Fix Common Problems More Information

[Squid]

Always include your diagnostics if you are creating any new post for additional help with any issues Fix Common Problems finds

 

This thread is locked.  If there are any further questions / clarifications needed, then use the associated questions thread. https://forums.unraid.net/topic/120244-fix-common-problems-more-information-feedback/

 

 

MODERATORS: Feel free to change or expand upon any instructions etc given here, but DO NOT add in (or delete) any additional replies.  These posts are all directly linked from within FCP

[Squid]

Share xxx is an implied array only share, but files / folders exist on the cache drive.

 

This error means that files or folders in your share exist on one or more of the cache pools, but the share's settings have never been adjusted and it is implied that the share's settings are to only store files within the array.

 

There are some rare edge-cases where you would want to do this, but generally it is not how you would set up the system.  Additionally, mover will not take any action in this circumstance.

 

This can also easily happen if within an app you have the docker directly referencing /mnt/cache/someshare instead of /mnt/user/someshare 

 

The easiest way to handle this is to go to Shares, select your share and set the Use Cache settings to be "yes".  New files will be written to the cache drive, and mover will move them to the array at the scheduled time.

 

If you don't want any files to be stored ever on the cache drive, then once again set the Use Cache settings to be "yes", then manually run mover from the Main tab.  After the move is completed, change the Use Cache settings to be "no"

 

NOTE: mover cannot move any files which are in use at the time, so the best way to accomplish the second scenario is to disable both VM's and Docker from the Settings tab.

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Share xxx is set to cache only, but files / folders exist on the array

 

This means that your share settings are telling the OS to only put files onto the cache pool, but for one reason or another files exist on the array.

 

This could happen if you have added a cache pool to a system previously set up without one.  

 

This most likely explanation for why this actually happened is that when adding the pool you didn't move the pre-existing files onto the cache drive. 

 

The ideal solution here is to change the Use Cache settings for that share to be "prefer", stop the VM and Docker services within Settings and then run mover. 

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Share xxx references non existent cache pool yyy

 

This error pops up if you have renamed or removed a cache pool from your system.  When doing this, the OS does not automatically adjust any existing share settings, and behaviour of your files may have unexpected results.  Your easiest solution is to simply change the reference to which cache pool the share is using within it's share settings

 

Share xxx set to use pool yyy, but files exist on zzz

 

This error can happen when re-arranging your system's cache pools.  At some point your share's settings were set to use a certain pool, but you've changed the settings and now there are abandoned files on the original pool

 

If you're running an OS version < 6.10-rc3 then the solution would be to.

 

1. Ideally, disable the docker and VM services in the settings tab
2. Change the applicable share's settings to be use cache: yes, and for the pool set it back to the original pool
3. Run mover from the Main Tab
4. Change the share's settings back to reference the new pool.
5. If the share in question should only have files present on the new pool, then change the share's settings to be use cache: prefer and re-run mover

 

On Unraid versions 6.10-rc3+ you also have the option of using Dynamix File Manager to manually move the files from one cache pool to another.

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Share xxx is to to not use the cache, but files / folders exist on the cache drive

 

This means that your share's settings are set to Use Cache: NO, but files / folders still exist on the cache drive

 

Assuming that you want all files to go directly to the array, your solution is to

1. Stop the docker and VM services within Settings
2. Change the share's settings to be Use Cache: yes and then run mover
3. After mover is finished, set the use cache settings back to be "no" and re-enable the docker and VM services

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Plugin update check not enabled

 

Plugins are like any other software.  Updates for them are constantly being released to fix any bugs or security issues or add additional features.  Your system is set to never check for plugin updates.  This can be fixed by going to Settings - Notification Settings and enabling the OS to check for updates on a schedule

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Docker update check not enabled

 

Dockers are like any other software.  Updates for them are constantly being released to fix any bugs or security issues or add additional features.  Your system is set to never check for plugin updates.  This can be fixed by going to Settings - Notification Settings and enabling the OS to check for updates on a schedule

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Fix Common Problems not set to auto update

 

Fix Common Problems is constantly being updated with additional tests, and it is highly recommended that it be automatically updated as new versions are released.  This test is only performed if the Auto Update plugin is also installed, and an error results if Fix Common Problems is set to not automatically update within Auto Update's settings

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Same share xxx exists in a different case

 

Unraid runs on Linux.  Linux allows (and it perfectly reasonably) to have a share named Movies, another named movies, and a third named MoViEs and they are all treated separately.  However, operating systems like Windows make no differentiation between shares only differing in name by "case"

 

This results in your files appearing to disappear when attempting to access them over the network.

 

The most common cause of this issue would be that you have a share named "Movies", but a docker app is referencing a share named "movies".

