[Support] Linuxserver.io - Nextcloud

[Geck0]

2 hours ago, Tucubanito07 said:

Can someone please look in there database and see if what you guys have under oc_jobs please? I don't know how to fix this issue and would like to see if someone can see what keys they have under the oc_jobs please. Any help would be greatly appreciated.

 

My suggestion is to restore your database and Nextcloud instance. I've only glanced at your posts, but my suggestion is to roll back to v 27. Disable your apps before upgrading. Don't upgrade to major versions. You need to do the intermediary upgrades as well. Go to the upgrade page and check that there are no database occ commands to be run before jumping to the next version. If you edit the database, you're going to have more problems later that will likely be harder to fix. You could also try using occ commands to repair and/or database maintenance. Remember to run an occ fingerprint after each upgrade.

[Geck0]

On 9/17/2024 at 4:33 AM, fritz_hh said:

29.0.7 
 

Dear all, 
For some time now, no files have been displayed in the browser under “apps/files/files” (All files), only the current circle appears. Since today's update, however, this is also the case under “/personal” (personal files). 

Under “/recent”, “/shareoverview”, “/tags” everything is still displayed correctly. 

I have already carried out every conceivable file operation and maintenance operation. But this does not bring any changes. 

Everything works perfectly in the app on the smartphone. 

Do you have the same problem? 

 

 

Have you done an occ files:scan --all?

Have you cleared your cookies and browser history?

You're not supplying sufficient information for people to respond, what do your logs state?

Are you using Elasticsearch as a separate docker?

 

Also, you may wish to look at these commands, run with occ from the command line and consider what details to post on the forum.

To run these commands, in CLI run "docker exec -it [exact name of docker] occ"

 

1.  check                                   check dependencies of the server environment
2.   completion                              Dump the shell completion script
3.    setupchecks                             Run setup checks and output the results
4.   status                                  show some status information 

Its possible that occ files:scan-app-data and / or files:cleanup (which cleans up the cache and I'm leaning towards this one, as it sounds like a cache issue) will resolve your issue, but I cannot state this definately, as you haven't given sufficient information. If I was you, I wouldn't run commands without first making proper backups of all data. Nextcloud is pretty good at not touching your information during an upgrade and afterwards, but you never know. Although, you've probably done this already.

 

 

When I run Nextcloud upgrades:

1. I have a recent backup of all my data.

2. I run a recent appdata backup, it takes way less than dealing with upgrade errors.

3. Disable apps that don't like upgrades, mainly elasticsearch apps tend to fail. I will then enable them after the upgrade manually.

4. Apps that don't like being upgraded manually afterwards (elasticsearch, again, is one of them), I tend to do via occ manually and that always seems to work. You normally get a non-compatible version error message via the gui.

 

To avoid:

5. Editing the database to correct errors can work, but it can unravel when you try and upgrade again.

6. Open user accounts, particularly with Talk, Circles, etc are a pain to get rid of, once the account is closed. You really want to sort those out before closing the account.dd

 

Let me know if this helps, but I can't offer more help without knowing more info. Also, I'm not a Nextcloud expert, but I've made mistakes in the early days that have taught me quite a bit.

You're also better posting on here, rather than the Nextcloud community forum. The moderator on there is a d*ck!

 

 

 

[fritz_hh]

8 hours ago, Geck0 said:

Have you done an occ files:scan --all?

Have you cleared your cookies and browser history?

You're not supplying sufficient information for people to respond, what do your logs state?

Are you using Elasticsearch as a separate docker?

 

 

If I had a meaningful error message somewhere, then of course I would have dealt with it, but so far I haven't had one. Hence the general question.

As already written, I have already carried out all the occ operations associated with this several times: 

occ files:scan --all

occ files:celanup

occ files:repair-tree

occ files:scan-app-data

occ db:add-missing-indices

 

 

In each case without error message. 

I have of course also deleted the cookies and tested them on devices that have never accessed the nextcloud. 

I will spare the output of “docker exec -it nextcloud occ setupchecks” here, as there are no anomalies there either.

I have Elasticsearch running in a separate Docker, but have already deactivated it for troubleshooting purposes.

Something that appears again and again in the protocol, but was already there before:

[fritz_hh]

I have now deactivated all apps again, but this has not brought any change.

[Geck0]

1 hour ago, fritz_hh said:

I have now deactivated all apps again, but this has not brought any change.

https://help.nextcloud.com/t/log-fills-with-error-appinfo-app-php-is-deprecated-after-upgrade-to-nc20-0-12/122604

 

I found this. To be fair, I was on my phone, so a bit difficult to.do.intensive searching. 

Seems similar to your issue.

[Tucubanito07]

22 hours ago, Archonw said:

 

 

 

Maybe you can recreate this with:

occ recognize:recrawl

 

 

When I try that command it says "There are no commands defined in the "recognize" namespace. Would you happen to know what other command needs to be done with recognize?

