Re: Cannot write more data to NAS (NV+ V2) (19489971)

Zaim · ‎2013-01-19

Can you tell the NAS to format to EXT3 ? Im thinking of buying this NAS and certainly don't want the issue's mentioned here.

Tangoman · ‎2013-01-24

Grrr - I wish I'd got a (cheaper) V1 now - bought the V2 because I wanted the ability to support 2Tb+ drives in the future.

However, I've ended up with a barely usable product.

janforman · ‎2013-01-26

Tangoman wrote:
Grrr - I wish I'd got a (cheaper) V1 now - bought the V2 because I wanted the ability to support 2Tb+ drives in the future.

However, I've ended up with a barely usable product.

As you said this is not problem with fragmentation on ext4. Just simply reformat volume as ext3 and everything is OK.
On volume with aprox. 5% free space with same data ext4 driver performs more than 10x slower than ext3 driver.

V2 is lot faster than V1

My speeds are

Write: 52MB/s (416Mbit/s)
Read: 78MB/s (624Mbit/s)

With WD-GREEN AV-GP / Windows 8 Drag&Drop file 12GB

readynas2012 · ‎2013-02-02

janforman wrote:
Tangoman wrote:
Grrr - I wish I'd got a (cheaper) V1 now - bought the V2 because I wanted the ability to support 2Tb+ drives in the future.

However, I've ended up with a barely usable product.

As you said this is not problem with fragmentation on ext4. Just simply reformat volume as ext3 and everything is OK.
On volume with aprox. 5% free space with same data ext4 driver performs more than 10x slower than ext3 driver.

V2 is lot faster than V1

My speeds are

Write: 52MB/s (416Mbit/s)
Read: 78MB/s (624Mbit/s)

With WD-GREEN AV-GP / Windows 8 Drag&Drop file 12GB

yeah upon reading I think my issues the same, If I delete a load of files it starts working again..

/c
X-RAID2 (Redundant)
381.3 GB Free Of 5.4 TB

I've a spare hdd slot..
So will buy another 3TB drive and that should sort it, hopefully Netgear will permanently fix it with a firmware update

readynas2012 · ‎2013-02-09

added another APPROVED Caviar Red Internal Hard Drive WD30EFRX 3.5" - 3 TB

now I have

3.1 TB Free Of 8.1 TB

and can write to the raid NO problem, have sent 20gb last night, took 30 hrs to rebuld raid with extra hdd.. suppose that will do me until netgear sort a fix! bah humbug

StephenB · ‎2013-02-09

readynas2012- no need to post in triplicate.

robclarke411 · ‎2013-02-14

I have 6 of these units (I do a lot of photography & CAD work) and they all now have this same problem. They are all have about 12% space free and I can no longer write to them without the unit crashing completely. Please tell me that there is a fix/solution for this now? Did anyone see this through with Netgear support? If not I will be demanding that they sort this faulty product immediately. Unbelieveable, it should never have gone to market in this state. I have 4 NV+ V1 units that all work absolutely fine - what is going on with quality control these days.

dcunning · ‎2013-02-16

Ladies/Chaps,

I have have the same issue - get to 8% capacity and cannot copy large files - most annoying.

Netgear rep advised the following:

Please check if you have set a limit on the users that you created on the Readynas. Please see the the troubleshooting steps below:
1. Click on Security and go to User and Group accounts.
2. Please check if there is a limit set for any users that you created.
3. If there is a limit set, please remove it and test if you will be able to transfer any file size on the readynas.

This did not help.

Stumbled onto this discussion. Any solutions here?

robclarke411 · ‎2013-02-17

The forums seem to be littered with complaints about this faulty unit, here are a few examples:

viewtopic.php?f=21&t=68342
viewtopic.php?f=21&t=66731
viewtopic.php?f=21&t=63992
viewtopic.php?f=21&t=64739
viewtopic.php?f=21&t=69560
viewtopic.php?f=23&t=61107

What a nightmare. Still waiting to hear back from Netgear support. I think I know what they will want to do which seems to be this dreaded conversion back to EXT3 which not only requires that you lose all your data but does nothing in the way of solving this problem. Why an earth was this device not tested?! If the fault is inherent and unfixable they should be recalling all these units and refunding people. I'll post back when I hear from Netgear although I am not hopeful.

janforman · ‎2013-02-19

This may help without data loss ...