 

The solution here is to 

1. Check all your settings for all of your docker apps (you might need to click "Show More Settings") to ensure that all of the paths are consistent with how you actually want your system to run
2. Manually move the files from one share to another.  This can be easily accomplished via a file manager such as the Krusader App, or if running 6.10-rc3+ Dynamix File manager).  There is unfortunately no easy automated method to move the files from one share to another

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Unable to communicate with GitHub

 

(Note: many other tests will also fail if this test fails, as Fix Common Problems does have to download auxiliary files to perform certain tests)

 

This error will appear if your server is unable to communicate with GitHub.com  This is where much of the OS has to communicate with in order to check for updates, download icons etc.

 

The most common problem causing this is that of DNS resolution, and the easiest solution is to simply set static DNS addresses within Settings - Network Settings of your favourite DNS server (I'll suggest 208.67.222.222 and 208.67.220.220)  Do not use your router's IP address as a DNS server.

 

Additionally, this can also get triggered if you happen to have the OS running through a pfSense computer or VM or piHole (to block ads) etc.  There is no real reason to do this, as the OS contains no ads at all, has no tracking etc and the extra layer may impede your server's ability to communicate with the outside world.  Ensure that GitHub, dockerHub etc are all whitelisted within the pfSense / piHole's settings

 

Some users in certain geographical areas (eg: China) may also see this errors due to government policies in those jurisdictions.  Solutions to that problem can easily be googled (or searched here on the forum) and are outside the scope of this posting.

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

 

[Squid]

Your drive is either completely full or mounted read-only

 

For the array, your drive is most likely completely full.  If this is an issue and you're not aware of it then you will have to manually move files around between disks using either the Krusader app or Dynamix File Manager (6.10-rc3+ only)

 

Usually however this error will notify you about the docker image being full or read-only.

 

This will happen usually because of an improperly configured docker app where within its internal settings it is storing information directly within the docker image instead of to an external share.  EG: On your download client you want the downloads to go to a share /mnt/user/Downloads which is mapped to /downloads, but within the app itself you're telling the app to store the files in /Downloads (linux is case sensitive)

 

There are many videos by space invader one and multiple topics in the Docker FAQ here in the forum about these types of problems

 

Alternatively, this may also happen on the cache drive itself (which in turn will affect the docker image) if the file system is corrupted, you have bad memory (run a memtest) etc.  For the error happening being caused by anything dealing with the cache drive, it will be best to create a new topic in General Support detailing your problems and always including your diagnostics.

 

 

[Squid]

Default docker appdata location is not a cache only share

 

This is a performance only error.  If you have a cache pool, it is highly recommended that the appdata location (Settings - Docker - Default Appdata path) reference a share that is set to be either Use Cache: Prefer or Use Cache: Only.  

 

While setting this share to be use cache: yes (or no) might seem a good idea (redundancy because it'll be stored on the array), your performance on any given docker container will have a very significant performance reduction.  If your cache pool isn't redundant, then consider making automated backups of your appdata share (eg: via the Appdata Backup / Restore plugin)

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

/var/log is getting full

 

With a correctly functioning server, you should never see this error even with uptimes measured in years.

 

The most common cause of this error is either you have installed the NerdPack plugin and it is set to install "atop"  - Stop NerdPack from installing it.  (NOTE: NerdPack / DevPack is for advanced users only, and if you do not know exactly what you are installing with it and more importantly why you are installing it then you should not be installing it)

 

You have a recurring error being constantly logged that is effectively spamming the syslog and causing it to fill up.

 

You should create a new topic in General Support and always include your diagnostics if you cannot determine the cause (and solution) by yourself.

 

Ultimately though your only recourse to clear this error and have your server functioning again is to reboot the system

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Rootfs is getting full

 

Unraid operates completely from RAM, and if the ram is getting full it will impact adversely your system.

 

The functional minimum amount of RAM for Unraid is 4G of memory.  This amount of RAM will allow you to run a basic file server with a couple of additional apps installed.  The recommended minimum amount of RAM for any purpose other than the basics would be 8GB

 

When running less than the minimums you will probably see this error.  This can also happen due to misconfigured system (eg: you're backing up a share that is supposed to be going to the cloud via rclone plugin, but the "cloud" isn't actually mounted)

 

This error is pretty much always going to require you to create a topic within General Support.  Include both your diagnostics, and the output generated by following the directions in this post:  https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/3/?tab=comments#comment-1090710

 

[Squid]

Share include / excluded drives both set

 

A common mistake many users make in setting up their system is to specify both included and excluded disks in their share settings.

 

Included and excluded are opposites of each other:

 

EG:

 

If you set up a share and set disk 1 to be included, there is no reason to tell the system that disks 2, 3, 4 are excluded at the same time.  The problem with setting both included and excluded simultaneously is that it is quite easy to include and exclude the same disks resulting in a conflict.  While rules within the OS are in place to handle this conflict, the net result of where the files get placed may not be where you actually want them placed.