[Geck0]

18 minutes ago, Tucubanito07 said:

When I try that command it says "There are no commands defined in the "recognize" namespace. Would you happen to know what other command needs to be done with recognize?

You appear to be trying to run the command from within the docker. You need to run it from Unraid CLI as a docker command.

 

Ie. docker exec -it [name of your nextcloud docker] occ

 

Try that, use the exact name of your Nextcloud docker.

 

 

[Tucubanito07]

20 minutes ago, Geck0 said:

You appear to be trying to run the command from within the docker. You need to run it from Unraid CLI as a docker command.

 

Ie. docker exec -it [name of your nextcloud docker] occ

 

Try that, use the exact name of your Nextcloud docker.

 

 

Ohh okay. I am runnign the repair in the console of the container and keeps giving me this.

 

[Tucubanito07]

23 minutes ago, Geck0 said:

You appear to be trying to run the command from within the docker. You need to run it from Unraid CLI as a docker command.

 

Ie. docker exec -it [name of your nextcloud docker] occ

 

Try that, use the exact name of your Nextcloud docker.

 

 

I just did what you recommended and still got the same error.

[Geck0]

1 hour ago, Tucubanito07 said:

I just did what you recommended and still got the same error.

I seem to recall you had a failed upgrade by trying to use the "latest label". I take it you can't restore previous versions of your database and nextcloud, which would be easier.

 

What happens when you type "occ upgrade"?

[Geck0]

I would stop trying to use recognise:recrawl, that's not going to help you with your problem.

[fritz_hh]

4 hours ago, Geck0 said:

https://help.nextcloud.com/t/log-fills-with-error-appinfo-app-php-is-deprecated-after-upgrade-to-nc20-0-12/122604

 

I found this. To be fair, I was on my phone, so a bit difficult to.do.intensive searching. 

Seems similar to your issue.

Unfortunately, this did not solve my problem, but I then realized that cron.php was not running correctly due to the backup app. I have now deactivated it so that cron runs correctly. And I have no more error entries in the log.



However, the problem with the files has still not been solved. And I suspect that it may have something to do with write permissions and owners. But why the problem is only occurring now is a mystery to me... 

[Tucubanito07]

2 hours ago, Geck0 said:

I seem to recall you had a failed upgrade by trying to use the "latest label". I take it you can't restore previous versions of your database and nextcloud, which would be easier.

 

What happens when you type "occ upgrade"?

So, I have a backup and from the 9/8/2024 but for whatever reason Backup/Restore Appdata is giving me this error. 

 

 

So, I was able to get the tar.zst file and get the OC_JOBS and place it on the current docker mariah database and the issue still persists. I change the permissions and the user for the file to be like if it was never changed, but didnt work.

 

 

The error says that the table does not exist in engine. Would that be the Nextcloud container instead mariadb database? I also did the OCC Upgrade like you said and get the same error.

 

Also, I tried to go to the same version 29.0.5.1 instead of latest and it does not work. I would think since I don't want to update it will work but it does not. I don't remember the error code though.

[Geck0]

4 hours ago, Tucubanito07 said:

So, I have a backup and from the 9/8/2024 but for whatever reason Backup/Restore Appdata is giving me this error. 

 

 

So, I was able to get the tar.zst file and get the OC_JOBS and place it on the current docker mariah database and the issue still persists. I change the permissions and the user for the file to be like if it was never changed, but didnt work.

 

 

The error says that the table does not exist in engine. Would that be the Nextcloud container instead mariadb database? I also did the OCC Upgrade like you said and get the same error.

 

Also, I tried to go to the same version 29.0.5.1 instead of latest and it does not work. I would think since I don't want to update it will work but it does not. I don't remember the error code though.

If you do a restore from backup, in this case I suggest you restore both the database and Nextcloud. 

PLEASE MAKE SURE YOU HAVE BACKUPS OF ALL YOUR DATA. Its unlikely that you would lose anything doing the below, but better safe than sorry.

I'm also going to point out that I'm not an expert, but I've had similar issues in the beginning.

 

I think the issue you have, is that you've tried to upgrade Nextcloud from one major version to the next, which hasn't worked, but the database will have entries that have been partially upgraded. So you need to roll back both of them from a previous backup that was working.

Don't use the restore option for appdata unless you want to overwrite all of your appdata folder. You also may want to backup the current appdata, in case you mess up. Stop your Nextcloud and MariaDB dockers first. Plus Redis, Elasticsearch or any other dockers that are related.