# disable journal
tune2fs -O ^has_journal /dev/c/c

# Required fsck after
e2fsck -f /dev/c/c

# this revert changes back (switch on journal)
tune2fs -O has_journal /dev/c/c

rabidh · ‎2013-02-21

Just to add, I have this problem too on an NV+ v2. My old Duo V1 had no problems at all, apart from being really slow. Nightmare - and after I found out (after the unit lost all data) that I can't use power save either! Really very poor of Netgear.

janforman: What are the bad points of disabling journalling? If there was a problem does it make it harder to back up?

janforman · ‎2013-02-22

rabidh wrote:
Just to add, I have this problem too on an NV+ v2. My old Duo V1 had no problems at all, apart from being really slow. Nightmare - and after I found out (after the unit lost all data) that I can't use power save either! Really very poor of Netgear.

janforman: What are the bad points of disabling journalling? If there was a problem does it make it harder to back up?

I decided to stay on ext3 / My Duo v2 can write 52MB/s on WD-AV without any problems.

A long time ago many professional sollutions don't use journaling.
When journal is enabled, then last modification can be rolled back (anyway some users lost data when journal is corrupted and replayed).
Without journal enabled you may see last modified/created file before crash - filled with some incorrect data.
It's about data integrity with power outage or fatal crash of OS.

StephenB · ‎2013-02-22

I would suggest a UPS, as that will reduce the risk of an unexpected shutdown.

Snipe3000 · ‎2013-03-01

When I try to disable journaling, I get this message:
The has_journal feature may only be cleared when the filesystem is
unmounted or mounted read-only.

is that normal?

lukeyates1981 · ‎2013-04-03

I have just raised this with Netgear Support and got passed straight to "level 2 support" who appear to be actually investigating the problem.

I am currently using 2 x 2Tb and 2 x 1Tb drives in X-RAID2. This gives me 3.6Tb available space with 1.8Tb redundancy. My NAS gets to around 0.5Tb available when it locks up. After restarting it and struggling to access it, I am able to delete some files (usually a few Gb) which is enough for the box to start working "normally" again.

As has been said a number of times, this is a crazy problem to have. How the product passed its testing phase is shocking.

I also sent a link to this thread to the engineer, pointing out that there are a number of other people with the same problem. I will have to wait and see what happens next!

Zaim · ‎2013-04-03

lukeyates1981 wrote:
I have just raised this with Netgear Support and got passed straight to "level 2 support" who appear to be actually investigating the problem.

I am currently using 2 x 2Tb and 2 x 1Tb drives in X-RAID2. This gives me 3.6Tb available space with 1.8Tb redundancy. My NAS gets to around 0.5Tb available when it locks up. After restarting it and struggling to access it, I am able to delete some files (usually a few Gb) which is enough for the box to start working "normally" again.

As has been said a number of times, this is a crazy problem to have. How the product passed its testing phase is shocking.

I also sent a link to this thread to the engineer, pointing out that there are a number of other people with the same problem. I will have to wait and see what happens next!

Thanks for the update, i'll also send in a support ticket, I agree, its shocking how this bug hasn't been fixed yet.

lukeyates1981 · ‎2013-04-03

The reply came back as follows:

Hi Luke

I have been digging around a bit here and I should have spotted this right away.
There is a known issue on the NV+ v2 with EXT4.
The only solution for it right now is to factory default the device and set it up with EXT3.
We have sent this up to the engineers for them to have a look at this and they are.
I have no date for when a fix will be sent out. But we are working on solving this issue for you and the other customers that may be experiencing this issue.

