ultra2 gave up on me, nothing helped (os re-install/factory default/usb recovery). No contact at all with the NAS. Just starting and fan goes on and off on an interval. Anyhow in case anyone of you in a similar situation this really helped on getting the data back from the drives in x-raid and formatted as btrfs as no app i found could manage.

install ubuntu with widows installer (usb stick install was messy)
http://www.ubuntu.com/download/desktop/install-ubuntu-with-windows

then with all rives connected use terminal:
http://www.readynas.com/forum/viewtopic.php?f=4&t=71423
#sudo su
#apt-get mdadm install
# mdadm --assemble --scan
# cat /proc/mdstat (mine was md127 change if diferent)
# mount -t btrfs -o ro /dev/md127 /mnt

and the drive can be opened from the disk tools mounted as /mnt

[mdgm-ntgr]

Another option would have been to migrate your disks to a new ReadyNAS running ReadyNAS OS.