 

So, if I have an issue with Nextcloud, I will do the following:

 

1.  Manually extract the chosen backup's MariaDB (I'm assuming its MariaDB) and Nextcloud from the tar backup. Its easier to extract to a separate folder that you can delete later.
2. Delete your appfolder's Nextcloud and MariaDB folders. If you don't, you're amalgamating files. (I normally delete the folders and extract the two folders directly into appdata, but I don't know what your experience level is, I'm keeping it simple).
3. Move the extracted folders into appdata.
4. Change your Nextcloud docker back to the tag that you were on before. If you don't know what I mean by this, then don't go any further and wait for it to install. If you use the wrong version, you're likely going to have a similar problem, as its going to be a mismatch. (could happen, not saying it will, depends on the version).
5. I'm assuming that you didn't upgrade your MariaDB docker, so that can stay as it is.
6. Start MariaDB first and any other containers that are related to Nextcloud. If you use Elasticsearch as a separate docker it should be up and running before Nextcloud. I've had Nextcloud fail to start with Elasticsearch down.
7. Start Nextcloud and look at the docker log whilst its starting, for any errors.
8. If its up and running, check the Admin overview page for any messages of errors and the main log file. Its likely that it will tell you to run an occ command in the overview page. If it does don't do anything yet.
9. At the most, at this stage, run a database fingerprint command and a files:scan --all command. Both should be run after each upgrade, in my opinion. Otherwise you may not be able to see your files.

Once its up and running using the previous version, post back with more info. I would do those 9 steps without cutting corners. No. 8 is important to run the occ commands in the Admin overview page, but you may want to post first.

A couple of things:

What do you mean by this:

Quote

So, I was able to get the tar.zst file and get the OC_JOBS and place it on the current docker mariah database

Is the database called OC_JOBS, or did you copy the database table OC_JOBS?? It sounds like the latter.

 

As a general rule, don't copy parts of the database over to correct a problem. Roll back to the most recent working version of Nextcloud and MariaDB, using your backup and then start again. Its quicker that way.

I strongly recommend that you set your Nextcloud version, don't leave it on :latest. If it auto upgrades and its an issue, its a pain later to fix. If you want to use the :latest tag, as a minimum set your dockers to auto upgrade AFTER an appdata backup has run. 

 

When new Nextcloud versions are out, I copy the tag from Linuxserver.io and paste it into Nextcloud's config to upgrade manually. I never auto update my MariaDB, I also don't upgrade it unless I have to. I do this for each version and I generally wait until a few people have upgraded, so I can see if its an issue. Your main issue, from what I can see, is not updating frequently and then jumping to another main version. That really works. You don't have to do every release, but you should be doing at least the final version before the next main upgrade, otherwise you're not getting all the corrections and bug fixes.

 

 

 

 

 

[Geck0]

8 hours ago, fritz_hh said:

Unfortunately, this did not solve my problem, but I then realized that cron.php was not running correctly due to the backup app. I have now deactivated it so that cron runs correctly. And I have no more error entries in the log.



However, the problem with the files has still not been solved. And I suspect that it may have something to do with write permissions and owners. But why the problem is only occurring now is a mystery to me... 

 

 

That's not the issue, here's mine. You generally don't have to play around in that folder anyway.

 

 

Here is the main folder in docker:

 

Personally, I would roll back to the previous stable version you've got and change the data folder permissions, then upgrade correctly. See my previous post to the other chap that is having issues, that should work for you as well. Assuming you have a decent backup of a working Nextcloud version that is recent and you follow all 9 steps.

 

You don't generally have to go into your docker, the important files are already under your appdata folder and in your Nextcloud data folder. Running docker exec -it "nextcloud" occ is generally all you need to do to manage the docker.

 

 

 

 

[Tucubanito07]

2 hours ago, Geck0 said:

If you do a restore from backup, in this case I suggest you restore both the database and Nextcloud. 

PLEASE MAKE SURE YOU HAVE BACKUPS OF ALL YOUR DATA. Its unlikely that you would lose anything doing the below, but better safe than sorry.

I'm also going to point out that I'm not an expert, but I've had similar issues in the beginning.

 

I think the issue you have, is that you've tried to upgrade Nextcloud from one major version to the next, which hasn't worked, but the database will have entries that have been partially upgraded. So you need to roll back both of them from a previous backup that was working.

Don't use the restore option for appdata unless you want to overwrite all of your appdata folder. You also may want to backup the current appdata, in case you mess up. Stop your Nextcloud and MariaDB dockers first. Plus Redis, Elasticsearch or any other dockers that are related.

