First off, I posted this issue before, but didn't get much traction or assistance.

Link to that post is here

 

So since then the issue has cropped up again. I've since updated to 6.7.2 thinking that it may help resolve it, but within a couple days of updating to 6.7.2, I encountered a similar issue. It does present itself differently than it did in 6.6.7 though.

 

Here's a couple screenshots of what I'm seeing in the UI

Main page

 

And the apps page

 

The UI is largely unresponsive. I cannot reboot it from the UI, it must be done via terminal. Once it's rebooted, everything works for a while, but after a few days the issue will crop up again. I have not rebooted yet, so the system is still in this state (has been since last night sometime), but Plex and everything still works. I tried to get a diagnostics logs, but unfortunately it won't download. I can click download, but then I get a blank page, and nothing happens. Not sure if there is any other way to get a diagnostics log elsewhere, or something I can run in terminal.

 

Any help with this would be greatly appreciated.

Type diagnostics on the console.

So it appears it's a disk space issue?

 

It's saying Warning: file_put_content(): Only 0 of 29 bytes written, possibly out of disk space in /usr/local/emhttp/plugins/dynamix/scripts/diagnostics on line 158

 

I'm fairly certain I am not low on disk space on any drive. Last I checked I was only using ~48gb of my 500gb cache drive, and had around 9tb of space on my Array available. The USB stick was a 32gb, and should be using less than 1gb total.

More likely it's a flash drive problem.

Just curious, how would this be a USB issue? The issue goes away upon boot, and it's my understanding that the entire OS is loaded into memory (RAM),  so the USB would barely be used. In fact should it really only be used on boot, and that's pretty much it? Would it make sense that maybe it's an issue with my RAM since the issue crops up not during or after a boot, but rather after several days of being online?

 

To note, I'm using a SanDisk Cruzer Fit.. any recommendations if it is a USB issue? I've only been running unraid for ~5-6 months and it was a brand new USB drive.

Edited July 16, 20196 yr by NVS1

Problems you are having are consistent with the flash drive dropping offline, if you're using a USB 3.0 port try a 2.0 instead.

Ah interesting. I'll have to take a look and see what I can do there. At work right now, but will check it out this evening.

 

Thanks!

I'd say its more the rootfs is full, due to a misconfigured app somewhere along the line, or a script attempting to save stuff onto an unassigned device that isn't present..

6 hours ago, Squid said:

I'd say its more the rootfs is full, due to a misconfigured app somewhere along the line, or a script attempting to save stuff onto an unassigned device that isn't present..

This could be entirely true. I'm not familiar with rootfs, when searching around I found another post that you had made last year. You mentioned that it's possible that a docker image is misconfigured and writing to RAM. Another user also commented that having something pointed to /mnt/ is not good enough, it needs to be /mnt/user, /mnt/disk1/, or /mnt/cache, etc.

 

The reason I mention this is because it may be my rclone setup, or even plex setup. Now my plex setup has remained largely how it is for a while, and I don't believe there is any issues with it. All my docker images are pointing to a variation of /mnt/user/ or /mnt/cache/.. except for 1 setting. On my Plex setup, I have the /media flag pointed to /mnt.... I don't believe this is necessarily an issue, as I don't believe anything is being written here. All this does is allow Plex to see the file structure from /mnt on. This helped me when pointing my TV or Movie library's to the correct file location. Correct me if I'm wrong here.

 

That all aside, I'm leaning to more of a rclone issue.

 

I setup rclone a while back, ran into issues, and it wasn't working. I got busy and let it sit for a few weeks before going back to it. I  worked on it, and got it working and I estimate not long after that was the first time I encountered this issue. Now I have rebooted my unraid server as I was trying to look at my rclone scripts. My rclone scripts are as follows:

 

Rclone Config

[gdrive]
type = drive
client_id = xxxxxxxxxxxxxxxxxxxxxxxxxxx.apps.googleusercontent.com
client_secret = xxxxxxxxxxxxxxxxxxxx
scope = drive
token = {"access_token":"xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx","token_type":"Bearer","refresh_token":"1/-AKF-xxxxxxxxxxxxxxxxxxxxxxx","expiry":"2019-07-17T00:10:59.87668022-04:00"}

[gcache]
type = cache
remote = gdrive:/gdrive
plex_url = http://192.168.1.43:32400
plex_username = xxxxxxxxxx
plex_password = xxxxxxxxxxxxxxxxxx
chunk_size = 5M
info_age = 1d
chunk_total_size = 10G

[gcrypt]
type = crypt
remote = gcache:/crypt
filename_encryption = standard
directory_name_encryption = true
password = xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
password2 = xxxxxxxxxxxxxxxxxxxxxxxxx


Rclone Mount Config

# Local mountpoint
mntpoint="/mnt/disks" 		 

# Remote share
remoteshare="gdrive:gcrypt" 	


#---------------------------------------------------------------------------------------------------------------------

#!/bin/bash
#----------------------------------------------------------------------------
#first section makes the folders for the mount in the /mnt/disks folder so docker containers can have access

mkdir -p /mnt/disks/gdrive
mkdir -p /mnt/disks/crypt


#This section mounts the various cloud storage into the folders that were created above.

rclone mount --max-read-ahead 1024k --allow-other --allow-non-empty gdrive: /mnt/disks/gdrive &
rclone mount --max-read-ahead 1024k --allow-other --allow-non-empty gcrypt: /mnt/disks/crypt &

I suspect that might be my issue. I'm pointing my rclone mntpoint to /mnt/disks.

 

I'll admit I followed a guide I found online for setting it up. I'm not well versed with rclone or traversing *nix file systems as you can probably tell. 

Edited July 17, 20196 yr by NVS1

@Squid Any suggestions or ideas based on the information I posted above?

 

Thanks

I suggest you look at DZMM's topic on how to set up rclone properly. Using /mnt/disks is inviting trouble.