TS-EC879u-RP with FW 3.5.1.1002 Unstable

[StarrCompany]

Sorry this is really beyond frustrating - copy of email sent to QNAP support:

Spoke to you earlier and you remoted on to our QNAP 879 re random reboots.

Conclusion was that the fact we had the EXT 4 cache enabled could be an issue so we disabled.

I can confirm that our QNAP 879 just rebooted so I guess its not that !!!

On the basis of this forum viewtopic.php?f=25&t=52191&start=15 where the number of incidents is increasing can you confirm the following:

What steps are being taken to resolve this ?
Is this device fit for the purpose it was sold ?

If the issue is created by our environment or those experiencing the same issues in the forum above - please can you tell us what configuration changes need to be made
to make the QNAP TS-879 work in our environment?

I don't think its unreasonable to expect a response within 24 hours of this email.

[patriots]

Dude, I think the QNAP's Flux Capacitor either got corrupted or crashed... Please verify that you have your "Mr. Fusion" properly installed and power is being generated. Then check to see if the "Time Circuit" is functioning properly, set the date before the time you purchased your 879 unit...

-Doc Brown.

[gm85]

I shut the system down tonight and applied the BIOS update. I'll keep an eye on it over the next few weeks to see if it restarts.

As it's been happening once a week, I'll keep my fingers crossed that this weekend everything operates as expected.

Update: System just crashed an hour after the update, so it didn't resolve the issue

[scr3w3d]

Sorry to be the bearer of bad news guys but don't expect this product to be production ready. I'm the "proud" owner of a TS-809U-RP for my home virtualization environment. The Qnap is a mildly decent if not frustrating piece of kit for that purpose. Now I can't speak for the 879's but but I would NEVER put an 809 into production, MAYBE test or validation, that's all.

Some rules I've come to follow:

1.) BACKUP! Your config and data religiously.--see subsequent rules.

2.) Use enterprise drives. I loaded consumer grade "aaproved" drives and spent a month figuring out mdadm commands to fix qnaps awful software RAID, which is most likely the casue of your problem even though you have some pretty nifty drives in there, one small hiccup casues a lot of reverberation in the QNAP software raid world, believe me. In QNAPs defense however software RAID in general is terrible, in my research I learned a lot of the fixes from intel forums where people were trying to fix intels implementation of software raid and had very similar issues to the QNAP software raid problems. google qnap drive dropped or qnap raid for more info.

3.) NEVER load a new firmware, there are bound to be bugs, I'm talking starship trooper size bugs. As an example the apache.conf on my 809 was just destroyed by FW 3.5.2, looks like my wifes gonna have to live without her gallery page for a while. lol. I actiualy found this post in the search to fix apache. double lol. BTW, If no bugs then you stand a decent chance of loosing your entire configuration along with your data.

4.) Make sacrifices to the NAS god's, it will be about as useful as QNAP support, GUARANTEED!


BTW QNAP supports default fix for all issues usually comrpise 2 options 1.) RMA's or 2.) "REMOVE ALL THE DRIVES AND RESET". These options rarely work.
Hope this helps.

[kevinliao]

I shut the system down tonight and applied the BIOS update. I'll keep an eye on it over the next few weeks to see if it restarts.

As it's been happening once a week, I'll keep my fingers crossed that this weekend everything operates as expected.

Update: System just crashed an hour after the update, so it didn't resolve the issue

Hi gm85,

Which version of BIOS you got from customer support? As far as I know, there are some power management settings in BIOS that may be related with the random reboot issue. Could you restart the system then go to BIOS screen by pressing DEL key and help to check:

1. In the Main screen, what are the "Project Version" and "Build Date and Time"?
2. In Advanced screen -> CPU Configuration: Can you see "Power Technology"? If yes, what is the setting?
3. In Advanced screen -> SATA Configuration: Can you see "Aggressive Link Power Management"? If yes, what is the setting?

[gm85]

Hi gm85,

Which version of BIOS you got from customer support? As far as I know, there are some power management settings in BIOS that may be related with the random reboot issue. Could you restart the system then go to BIOS screen by pressing DEL key and help to check:

1. In the Main screen, what are the "Project Version" and "Build Date and Time"?
2. In Advanced screen -> CPU Configuration: Can you see "Power Technology"? If yes, what is the setting?
3. In Advanced screen -> SATA Configuration: Can you see "Aggressive Link Power Management"? If yes, what is the setting?