 

Only specify either included or excluded and never both to avoid any possible conflict or confusion.

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

 

[Squid]

Docker application xxx has volumes being mounted by Unassigned Devices that aren't mounted slave or shared

 

Due to how Docker works, if you have an disk or share being mounted by Unassigned Devices and that device or share is being passed through to a docker container under certain circumstances the docker app may not always see the files present within it.

 

The solution here is to edit the docker app in question, and next to the path that you're mapping to the container hit EDIT.  On the popup window that appears, for the access mode select one of the Slave access modes

 

NOTE: If this is your appdata share that is triggering this error, then to fix it you may have to stop the Docker Service in settings, then enable Docker Authoring Mode, and finally restart the service.  You will then have the EDIT button on that path mapping (You may also have to press Show More Settings)

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Out of memory errors detected on your server

 

On a properly running server this error should never appear.  

 

Effectively, one process on your server requested to use some memory and it wasn't available so the OS killed off a low priority (or very infrequently used) process so that the higher priority process could keep running.

 

The consequences of this error could either be unnoticeable or severe depending upon what process was killed off.  You should create a new post in General Support and include your diagnostics.  Ultimately your recourse to clear this error is a reboot, but the forum users will need to see the diagnostics before you reboot

 

[Squid]

Machine Check Events detected on your server

 

Every CPU has the ability to trigger a Machine Check Event to alert the user of an abnormal situation that it is detecting.

 

Installing the NerdPack plugin and then having it install mcelog (if running OS <6.10) should return more information about the error when it happens again.  (Note that mcelog does not return more information under all hardware combinations)

 

In many circumstances, this error happens during processor initialization due to the upgrading of the processor's microcode etc and is safely able to be ignored.  If you look at the syslog and see the mce looking like this, relatively early in the bootup sequence, then it's harmless.  The start of the CPU initialization is the "smpboot CPU0" line.  Note that the mce was triggered right away.  This mce is harmless

Mar 26 08:43:11 MediaServer kernel: smpboot: CPU0: Intel(R) Core(TM) i7-10700 CPU @ 2.90GHz (family: 0x6, model: 0xa5, stepping: 0x5)
Mar 26 08:43:11 MediaServer kernel: mce: [Hardware Error]: Machine check events logged
Mar 26 08:43:11 MediaServer kernel: mce: [Hardware Error]: CPU 0: Machine Check: 0 Bank 6: ee2000000040110a
Mar 26 08:43:11 MediaServer kernel: mce: [Hardware Error]: TSC 0 ADDR fef1e680 MISC 43880004086 
Mar 26 08:43:11 MediaServer kernel: mce: [Hardware Error]: PROCESSOR 0:a0655 TIME 1648244554 SOCKET 0 APIC 0 microcode e2
Mar 26 08:43:11 MediaServer kernel: Performance Events: PEBS fmt3+, Skylake events, 32-deep LBR, full-width counters, Intel PMU driver.
Mar 26 08:43:11 MediaServer kernel: ... version:                4
Mar 26 08:43:11 MediaServer kernel: ... bit width:              48
Mar 26 08:43:11 MediaServer kernel: ... generic registers:      4
Mar 26 08:43:11 MediaServer kernel: ... value mask:             0000ffffffffffff
Mar 26 08:43:11 MediaServer kernel: ... max period:             00007fffffffffff
Mar 26 08:43:11 MediaServer kernel: ... fixed-purpose events:   3
Mar 26 08:43:11 MediaServer kernel: ... event mask:             000000070000000f
Mar 26 08:43:11 MediaServer kernel: rcu: Hierarchical SRCU implementation.
Mar 26 08:43:11 MediaServer kernel: smp: Bringing up secondary CPUs ...
Mar 26 08:43:11 MediaServer kernel: x86: Booting SMP configuration:
Mar 26 08:43:11 MediaServer kernel: .... node  #0, CPUs:        #1  #2  #3  #4  #5  #6  #7  #8  #9 #10 #11 #12 #13 #14 #15
Mar 26 08:43:11 MediaServer kernel: smp: Brought up 1 node, 16 CPUs
Mar 26 08:43:11 MediaServer kernel: smpboot: Max logical packages: 1
Mar 26 08:43:11 MediaServer kernel: smpboot: Total of 16 processors activated (92796.35 BogoMIPS)

 

In other circumstances, this will be triggered due to an actual hardware fault (bad memory, cache errors on the CPU die etc).

 

If you're not sure about what is causing the error, then create a new post in General Support and include your diagnostics for a forum user to review

 

[Squid]

Possible Hack Attempt

 

Fix Common Problems during it's scans looks for failed login attempts to your server.