Kind Regards,

Xxxxxx xxxxxx
Level 2 Technical Support

I hashed out the guys name to save him some direct flack, but seems to be the same response as everyone else got. I will see how I get on with my next reply:

Hi,

Thanks for the reply.

Unfortunately this doesn't solve my problem and looks as though the only "quick" fix is to buy a new, bigger, drive and upgrade one of my current 1Tb drives. This will then require a full rebuild of the RAID array etc etc etc, which is going to take at least a couple of days. As I use this device for work this is not an ideal option. Also, returning the device to the "factory default" and setting up in EXT3 might be a tad complicated as I don't have a spare 4Tb laying around to backup my backup device.

Please could you let me know who I need to contact in order to arrange a full refund for the NAS. I feel that my best option at this point will be to buy a different device and transfer all of my data to it. While this will be a gigantic problem for me, it seems like a better solution than not being able to use in excess of 10% of my storage space. Considering the current price of hard drives, missing out on over 10% of the available storage space has a real cost implication to a small business like mine, as well as being a general nuisance.

The device is simply not fit for purpose, is not as described, and has become a real nuisance piece of hardware.

Thanks,
Luke Yates

MikeS1531 · ‎2013-04-06

Zaim wrote:
lukeyates1981 wrote:
I have just raised this with Netgear Support and got passed straight to "level 2 support" who appear to be actually investigating the problem.!

Thanks for the update, i'll also send in a support ticket, I agree, its shocking how this bug hasn't been fixed yet.

I can report I have a similar problem.

In early March I bought an empty ReadyNASDuoV2 and filled it with two Western Digital 3TB (WD30EZRX) drives in X-RAID2. It worked well until it became 80+% full, and then the unit became useless at accepting additional data -- small files went at a snail's pace and large files wouldn't go at all. Retrieving data from the unit worked at a reasonable speed.

I contacted Netgear telephone support, but after the person I spoke with spent over an hour with me we had gotten nowhere and they said they'd have to forward the case to someone else. A week later, having had no feedback, I phoned Support again. This time, the person I talked to acknowledged that this was a known issue, and we tried a few tests to confirm the unit's performance -- or lack thereof.

The most important point for others here with a similar problem is that Netgear have developed a patch -- still in Beta -- that might help. I say 'might' because, although it was reported as having helped a couple of other people, when we tried it on my unit there was no performance improvement. I had to leave the issue with Support while my contact reported the patch's failure up the line. If you haven't already been offered the patch to try, you might want to contact Support and ask for it. It might help, or it might not, but it ought to help Netgear identify the problem better so that they hopefully can improve the patch to the point where it works reliably for everyone and they can build it into their next firmware update. (I'm using RAIDiator 5.3.7)

I last spoke to Netgear on 2/Apr. Since then, I've done some more testing and have found that if I delete enough data from the NAS to have 20+% of the space free then it works as it should. So for now I just have to forget that my ReadyNAS has 3TB HDDs installed and pretend that it has just 2.5TB of capacity. Leaving half a TB of HDD space unusable isn't an ideal solution, but it is a workaround that I can live with for a little while. Hopefully Netgear will come up with a proper solution before too long.

The bad news is that I've not heard anything further from Netgear in four days, and when I log into MyNetgear account the ticket that had been raised by my telephone contact -- and had been visible shortly thereafter -- seems to have disappeared. So I guess I'll have to raise another one.

I'll post here if I make any further progress.

PS. Both of the people I've dealt with have indicated that they're with "Level 2 Technical Support".

lukeyates1981 · ‎2013-04-08

I have now had an update on my support ticket. I have been sent a link to a "program to fix the issue".

The link I have been sent to the file is: https://dl.dropbox.com/u/90544406/Filesystem_fix_5.3b-arm.bin
In case that dies after a while, I have copied the file to my website here: http://www.lukeyates.co.uk/netgear/Filesystem_fix_5.3b-arm.bin

According to the support guy, "You just go through the same steps as you would when applying a firmware. You select this file and upload it to the unit and it will fix the issue for you."

