Have it working... so far, my findings so far (P.S. Netgear support were NO USE AT ALL)
Issues :- 1) The Linux hard drive drives are not reporting errors correctly and more importanly no sensible error message in the logs or the GUI 2) There are physical coding errors in the firmware which are not capturing errors, see my post earlyer 3) One of my New drives turned out to have a bad sector which could not be written too. 4) After replacement, it did sysnc and is now Raid1, but while it was raiding/Sysncing there were hundreds for unhandled errors. Here are a few example. The problem is they get deleted as soon as the box reboots, so unless you watch the sysnc run you don't see them
UBI: VID header offset: 2048 (aligned 2048) UBI: data offset: 4096 UBI warning: ubi_eba_init_scan: cannot reserve enough PEBs for bad PEB handling, reserved 19, need 38 UBI: attached mtd2 to ubi0
And
Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction
Have it working... so far, my findings so far (P.S. Netgear support were NO USE AT ALL)
Issues :- 1) The Linux hard drive drives are not reporting errors correctly and more importanly no sensible error message in the logs or the GUI 2) There are physical coding errors in the firmware which are not capturing errors, see my post earlyer 3) One of my New drives turned out to have a bad sector which could not be written too. 4) After replacement, it did sysnc and is now Raid1, but while it was raiding/Sysncing there were hundreds for unhandled errors. Here are a few example. The problem is they get deleted as soon as the box reboots, so unless you watch the sysnc run you don't see them
UBI: VID header offset: 2048 (aligned 2048) UBI: data offset: 4096 UBI warning: ubi_eba_init_scan: cannot reserve enough PEBs for bad PEB handling, reserved 19, need 38 UBI: attached mtd2 to ubi0
And
Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction Alignment trap: not handling instruction e1043090 at [<40387a0c>] Unhandled fault: alignment exception (0x801) at 0x464c457f Alignment trap: not handling swp instruction
You should have let us know what you were doing. We've been monitoring your box, we saw the IO errors on your disks but upon your advice that "the disks worked in jbod" we let it resync to see if we could catch anything. All of a sudden today it's working fine and we can't cause it to break (I just rebooted your Stora over 10 times).
The alignment trap errors are kernel related, they pose no issues. As for the actual resolution... bad disk. Thanks for the confirmation!
I guess my issues are the really poor error handling and poor coding.
The fact the you were not able to identify the issues shows there are serious issue with the firmware. And the number of people here report problems again show there are issuse. Had the firmware reported disk bad block and advised a disk change we would all be happy. Also the error logging the system does is not large enought, its rolls round and starts loosing data so you lose what happened.
Having been a computer professional for 20 years+, I can say, warnings and alignment errors which are not handled and reporting lack of parameter space are problems and show poor coding.
I accept that computer programs have errors, but when people report errors they need to be investigated and fixed, this type of error has been going on for year, some logs are back in 2009. The OS need to be improved to capture errors better, and then once the errors are captured they should be email to support. Support would then understand the problem in there code and address it.
Alignment errors are real errors and should not be happening.
One last comment, People purchase this unit becasue they want a RAID device, that means redundany was high on thier list, knowing what I now know, this device has a number of issues which mean I can not trust it, the most important thing we all have is data, and this unit is meant to protect that.
From the errors I've seen while building the Raid array, (program addressing errors) I can't say its safe to use and rely on 100%.
That distroys the reason to own one of these. :eek:
The self check seems to have fixed the RAID issues with my box...
Does any one else get beeps when your stora is spinning up? -
Mine beeps intermittently now ive run the self check.
But before the RAID failed it bleeped multiple times, every time both disks were spun up.
Mr Squeezy
The beeps have returned today... Only 2 days after self diagnostics... Aw well - I guess I'll jest have to keep repeating the self check everytime the raid fails
I have tried to replicate your concern, formatted the drives, set it as raid 1, wait till it finished to reconfigure, reboot the Stora, but still showing 2 disk, am I doing something wrong? My Stora is running on 2.6 firmware.
