Upgrade to 4.1.8 on NV+ stuck on "Booting" case 16910762

[wwalkersd]

On Saturday morning I started the upgrade of my NV+ from 4.1.5 to 4.1.7. The download was successful, I was prompted to reboot the ReadyNAS, and I did so. I checked back on it a couple of hours later and RAIDar showed the status as "Booting". The next chance I got to check it was Sunday morning, over 24 hours later. RAIDar still showed its status as "Booting". So, I powered down the NAS with the power button and rebooted it. It's been over an hour, and it still shows "Booting" in RAIDar and on the front panel.

Was there some reason I needed to upgrade to 4.1.6 first? What should I do now? There's no access to the NAS via Frontview.

OK, I just had a brainwave and tried connecting via ssh using the IP address, since I had previously had ssh access enabled. This was successful. The prompt I get is "Linux 2.4.18 Padre".

Sadly, I don't enough at this point to figure out what's going on. Wait, maybe I know a little.

/var/log/raidiator_version indicates that I'm still on 4.1.5 (RAIDiator!!version=4.1.5,time=1239467045). I'm not finding anything in any of the other log files (examined using tail) that would seem to indicate a problem.

ps -e shows these processes running:

WalkerNAS:/var/log# ps -e
  PID TTY          TIME CMD
    1 ?        00:00:05 init
    2 ?        00:00:00 ksoftirqd/0
    3 ?        00:00:00 events/0
    4 ?        00:00:00 khelper
    5 ?        00:00:00 kthread
   10 ?        00:00:00 kblockd/0
   13 ?        00:00:00 khubd
   41 ?        00:00:00 pdflush
   42 ?        00:00:00 pdflush
   43 ?        00:00:00 kswapd0
   44 ?        00:00:00 aio/0
   45 ?        00:00:00 cifsoplockd
   46 ?        00:00:00 cifsdnotifyd
   92 ?        00:00:00 kvblade
   93 ?        00:00:00 mtdblockd
  106 ?        00:00:00 hotplug-sata
  131 ?        00:00:00 djsyncd
  132 ?        00:00:00 djcheckd
  135 ?        00:00:00 hotplug-gmac
  394 ?        00:00:04 raidard
  553 ?        00:00:00 sshd
  562 ttyS1    00:00:00 getty
  563 ?        00:00:00 sshd
  565 pts/0    00:00:00 bash
  597 pts/0    00:00:00 ps

I think I'm out of ideas at this point. The force is weak with me, O Great Jedi Masters.

[chirpa]

You can do direct updates from 4.1.5 to 4.1.7, it should have worked.

Can you give me access to SSH on it (port forward setup) ?

[wwalkersd]

Um, sure. However, I just found a file /.update_fail which contains "SYSTEM_UPDATE_FAIL_EXTRACT"

Does that mean what I think it means, that the downloaded update was not properly extracted from an archive? Odd that the system didn't tell me about it.

I think I have to reboot my router to get the port forwarding to work, so I'll get back to you when it's set. Thanks, Chirpa, that was fast!

[WhoCares_]

While you're on it, can you do a

df -kh

Sounds like your root file system hasn't got enough space left to successfully extract the firmware image.

-Stefan

[wwalkersd]

Chirpa, port forward for SSH (port 22, TCP, UDP) in place. IP in PM Let me know when you're done.

[wwalkersd]

Stefan:

WalkerNAS:/# df -kh
Filesystem            Size  Used Avail Use% Mounted on
/dev/hdc1             1.9G  438M  1.4G  23% /
tmpfs                  16k     0   16k   0% /USB

That looks kinda small. See sig for drives.

[markwoll]

edit your message to chirpa and take out the ip address.
I would just PM it to him.

[mdgm-ntgr]

OS partition is 2GB on Sparc devices (e.g. NV+). That's normal. Plenty of free space. Your data volume isn't mounted (so either you haven't reached that far in the boot process or it can't mount your volume for some reason).

[kiwi11]

Sounds like a simular issue to mine 😞

[sphardy1]

Sounds very similar to these reported issues:

viewtopic.php?f=7&t=43468#p245129
viewtopic.php?f=17&t=45333&p=258275
viewtopic.php?f=17&t=47442

Workaround in each of these cases is to pull out one of the drives and reboot, then reinsert the drive and await the resync.

kiwi1 wrote:
Sounds like a simular issue to mine 😞

Actually, completely different judging by your other posts

[kiwi11]

Funny all my posts/issues have been caused by having the second drive in place. Front view took ages to load I mean 20 minutes+ per instruction. I tried everything then found this post and decided to pull drive 2 out just to see as I'd tried everything else i.e firmware upgraded, tested & swapped drives, new router I even isolated my network all to no avail. Took drive 2 out and bang front view works a charm and have just uploaded full release of 4.7 and now re-booting etc before I add the second drive back, hopefully then it will re-sync correctly.(God I hope so as the box has nearly been thrown out the window a few times now)

I know I've used a WD drive in a netgear product but what a load of balls this process is, there should be a warning on buying third party drives or something. Maybe there is and I missed it but the time wasted and frustration caused I feel like a bloody NV+ guru now and all I wanted to do was stream movies from a back up.... rant over 🙂

[mdgm-ntgr]

4.1.7 should fix most compatibility issues with WD disks. If it's a WD20EARS, I'd suggest you backup your data, verify backup is good, do a System > Config backup, then factory default with both disks in place. A factory default on 4.1.7 is needed for 4k sector alignment.

There is warnings in the manuals to only choose drives from the compatibility list.