Hi kevinliao, I performed a shutdown and checked the following settings:

The version of the BIOS provided is Q61AR1B - 12/26/2011 (the factory version is Q61AR16).

CPU - Power Technology: Disabled
SATA - Agressive Link Power Management: Disabled

Now, I also noticed the BIOS has Intel Active Management Technology extensions (which provides an "out-of-band" web interface that also includes an hardware event log, which might document if these reboots are hardware or software). I tried to enable the AMT, but since my Nework Cards + Switch are configured with 802.3ad link aggregation, I am unable to access the interface.

I wonder if anyone else has experience and is able to access the AMT on their system. The event log might provide a cause to the reboots.

[ProTeck]

Is there any update from QNAP regarding this issue?

[gm85]

Is there any update from QNAP regarding this issue?

I've been working with support (it looks like they're attempting to replicate my setup there).

Since Monday, the system has not crashed. Here is what I have changed since then:

-Applied the BIOS update as mentioned (although it crashed an hour after... that MIGHT have been a fluke since it hasn't crashed since)
-Removed multiple iscsi initiators to one target (I used to have 3 ESXi servers working with 2 iSCSI Targets.... all configured through dynamic discovery. I have changed the setup so we now only have 2 ESXI servers and each server is statically configured for one iSCSI target).
-Entered the Intel AMT settings (CTRL+P at bootup) and disabled the feature.

So I'll keep an eye on it. Not sure which of those settings has made the system stable so far, but we'll see. I'm thinking of placing the iSCSI setup back to the way it was to determine if that was the issue (or to rule it out).

[ProTeck]

With only on reply from a QNAP rep(Jason), everyone there has be silent on this issue. I wonder if they have been too busy with the CES show. Is there anyone at QNAP willing to acknowledge that there is some problem with x79 units?

[dbaps]

I'm having the same issue on my TS-509. I don't use ESX or ISCSI. The reboots are happening once or twice in a 24 hour period. Just curious, forgive me if this was already mentioned. Did you update the firmware on your drives? The part that is so frustrating is there is absolutely no relevant messages in the logs. I want to try and stress my memory too, though I need to lookup how to do that process. I guess I'm a newbie but how did you enter CTRL + P at bootup? Do you attach a USB keyboard directly to the NAS? I am getting the check disk errors in the log but I could sneeze the wrong way and that message would show up. It's totally unreliable. Is there anyway to keep a historical record of the internal temperature as this can cause random reboots?

[StarrCompany]

Our device was RMA'd new device installed - less than 48 hours later a random reboot !! Have to agree there is something wrong with these devices - not sure where to go from here !!!

[gm85]

Just an update....

Since my last post, the sytem has been stable and has not crashed. My plan this week is to reconfigure iSCSI to the way it was (each ESXi machine having access to all iSCSI targets, with both ESXi machines accessing VMs on each target)

As mentioned, I think the reboot was either caused by
-The previous BIOS version (hardware)
-The Intel AMT extensions (hardware)
-or the iSCSI (OS/software).

I'll make this change to iSCSI, and if it DOESN'T crash afterwards, then the obvious answer will be to update the bios and disable the Intel AMT.

dbaps: Yes I attached a USB Keyboard and Monitor to the NAS and the screen appears during bootup. However, I don't think this AMT is on the older NAS's. My TS-259 Pro does not have it as an option.

[StarrCompany]

Thanks for the update - great info

[gm85]

Well i spoke too soon (maybe i jinxed it). The Qnap crashed tonight, 2 Weeks and 2 Days after the last. Checked the logs (and sent them to my Qnap support contact) and once again I could not see any indication that could have led to the Crash.

As mentioned, i previously changed my iSCSI setup to one-host-per-target and it still crashed (so iSCSI may not be a factor at all, or it may be some other underlying issue.)

The only thing i can think of trying is going into the BIOS and turning off the Intel AMT/ME extensions alltogether (previously i just went into the AMT/ME setup and unconfigured it from there). I know i keep going back to that feature, but its the only thing i can think of other than a BIOS fault or hardware design defect that might cause this. The AMT/ME feature has the ability to control the power state of the system, as well as features like a watchdog timer.

[gm85]

Heard Back from QNAP.

They said they are releasing a firmware update in Early Feb that should resolve this issue. Apparently the issue is with Port Trunking / Bonding.

In the mean time they told me to try disabling Port Trunking (which I plan to tonight. I'm going to assign IPs to each network port and stagger the iSCSI connections from my VMWare servers across those two addresses).

So my question is for those who have been experiencing these crashes, do you have Port Trunking enabled?