With regards to replicating this test - These issues slowly get worse (snowball) over time - try leaving it for a week without turning it off - only accessing occasionally.
With regards to replicating this test - These issues slowly get worse (snowball) over time - try leaving it for a week without turning it off - only accessing occasionally.
Mr Squeezy
Also to point out the obvious - this thread has had almost 1,400 views... It would appear that this issue is affecting quite a few people.
Also to point out the obvious - this thread has had almost 1,400 views... It would appear that this issue is affecting quite a few people.
Mr Squeezy
a 'view' means people have looked.. it doesn't mean people have the issue.... so far as no one else from the 1400 views has responded it seems to be very limited to a few people..
My bet is on the drive not being 100% conpatible with the Stora
a 'view' means people have looked.. it doesn't mean people have the issue....
Including bots!!!
As well, I've posted in this thread to contact us at support@axentra.com. Only one person did so and his issue was a bad disk and he's posted here saying his issue is resolved... I've tried reproducing this(with good disks) and I can't!
There's about 6000 units updated to 2.6 and we're not seeing much concern over this issue. I'm not saying your Stora doesn't have a problem, but the issues are most likely unrelated and specific to each Stora.
As well, I've posted in this thread to contact us at support@axentra.com. Only one person did so and his issue was a bad disk and he's posted here saying his issue is resolved... I've tried reproducing this(with good disks) and I can't!
There's about 6000 units updated to 2.6 and we're not seeing much concern over this issue. I'm not saying your Stora doesn't have a problem, but the issues are most likely unrelated and specific to each Stora.
We'd be happy to investigate!
Cheers, Roby!
and mine is one of the good ones.. 2.6 and not a peep of this issue...
I have been in touch with netgear support I think they have bumped me up to level 2 support.
Samsung spinpoint F4 HD204UI are very popular drives - and very well reviewed... (old firmware aside)
The HD204's are also included in the pinned thread compatibility.
I certainly won't rule out the possibility of the issue being a faulty drive - but it will definitely be one of the last things I explore considering my luck with this stora so far.
I suffered the hard drive drive activity issue on 2.5.3 - Both HDD worked though - I did keep my stora turned off when not in use to prevent the hdd from burning out.
On 2.6 the drive activity has stopped - But when I need access to the stora, the drives began to struggle to spin up. I had to disable the power timer because the stora occasionally had trouble booting up. Could this fault be a power issue? I doubt it will be but I have never heard of a consumer hard drive struggling to spin up before....
Robyb you're suggesting that its purely a firmware 2.6 issue - would a roll back to an earlier firmware help diagnose this error?
Any help would be gratefully received.
Mr Squeezy
p.s. Sorry about the comment about views - I figured there were similar people with the same issue. It makes me even more thankful that you are both going out of your way to help.
I have been struggling with a very similar "disk2 disappears" problem after going to 2.6. In my case I have two Seagate ST31000528AS drives and have a scheduled power up and power down scheme. One day after the upgrade I got an io error during a scheduled backup and I had to reboot Stora. Then it turned out the directory where the backup was to be written was corrupt as I could not remove it while I was root on Stora. So I took out disk2 and found that now I could remove the directory (on disk1) so I suspected disk2 was the trouble maker. I checked and zeroed the disk2 with Seagate's Seatool and reinserted it back in Stora. After 2 hours everything was fine again. In the evening the backup went fine. The next day (after a scheduled power and power down) the scheduled backup got stuck. Guess where? Indeed the same place under MyComputers. However now I saw that Stora started again to syncup disk2. This time I tried to take out disk1 first. Everything is fine. Take out disk2, everything is fine. Inspecting the two disks with Seatools shows no problems. Inspecting them with gparted showed a problem with disk1 (disk2 was fine), and testing both disks structure with TestDisk showed no problems.
Something is corrupting things here. Furthermore ,as it was said before, it is very annoying that one does not get propper failure/error reports.
I will now zero disk1 and put it back again and see what happens. In the mean time I have removed the parent directory of the backup target and recreated it.