[sphardy1]

kiwi1 wrote:
Funny all my posts/issues have been caused by having the second drive in place.

That maybe correct, however this post (it seems) and the 3 I referred to report symptoms where the NAS fails to boot at all with all drives installed - not just go slow. That and the fact WD drives you have are well known to have issues, including affect performance, that 4.1.7 specifically addresses strongly suggests yours is a very different issue.

Good luck with the resync

[wwalkersd]

Well, this has wandered a bit from my original request. Anyway, I opened a tech support ticket, number 13710255.

[kiwi11]

Box is now failing to boot with 2 drives with new firmware! So I'll boot with 1 but can some one please explain why I need
'A factory default on 4.1.7 is needed for 4k sector alignment.'
And how I'm supposed to do that if it hangs on booting!
Thx

[mdgm-ntgr]

Some manufacturers have started to release 4k sector disks. If the partitions aren't properly aligned for 4k sectors you may have poor write performance.

You could try http://www.readynas.com/forum/faq.php#How_do_I_reset_the_ReadyNAS_to_factory_default%3F. Bear in mind that this will wipe all data, settings, everything.

[wwalkersd]

Since I don't have 4K sector disks, is there any reason to believe a factory default is going to be the solution to MY "Booting" problem? I haven't heard anything from Chirpa.

[sphardy1]

wwalkersd wrote:
Since I don't have 4K sector disks, is there any reason to believe a factory default is going to be the solution to MY "Booting" problem? I haven't heard anything from Chirpa.

A factory default covers a multitude of problems and is highly likely to solve your issue, but without getting to the root cause.

Have you tried pulling a drive & then rebooting as I suggested? Woudl be good to know if you are havgin the same issue as others as then we can bug Netgear to take a look and perhaps finally solve this (I struggled for a week and ended up unnecessarily factory defaulting, wasting my time and putting my data at risk)

[wwalkersd]

sphardy wrote:
Have you tried pulling a drive & then rebooting as I suggested?

I'm giving it a try now, although it seems like rather a "black magic" solution. I'm thinking the only outcome of this will be to force a resync, and that there ought to have been some way to do that through the command line.

Interestingly, all four drive lights were green. I pulled drive #4, and its light stayed solid green. Upon reboot, it's now flashing. We'll see how it goes.

Well, I've discovered one thing: you can't trust the drive status indicators in RAIDar when the NAS is in "Booting" state. It's showing me four green lights!

[wwalkersd]

No joy. Still in "Booting" state, still have SSH access but no frontview. But now tech support wants in.

[wwalkersd]

Well, working with tech support I've now tried the USB Flash boot recovery process. One flash drive results in a kernel panic after the "USB FW Recovery" step, while the other never gets past "USB Init." Grr. I tried reinstalling the FW from flash (even though I had failed to update the flash), and got back, as far as I can tell, to where I started: "Booting".

This is getting to be a long time without access to my data.

[wwalkersd]

The following flash drives will NOT work with the NV+: Kingston Data Traveler 2 GB and LaCie imaKey 8 GB.
The following flash drive WILL work with the NV+: SanDisk Cruzer 4 GB.

So, I was able, using the flash drive, to reload 4.1.5 into the onboard flash. I then did an OS reload to disk from flash, and after resetting my admin password my NV+ seems to be back up and running 4.1.5 (and is busy resyncing drive 4).

So now I'm back where I started before I attempted to update to 4.1.7. Is there a way to safely update to 4.1.7?

[mdgm-ntgr]

You can update using System > Update > Remote in Frontview. Or download the 4.1.7 image and update via System > Update > Local. See http://www.readynas.com/RAIDiator_4_1_7_Notes

If it still doesn't work for you, best to contact tech support for assistance.

Btw there is a 4.1.7 USB image: http://www.readynas.com/forum/faq.php#How_can_I_perform_a_Boot_Recovery_using_a_USB_flash_device%3F

[wwalkersd]

This just keeps getting weirder. I tried updating via Frontview to a local copy of 4.1.6 (NOT 4.17), and got the same result: stuck forever at "Booting...", no Frontview. Tried a USB upgrade to 4.1.7. Same result. Used USB boot to go back to 4.1.5, and everybody's happy!

The only thing I can think of is that this must be due to the changes I did to my system, which were as follows: install qpopper, and configure the already-installed exim to host mail on the NAS. This has been working fine for me for quite some time. But I figured I would try backing out those changes. The first thing to do is to take the relevant lines out of /etc/inetd.conf. Well, it turns out that there's now no way for me to do that! I can still connect via ssh and get to the files, but even though I'm logged in as root and have write access to the file, the system will not let me delete it or overwrite it! inetd.conf has become untouchable. I can't even chmod it!

WalkerNAS:/etc# cp -f inetd.conf.orig inetd.conf
cp: cannot remove `inetd.conf': Operation not permitted
WalkerNAS:/etc# chmod -t inetd.conf
chmod: changing permissions of `inetd.conf': Operation not permitted
WalkerNAS:/etc# su
WalkerNAS:/etc# chmod -t inetd.conf
chmod: changing permissions of `inetd.conf': Operation not permitted
WalkerNAS:/etc# sudo rm inetd.conf
rm: remove write-protected file `inetd.conf'? y
rm: cannot unlink `inetd.conf': Operation not permitted

I even tried killing inetd, in case the issue was that it had the file open. No joy. So, I'm thinking I can't uninstall qpopper, because inetd will still be trying to run it.

Does anybody have any idea how I can back out these changes?

[WhoCares_]

Sounds like your file system was mounted read-only. Can you verify this using the command

mount

?

-Stefan