So I might be the first one with good news to report here.
After upgrading to the 6.2.3-T1652 beta last weekend, my RN104 has been stable. No lockups. It's been up for 8 days straight now, when previously it wouldn't make it more than 4 hours or so.
I don't know for sure, but I'm guessing the factory reset that I did prior to the beta upgrade might have helped. Also, I'm not sure if others are having the same issue that I had, but my unit would only lockup or freeze when it was idle. It never locked-up while I was actively using it.
Anyway, hope this information helps somehow. I also I hope that I don't jinx myself by writing this success post.
In less than 24 hours, the RN10400 locked up 7 times, until the last one when not even 'kernel panic=1' was able to reboot the system. Most of the lockups happened while it was idle.
I ran 1672 (Beta3) with the requested logging and it failed fairly quickly after going idle (I was doing some data copy first). Logs have been sent in.
I also ran journalctl -af in an SSH session and captured the output, although to my eye it looked unremarkable.
Simultaneously, I ran a bash script in another SSH session to capture the dmesg output every 30 seconds in case something spit up there, however, nothing out of the ordinary occurred. I *do* get the advantage of being able to tell within a 30 second window of when the machine crashed which may be helpful for the log analyzers. 🙂
I then downgraded to 6.2.0 and my NAS104 has been up for 12+ hours while idle. For me, that's a new record. More data as it becomes available.
Does anyone remember if 6.2.1 was locking up? Or did the problems start with 6.2.2? I *can* upload 6.2.1 to find out, but my wife would hate to be without her data..... 🙂
Mine locked up with 6.2.0, 6.2.1, 6.2.2 after 20 to 180 minutes or so and after installing the beta2 of 6.2.3 it locks up after 3 days or so.
I can however make it freeze/lock up on purpose if i read or write multiple times at once. Then it locks up straight away. An example would be if i unpacked 2 archive files on the nas drive at the same time. Or i upload via ftp and try copying out some large files via smb at the same time. Then bam, total freeze within seconds.
I know that this may sound weird (it sounds weird to me), but all the people here that are reporting having their system running more than 24 hours without lockups, like me, they have done some sort of downgrade.
I downgraded from 6.2.3 -T1762 to T1752 and it's been 48 hours without any lockup. Others downgraded to 6.2.0 and so on, and so on... just read back in the forum.
Don't know if this helps anyone, but a pattern is starting to show here.
We have a lot of users running the 6.2.x firmware without problems and generally we've found feedback indicates it is an improvement over previous releases.
However these lockups appear to be difficult to reproduce issues and when these lockups occur log entries indicating what things are causing them are not being written to disk.
I had no issues with 6.2.x myself - I wasn't meaning to imply that all users were having trouble. Though there clearly are a lot of users who've run into the loss of access to the Web UI, and a few who've seen the lockups.
I'm using 6.3.3 on my RN102 at present, and so far it is working ok.
I tried a downgrade to 6.2.1 and the NAS started flooding the network with activity resulting in my DSL router freezing up. I am not sure what it was trying to do. But disconnecting the patch cord resulted in the router and switch behaving normally. The NAS was flashed with 6.2.2 and is now running fine.
The NAS is normally shut down at the end of the day. But this time it was left powered on and has been running for over 24 hours continuously. No issues or lockups. All seems normal.
The box is nearly always idle. No automated backups tasks are being performed. Activity is restricted to manually initiated directory/folder syncs from my PC to shared folders on the NAS. These synchs are performed once or twice a day. So far so good. <fingers crossed>
Running on 6.2.0 firmware the NAS got to the 4th day running straight without lockups then when I came into work on the 5th day it was locked up again. Does sound like people are getting more days before lockup after downgrading their firmware. Any chance we can downgrade to 6.1.x firmware for testing? I would be happy to go back to this magical firmware that just worked...
We have no plans for any further 6.1.x firmware. There are several reasons for this.
We made changes in 6.2.x which are one way, for example a major upgrade to snapshots. Once you have updated to 6.2.x you cannot downgrade back to 6.1.x.
We have added a number of new features, enhancements and bug fixes in 6.2.
We would like to get to the bottom of why your systems are locking up and if a firmware update is necessary to fix it to get it addressed in a firmware update.
Yesterday I stepped back to Firmware 6.2.0 but my RNAS 104 froze again within a few hours. It looks like for me this downgrade didn't improve my situation.
By the way, should we still provide logs when freezes happen? (I would have no problem to upgrade to Beta3 again).