I have applied the patch to my NAS and it seems to be running the same as it was before (I am always a little nervous in case it kills my NAS and I lose all the data!).

I haven't tried to overfill the NAS as yet, but that will be my task for later this evening! I am pretty close to the breaking point, so it shouldn't take much to kill it! I think if I copy 100Gb or so across that should be enough to pass the point where it died last time!

I will update the forum once I know what happens next.....

Zaim · ‎2013-04-08

lukeyates1981 wrote:
I have now had an update on my support ticket. I have been sent a link to a "program to fix the issue".

The link I have been sent to the file is: https://dl.dropbox.com/u/90544406/Filesystem_fix_5.3b-arm.bin
In case that dies after a while, I have copied the file to my website here: http://www.lukeyates.co.uk/netgear/File ... 3b-arm.bin

According to the support guy, "You just go through the same steps as you would when applying a firmware. You select this file and upload it to the unit and it will fix the issue for you."

I have applied the patch to my NAS and it seems to be running the same as it was before (I am always a little nervous in case it kills my NAS and I lose all the data!).

I haven't tried to overfill the NAS as yet, but that will be my task for later this evening! I am pretty close to the breaking point, so it shouldn't take much to kill it! I think if I copy 100Gb or so across that should be enough to pass the point where it died last time!

I will update the forum once I know what happens next.....

Hmm, that's newer than the one I have been given today, 5.3A

MikeS1531 · ‎2013-04-08

Zaim wrote:

Hmm, that's newer than the one I have been given today, 5.3A

The one I was given was 5.3a, so maybe I should ask whether I should try the 5.3b version. Or perhaps the two different patches relate to different hardware -- NV+ vs. V2. (My NAS is a DuoV2.)

However, I also note that when I agreed to give the patch a try I also had to "promise not to disseminate it in any way, other than as expressly agreed by the representative who sends it to you including (but not limited to) that you will not post to forums, web sites and FTP servers for others to download, nor will you forward it on by more conventional means like the postal system."

Am I the only one who was asked not to allow anyone else to use the patch they were given?

With respect to my Netgear contacts, the person I dealt with over the phone has written back to say "At the moment we do not need a remote connection to your NAS since it is a known issue meaning the engineering team also experience the issue. We are getting a mixed result on the add-on and engineering team is targeting to release the add-on at the end of this month." I take the fact than Netgear can reproduce the problem to be good news -- there's nothing worse than trying to solve an intermittent problem, or one that can't be reproduced.

While waiting for the above reply, I did put in an online ticket, which generated the response "We had checked our system and our Engineering is still on the process in creating the add-on fix for the issue you had reported for your NAS DUOv2.

Again, I apologize for the inconvenience. Rest assured that we are committed to provide a resolution on your case at the least possible time."

So Netgear do recognise they have a problem, and are working on a solution. Let's all wish them success.

Zaim · ‎2013-04-08

Mine is a NV+ V2,

hauser1 · ‎2013-04-08

MikeS1531 wrote:

However, I also note that when I agreed to give the patch a try I also had to "promise not to disseminate it in any way, other than as expressly agreed by the representative who sends it to you including (but not limited to) that you will not post to forums, web sites and FTP servers for others to download, nor will you forward it on by more conventional means like the postal system."

Am I the only one who was asked not to allow anyone else to use the patch they were given?

So Netgear do recognise they have a problem, and are working on a solution. Let's all wish them success.

I was asked for the same NDA for the 5.3a patch I was given. what a joke these so called "support experts" so far it seems each of us have gotten a different level of competence & “drama” when trying to go through the steps of this support process.

I haven't been able to fully break apart what the patch does but the first one I was able to test (5.3) seemed to only change the number of extents the mutiblock allocator searched through before deciding to use a block. Really only useful for copying small files <128mb which happens to be the max size the mutiblock allocator is configured for.