 

It is now becoming more common for certain routers to try and be "helpful" by every once in a while attempting to log in to every device on your network (including your server) to look for vulnerabilities (I'm undecided if this however is a vulnerability in the Router itself  )

 

However this will also happen if you have forwarded your management ports on the OS's WebGUI to be openly accessible to the internet at large or have placed your server into the router's DMZ.  If you require access to your server remotely, then always use a VPN or reverse proxy or MyServers to remotely access the server.  Every minute, the entire internet gets scanned by bad actors automatically.  It is only a matter of time before you actually get hacked.

 

There is a threshold of the amount of invalid login attempts allowed that can be adjusted via Fix Common Problems' settings.

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Plugin Not Compatible

 

The lists maintained by Community Applications are constantly monitored and moderated.  If Fix Common Problems is telling you that something is incompatible then it most likely is and will quite possibly cause issues with your server. 

 

If on the other hand a plugin is deprecated, then this means that it is no longer supported.  While it *may* cause problems, it possibly won't.

 

Note that when your server is running a beta or release candidate of the OS, certain plugins may have compatibility issues with the OS.  Simply having issues with a beta or release candidate is not necessarily cause to label a plugin as being incompatible (depending upon the issue).  More leeway is given to the plugin authors in terms of compatibility in this circumstance than when running on a stable version of the OS.

[Squid]

Share name same as disk name or Reserved Share name used

 

These errors will happen if you have a share named something like "disk1".  (Or a shared named "cache" when you also have a cache-pool named "cache")

 

The OS will never allow you to create a share in these circumstances via the OS itself, but if you've created a share inadvertently (eg: via an incorrect mapping on a docker app), this results in the share being created automatically.

 

The consequences of this error are indeterminate because you are winding up using a reserved name for a share (other reserved names would be "user", "user0", "disks", "remotes" etc) and when accessing your files over the network, unexpected results may happen.

 

You would need to move the files out of the offending share via a file manager like Krusader or Dynamix File Manager (6.10-rc3+ only) and then delete the share (and ensure it's not being created by a docker app automatically)

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Directory Read Errors found

 

This error is indicative of two possible issues:

1. Your flash drive has major corruption issues
2. The flash drive has dropped offline and then reconnected itself

 

The only solution in either situation is to reboot.   

 

If these problems continue, then you should ideally be using a USB2 port on the server for your boot flash device.  If these errors continue to happen on a regular basis, then check for corruption issues (place the flash into a barebones (windows) computer and run the file system check on it) or also possibly install the Flash Remount plugin which will recover the OS automatically from a disconnect / reconnect situation

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

xxx.xml file corrupted

 

This error will happen if one or more of the docker template files stored on your flash drive is corrupted.  A corruption could occur if your server's memory is defective, the flash drive is defective, or if the server lost power when writing to the flash drive.

 

You should always run a file system check on the flash drive in a Windows / Mac computer.  The offending file(s) however will not be fixed.  Ideally you would restore the files from a backup you have made (the MyServer's plugin is ideal for doing this).  You may also need to replace the flash drive and transfer your registration to a new one.  https://wiki.unraid.net/Manual/Changing_The_Flash_Device

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[Squid]

Invalid file (or folder) contained within /mnt

 

NOTE: This has been edited for the new recommendation of /mnt/addons instead of /mnt/remotes or /mnt/disks (3/3/23)

 

Unraid maintains exclusive control over files / folders that are stored within /mnt  Any file which may get written there that isn't stored on a disk (/mnt/diskX) or a share (/mnt/user) will be lost on a reboot

 

Similarly, any folder which you are adding to within /mnt should not be there.  This usually happens when using something like rclone (or ZFS) and mounting the remote share within /mnt  You should instead mount the remote path within /mnt/addons for the least amount of problems

 

Additionally, this can also happen if you have a docker container which is directly accessing a non-existent drive (eg: /mnt/cache when you don't have a cache drive).  Usually something like this will happen when you remove or rename a cache pool, as the Apps tab will automatically ensure that the paths when installing any new container correctly match your system)

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems 

[ljm42]

Blank TLD

 

On most networks, the Local TLD field should not be blank. It should be set to "local" (the default) or an actual domain name (if you have a corresponding DNS record for [servername].[LocalTLD])

 

A blank Local TLD is only valid if NetBIOS and SMBv1 are both enabled on the server and all clients. Due to security issues, SMBv1 is deprecated and/or disabled in modern operating systems, including Windows.

 

The Local TLD can be set on the Settings -> Management Access page.

 

If you have any questions, post the full FCP warning and your diagnostics (from Tools -> Diagnostics) to this thread https://forums.unraid.net/topic/47266-plugin-ca-fix-common-problems