QNAP firmware issues : Major file system bug

[pmw]

It says 3.6.1 Build 0302T in information.

Filesystem Size Used Available Use% Mounted on
/dev/md0 4.0T 1.7T 2.3T 42% /share/MD0_DATA

[/share/MD0_DATA/Public] # time mkdir foobar
real 0m0.076s

Now it works fine, but when usage was ~65% and more, it was _really_ slow.

[moritzthecat]

This thread has meanwhile the most critical posts I have ever seen - quantity and quality.

While observing and participating in this and related threads (it is not the only thread the issue is reported) and being affected by the issue, there is still no evidence from QNAP that they are taking this serious. NO official statements from QNAPSimon (or others) about possible workarounds, clarifications or detailed desription so that we could perform our own risk assessment. So my interpretation is that QNAP is hiding something very serious - forced business case worst case assumption.

Furthermore is is definetely wrong info from QNAP support (as reported by Liperty above in viewtopic.php?f=25&t=51741&start=360#p261746) that the issue has been fixed in 3.6.1 (as it is still the same kernel 2.6.33.2). Even the info I received recently that it will be fixed in 3.6.4+ together with a required kernel update later in Q2 is questionable and not consistent given the level of customer communication maturity.

As mentioned before I could not just wait to take further risk and had to prevent my applications from these spurious failures. Hence I started buying 2x DS411 and are very happy with the performance of these machines AND the level of support I received from competition. For me the level of support makes a big part of the difference - and finally saves time which equal hard $$. In case you are interested pls have a look how effective customer RMA communication (my opinion) looks like. Here is an example of a most recent issue handled professionally: http://www.synology.com/support/news.ph ... ews_id=338

Again I will wait until Q2 and if there is no solution then I am forced to even sell and exchange all installed TS-machines (currently below70% disk usage).

For me the time budget I spent of this issue is finally exceeded and I have nothing more to add.

[schumaku]

In case you are interested pls have a look how effective customer RMA communication (my opinion) looks like. Here is an example of a most recent issue handled professionally: http://www.synology.com/support/news.ph ... ews_id=338

That's not yet another warranty RMA - much worse, this is caused by highly substandad quality assurance by their hardwae sub-contractor and/or poor work and test procedures defined.

[moritzthecat]

Hi schumaku I am not referencing the root cause or severity of this issue but the way this issue is handled and addressed. RMA means return merchandise authorization and in this context the open communication was meant to done properly to my opinion. This is completely missing from QNAP concerning the issues addressed in this tread and maybe others. - that is the difference in customer communicatio I am highlighting.
I wish QNAP would change their behaviour in same direction though but I have no time any more left to keep myself busy with this...

[schumaku]

A CEO does easily understand the negative effects of a blue LED behaving obviously wrong. Vey different, a CEO does not recognize something to be a big problem where the effects can not be - after all the years - reliably reproduced by the QA departement. In their mentality, a problem that can't be reproduced, will not be considered an issue to be addressed. The only indication might be genrated by us - by placing much more support requests on that issue.

No, this is not the only issue where we are facing that...

[moritzthecat]

good point. but issuing just more support requestes did not work at least for me... prob. they are marked solved/answered internally anyway.
As next step escalation I have asked QNAP sales/support to provide me with the name of an executive in charge fo the quality and support to issue a formal complaint request, but I could not get any names from them. Also there is no board management listed on their web sites.... any ideas / names ? I think we need to escalate ...

[wallstrum]

if it helps, after doing some digging, here is the linked in profile for the QNAP CEO, Meiji Chang

http://tw.linkedin.com/pub/meiji-chang/10/366/926

i've been following this thread since the beginning and hope this can help someone prompt a response and get a resolution to it.

wally

[QNAPJason]

Hi everyone,
We're sorry about the late response as it took some time to prepare the testing environment.
Based on our preliminary study, a newer Linux kernel does exhibit better performance in this situation.
We have begun the kernel upgrade and expect to release it in July for the x86 NAS.
Then the ARM-based NAS will follow.

[forkless]

Hi Jason,

I have to ask again since your reply does not give us an actual update/confirmation about the actual freeze on the filesystem people encounter.

While it's nice to see a newer kernel shows performance improvement but has the filesystem freeze issue described in this thread already have been resolved as may have been indicated by a mail exchange by one of the members and your sales department (earlier in this thread) or is this problem still trying to be addressed by upgrading the kernel version?

Regards,
fork

[dexterpicard]

Hi everyone,
We're sorry about the late response as it took some time to prepare the testing environment.
Based on our preliminary study, a newer Linux kernel does exhibit faster performance.
(Edit: faster performance when the NAS is almost full).
We have begun the kernel upgrade and expect to release it in July for the x86 NAS.
Then the ARM-based NAS will follow.

2yrs+ is probably a little long to fix something that is broken? but im sure users are long way gone now and went for alternative solutions if possible.

[rskadk]

Like i wrote in another post:

Currently major firmware problem in v. 3.6.1.

Problem on TS-410 (And many other NAS as far as i can read on the internet. I think everyone running 3.6.1 is having the problem). I can upload to the NAS with 200-800kb/sec (Before 30Mb/sec +). I can currently download from the NAS (Read), at 30Mb/sec.

So the problem is only writing to the NAS.

FW: 3.6.1 Build 0302T

I'm also going to write to QNAP support. This should be fixed right away, since the product is currently not working, with theese speeds. Very unacceptable!!

And i'm not gonna spent time restoring the whole NAS, making backup etc etc.

So i guess if it's not fixed in the coming week, (because this is so critical, and basically making the produt totally unuseable) - Then QNAP should gurantee that people can return the product and have their money back.

[Rémon]

Same problem here.

And also waiting for a solution...

[joelang1699]

I think we are being fed a load of bs tbh. They either dont have a clue how to fix it or they are simply stringing us along. This is an absolute disgrace even if the projected firmware is out in july ARM units will "follow"...that means nothing Qnap. Is was planing to upgrade to a TS 1079, i think not now! A 1079 with ten 3tb would have roughly ten percent of unusable storage!!!!

Get you act together Qnap.

[doktornotor]

I think we are being fed a load of bs tbh. They either dont have a clue how to fix it or they are simply stringing us along. This is an absolute disgrace even if the projected firmware is out in july ARM units will "follow"...that means nothing Qnap. Is was planing to upgrade to a TS 1079, i think not now! A 1079 with ten 3tb would have roughly ten percent of unusable storage!!!!

Get you act together Qnap.

Well, I definitely would not suggest upgrading ATM. Having 10 HDDs and being unable to remove any drive without a system crash definitely does not rock. Sounds like all these NAS vendors would benefit from hiring some kernel maintainer. Dunno what is behind this "the shipped kernel version is set in stone no matter how buggy it is" mentality.

[Egg7]

Just move on (to another brand). I sold the faulty unit (before this issue becomes well-known, now that it is it might be more difficult -you know who to blame-) and I'm happy with my new Syno unit bought on 2011/02.

I reported this problem on 2010/07 already!!: viewtopic.php?f=45&t=33431

I waited 7 months (!!, seriously, totally unacceptable) to get a working solution. Here it's even worse, some of you have been waiting for almost 18 months. I work in computers on little jobs here and there for medium-sized companies, this issue is now quite well-known and from what I gathered qnap lost some sales in ~1.5 year because of that.

What are you waiting for ? A miracle ? Companies love people like you. You can't blame them for bs'ing you since you're so willing to believe anything.