Once files get above that size copying larger files the multiblock allocator seems to wait until some form of data is flushed and allocated (in full) to disk before it allows the client to save the file.

hauser1 · ‎2013-04-08

lukeyates1981 wrote:
I have now had an update on my support ticket. I have been sent a link to a "program to fix the issue".

The link I have been sent to the file is: https://dl.dropbox.com/u/90544406/Filesystem_fix_5.3b-arm.bin
In case that dies after a while, I have copied the file to my website here: http://www.lukeyates.co.uk/netgear/File ... 3b-arm.bin

According to the support guy, "You just go through the same steps as you would when applying a firmware. You select this file and upload it to the unit and it will fix the issue for you."

I have applied the patch to my NAS and it seems to be running the same as it was before (I am always a little nervous in case it kills my NAS and I lose all the data!).

I haven't tried to overfill the NAS as yet, but that will be my task for later this evening! I am pretty close to the breaking point, so it shouldn't take much to kill it! I think if I copy 100Gb or so across that should be enough to pass the point where it died last time!

I will update the forum once I know what happens next.....

i've noticed that that version for 5.3b that you were givin is not encrypted and we can clearly see what they are trying todo to fix the issue!


bye() {
  . /frontview/bin/functions
  cd /
  rm -rf $orig_dir
  echo -n " $1"

  log_status "$1" 1

  exit
}

orig_dir=`pwd`

cd / || bye "ERROR: Could not change working directory!"
tar xfz $orig_dir/files.tgz || bye "ERROR: could not install files!"


if ! grep -q "For bug 10425" /frontview/bin/frontview_rc;then
echo "
#For bug 10425
volume=\`dir /sys/fs/ext4\`
for dev in \$volume;do
  if echo \$dev | grep -q \"dm-\"  ; then
    echo 20 > /sys/fs/ext4/\$dev/mb_max_to_scan
  fi
done
" >> /frontview/bin/frontview_rc
fi

if ! grep -q "ext4-delalloc" /etc/init.d/readynas_startup ;then
  sed -i '/chmod 1777 \/ramfs/a\\n# ext4 mount option \n\/frontview\/bin\/ext4-delalloc' /etc/init.d/readynas_startup
fi

bye "Please reboot your box to let this addon work, thanks."

# Remove the installation files
cd /
rm -rf $orig_dir

exit 0

and the files.tgz in the payload has one file "/frontview/bin/ext4-delalloc":


#!/bin/sh
### BEGIN INIT INFO
# Provides:          ext4-delalloc
# Required-Start:
# Required-Stop:
# Should-Start:      glibc
# Default-Start:     S
# Default-Stop:
# Short-Description: Do every thing in our readynas system.
# Description:       Fron mount kernel fs to start the applicaton.
### END INIT INFO

# add nodelalloc to ext4 mount
echo "Running ext4 mount option tweak..." > /dev/console

for v in c d e f;do
 if grep "/dev/$v/$v" /etc/fstab > /ramfs/$v.fstab;then
   if ! grep -q nodelalloc /ramfs/$v.fstab;then
     grep -v "/dev/$v/$v" /etc/fstab > /ramfs/fstab$$
     echo "/dev/$v/$v        /$v      ext4    defaults,acl,nodelalloc,user_xattr,usrjquota=aquota.user,grpjquota=aquota.group,jqfmt=vfsv0,noatime     0 2" >> /ramfs/fstab$$
     cp /ramfs/fstab$$ /etc/fstab
     rm -f /ramfs/fstab$$
   fi
 fi
 rm -f /ramfs/$v.fstab
done

so basically their attempt to "fix" the issue is to reduce the scan amount for the multi block allocator (from 200 -> 20) and disable ext4's delayed allocator...

Tokfan · ‎2013-05-04

IHello,

I'm new to the forum, been reading a bit, but i can't find any sollution to this problem. The people in this thread that have had contact with the support have you heard any news and/or sollutions regarding the problem with file transfers when your NAS is almost "full". "Full for me means 723Gb still unused wich really sucks to be honest.