I have contacted the email address as you suggest. I would like to add my voice here and say that this is still ans issue for me and makes my stora unfit for purpose i would.also.love to see a tool to.downgrade to 2.5 where i had no problems and a working stora.
Given that my problems happened after the firware update i really cant see this being a drive issue.
Robyb has had a look at my Stora and can't get the raid to fail.
I explained to Robyb that the stora was still beeping.
He replied saying that the stora has no internal speaker.
I've never heard of HDD having an internal speaker and searching through youtube clips there is certainly no sound like the one my stora/hdd is making.
I have emailed samsung/ seagate describing the potential fault - I will reply with their response.
Its looking like a potential HDD failure - When the raid does eventually fail I will stick the HDD in a caddy and see if I can diagnose anything else.
Robyb has had a look at my Stora and can't get the raid to fail.
I explained to Robyb that the stora was still beeping.
He replied saying that the stora has no internal speaker.
I've never heard of HDD having an internal speaker and searching through youtube clips there is certainly no sound like the one my stora/hdd is making.
I have emailed samsung/ seagate describing the potential fault - I will reply with their response.
Its looking like a potential HDD failure - When the raid does eventually fail I will stick the HDD in a caddy and see if I can diagnose anything else.
So, I did zero disk1 and put it back and I was just planning to go back to the forum and report that everything was working fine. But yesterday it again failed during a backup to write one of the files of the backup. The directory that was mounted (on the win7 machine) to which the backup program was writing became corrupt. Try each disk separately I could cd into the directoty below but got io error when trying to the delete the directory where the backup failed and subsequenly I lost my /home dir. dmseg reported a failure but in the web interface no information on the error is available (at least I cannot find it). I did then do repair (using the original inst disk) with only disk1 installed and then I could erase the directory and recreate it.
The disks themseleves are ok when checking with seatools. I'm getting more convinced that the 2.6 is suspicious.
I will now try if for two weeks or so with only one disk and check if that works. Next I can insert a new similar disk (luckily I bought 2 spares before prices went up).
BR Ben
PS: Backup program is Paragon Backup & Recovery 11 Home. Failure was while doing a new full weekly backup.
Hi guys, I have the same problem: frequently, I lose disk 2. I just have to reboot my unit and disk2 is available again (with need of rebuild) I receive and email alert when disk 2 disappears.
I always receive this email at 04:04AM !!! Why always at same time ? Does the unit check some times only and doesn't send mail in real time ?
I tired to upgrade: same issue I've just activated the scheduler to power off the unit during the night... I don't know if it will solve the pb.
Thanks
ps: I had the issue since the first times, it started only one week after first install
Well, when the error occurs get root access to your Stora and use e.g. putty to login and become root and then issue the dmesg command and inspect the printout.
I suspect that the file system somehow gets corrupted during writing. How large are the files that you are working with when the problems start? (see below why I ask)
Well, back to my previous submission to this thread. The backup again failed (now only with one disk). Restoring access again I created a new subdirectory and then via FreeFileSync I copied to directories/files that Paragon was backing up to this subdirectory. No problems, everything worked smoothly.
So could it be that the file corruption occurs when Paragon is pumping its 24G backup image to the Stora backup directory? With 2.5 everything worked fine but after the 2.6 update I started to have this problem with lost drive. (I mean Paragon worked smoothly with the 2.5 Stora sw version). The above would explain why the Paragon daily incremental backups have no problem. They are small!
Furthermore all other file systems and normal read their content on Stora works ok. Since Paragon worked with Stora 2.5, and with FreeNas, and with a Synology NAS I do not consider the Paragon backup program the bad guy but the changes between 2.5 and 2.6.
For the moment I stop experimenting and move to backup to my Synology and use FreeFileSync to my Stora for the daily incremental backups.
the stora is not used for backup. Inside, there are only personal files (pictures, working docs, docx, xls....) The units was powered on 24/24. Sometimes, the problem occurs many days after the last access (write OR read)
According to me, it's not linked to big file size, fs corrupt...
what is very surprising for me: the problem always occurs at the same time (04:04 AM, when all other devices/computers are powered off and every one is sleeping)
