fersal Posted September 16, 2018 Share Posted September 16, 2018 While checking the log to determine what was causing some drives to spin up, I saw an entry that mentioned "Warning: Rootfs file is getting full (currently 81 % used)". What causes this usually, and how can I prevent it? I'm attaching diagnostics file, thanks for your help! tower-diagnostics-20180916-0538.zip Quote Link to comment
itimpi Posted September 16, 2018 Share Posted September 16, 2018 Difficult to say! Most likely cause is a misconfigured docker that is writing files to RAM. The thing to remember is that all files that are not under /mnt or /boot are in RAM. Quote Link to comment
fersal Posted September 16, 2018 Author Share Posted September 16, 2018 Thanks, as far as I know I have all my data files under /mnt. I tried looking at the Diagnostics file but I don't know where to look... has someone else had this issue before? Quote Link to comment
trurl Posted September 16, 2018 Share Posted September 16, 2018 5 hours ago, fersal said: as far as I know I have all my data files under /mnt Is this a new problem? Have you made any recent changes or additions to dockers? 10 hours ago, itimpi said: Most likely cause is a misconfigured docker that is writing files to RAM. The thing to remember is that all files that are not under /mnt or /boot are in RAM. This is the most usual cause of this problem. And /mnt isn't good enough actually. It must be an actual user share path, /mnt/user/... or an actual disk, /mnt/disk1/... or /mnt/cache/... for example. Just /mnt/somefolder will still be in RAM. And you can also get a folder in RAM if you try to use a path to an unmounted Unassigned Device. 1 Quote Link to comment
fersal Posted September 17, 2018 Author Share Posted September 17, 2018 (edited) 4 hours ago, trurl said: Is this a new problem? Have you made any recent changes or additions to dockers? This is the most usual cause of this problem. And /mnt isn't good enough actually. It must be an actual user share path, /mnt/user/... or an actual disk, /mnt/disk1/... or /mnt/cache/... for example. Just /mnt/somefolder will still be in RAM. And you can also get a folder in RAM if you try to use a path to an unmounted Unassigned Device. Actually no major hardware or software changes in the past year other than upgrade to 6.5.3. However maybe it's happened before because I don't look at the log file very often. The only docker is a Plex Media Server. And yes, all files are in user shares under /mnt/user/; I'll keep monitoring the log file to see of this happens again. If anyone else has any ideas please let me know, thanks! Edited September 17, 2018 by fersal Quote Link to comment
fersal Posted September 17, 2018 Author Share Posted September 17, 2018 (edited) One thing that I did notice was that in the Docker volume mappings for Plex I have everything under /mnt/user/ except one entry that says: /transcode <--> /tmp Could this be causing this issue? Edited September 17, 2018 by fersal Quote Link to comment
trurl Posted September 17, 2018 Share Posted September 17, 2018 /tmp is in RAM but some people set it up that way on purpose so transcodes go to RAM. I guess it depends on how much RAM you have and how big transcodes are. Quote Link to comment
anindianforor Posted November 3, 2018 Share Posted November 3, 2018 I just upgraded to 6.6.3 and installed FCP and ran a diagnostics. It is showing an error for me saying "Rootfs file is getting full (currently 99& used). Not sure why I am getting this now. I don't have docker enabled. Attaching my diagnostics.. storage-diagnostics-20181103-2354.zip Quote Link to comment
Squid Posted November 3, 2018 Share Posted November 3, 2018 Running unRaid 6.6.3 on 2G of memory is really pushing it. I know that LT's website does state that is the requirement, but realistically that is the bottom of the barrel requirement. 1 Quote Link to comment
anindianforor Posted November 3, 2018 Share Posted November 3, 2018 (edited) 1 hour ago, Squid said: Running unRaid 6.6.3 on 2G of memory is really pushing it. I know that LT's website does state that is the requirement, but realistically that is the bottom of the barrel requirement. Thanks for responding and checking my logs! How much should be the RAM then? Any recommendation? Edited November 3, 2018 by anindianforor Quote Link to comment
Squid Posted November 3, 2018 Share Posted November 3, 2018 2 hours ago, anindianforor said: Thanks for responding and checking my logs! How much should be the RAM then? Any recommendation? TBH, even with no additional apps running, I'd say 4G is the functional minimum Quote Link to comment
bombz Posted August 8, 2021 Share Posted August 8, 2021 On 11/3/2018 at 6:19 PM, Squid said: TBH, even with no additional apps running, I'd say 4G is the functional minimum Hello, I have posted my diagnostics for some assistance. Keep in mind this NAS is sort of a test NAS I created from old hardware and upgraded RAM and CPU to what the motherboard would support. Fix uncommon problems is reporting: Rootfs file is getting full (currently 92 % used) Is this cause of the RAM? unraidNAS-diagnostics-20210808-1202.zip Quote Link to comment
Squid Posted August 8, 2021 Share Posted August 8, 2021 You have 2GB of memory On 11/3/2018 at 6:19 PM, Squid said: I'd say 4G is the functional minimum Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.