 

So, if I have an issue with Nextcloud, I will do the following:

 

1.  Manually extract the chosen backup's MariaDB (I'm assuming its MariaDB) and Nextcloud from the tar backup. Its easier to extract to a separate folder that you can delete later.
2. Delete your appfolder's Nextcloud and MariaDB folders. If you don't, you're amalgamating files. (I normally delete the folders and extract the two folders directly into appdata, but I don't know what your experience level is, I'm keeping it simple).
3. Move the extracted folders into appdata.
4. Change your Nextcloud docker back to the tag that you were on before. If you don't know what I mean by this, then don't go any further and wait for it to install. If you use the wrong version, you're likely going to have a similar problem, as its going to be a mismatch. (could happen, not saying it will, depends on the version).
5. I'm assuming that you didn't upgrade your MariaDB docker, so that can stay as it is.
6. Start MariaDB first and any other containers that are related to Nextcloud. If you use Elasticsearch as a separate docker it should be up and running before Nextcloud. I've had Nextcloud fail to start with Elasticsearch down.
7. Start Nextcloud and look at the docker log whilst its starting, for any errors.
8. If its up and running, check the Admin overview page for any messages of errors and the main log file. Its likely that it will tell you to run an occ command in the overview page. If it does don't do anything yet.
9. At the most, at this stage, run a database fingerprint command and a files:scan --all command. Both should be run after each upgrade, in my opinion. Otherwise you may not be able to see your files.

Once its up and running using the previous version, post back with more info. I would do those 9 steps without cutting corners. No. 8 is important to run the occ commands in the Admin overview page, but you may want to post first.

A couple of things:

What do you mean by this:

Is the database called OC_JOBS, or did you copy the database table OC_JOBS?? It sounds like the latter.

 

As a general rule, don't copy parts of the database over to correct a problem. Roll back to the most recent working version of Nextcloud and MariaDB, using your backup and then start again. Its quicker that way.

I strongly recommend that you set your Nextcloud version, don't leave it on :latest. If it auto upgrades and its an issue, its a pain later to fix. If you want to use the :latest tag, as a minimum set your dockers to auto upgrade AFTER an appdata backup has run. 

 

When new Nextcloud versions are out, I copy the tag from Linuxserver.io and paste it into Nextcloud's config to upgrade manually. I never auto update my MariaDB, I also don't upgrade it unless I have to. I do this for each version and I generally wait until a few people have upgraded, so I can see if its an issue. Your main issue, from what I can see, is not updating frequently and then jumping to another main version. That really works. You don't have to do every release, but you should be doing at least the final version before the next main upgrade, otherwise you're not getting all the corrections and bug fixes.

 

 

 

 

 

Dude thanks so much for the help. I got one of my docker containers to work. Now, what do you mean by run a database fingerprint command? I have never ran this so do you know the command?

[Geck0]

54 minutes ago, Tucubanito07 said:

Dude thanks so much for the help. I got one of my docker containers to work. Now, what do you mean by run a database fingerprint command? I have never ran this so do you know the command?

After an upgrade, ideally run this:

 

Quote

occ maintenance:data-fingerprint

     

Effectively, updates Nextcloud with files that have been synced, etc. 

Then run this:

Quote

occ files:scan --all 

   Effectively tells Nextcloud which files are in the system.

Think about it, you restore an older version. Older version doesn't know whats been synced or what files are there. Nextcloud will not show files that you have added recently, because "old Nextcloud" doesn't know they're there.           

(You can also do this if you add files manually to the data folder and want Nextcloud to discover them. If you add them to a specific user folder ie. username geck0, then you can do it like this for that user:

occ files:scan geck0

Otherwise you're updating for all users, which could take a while.

 

These are also likely. You shouldn't / shouldn't need to run them unless you see it as recommended in Admin overview page. You should always check there and your log after an upgrade or restore. NC may be running, but have errors. Don't run any old command to adjust your database willy nilly. You'll break it (possibly)

 

Quote

 
  occ db:add-missing-columns                  Add missing optional columns to the database tables
  occ db:add-missing-indices                  Add missing indices to the database tables

This is also common, make sure you have backups, in case it clears files you may still want.

 

Quote

 occ files:cleanup                           cleanup filecache

 

I hardly have to use repair mode. Maintenance mode is useful, particularly if an upgrade is stuck in maintenance mode. You can turn it on / off. You can also set this under your config file directly, if you want.

Quote

 occ maintenance:mode                        set maintenance mode
 occ maintenance:repair                      repair this installation

Finally:

occ upgrade will seomtimes get you over the line, if you've got a faulty upgrade. Linuxserver.io disabled this in the GUI and you shouldn't need it. You also shouldn't use it for normal upgrading, if Linuxserver.io enabled upgrading when going to new versions. I haven't used it in ages

 

You can upgrade apps using OCC apps command, which is also my preference. Sometimes the GUI doesn't do it properly after an upgrade and gives you an error message of incompatible app version. I will then remove the app and reinstall via command line.

 

Bottom line, get used to NOT running commands inside the docker, when you can run them from CLI, and don't edit files or database unless you absolutely have to. Modifying a DB or main files will leave you struggling to upgrade or get rid of errors in your logs. If you can't edit it in appdata folder, then ask if you really should be?

 

docker exec -it "nextcloud dockername" occ 

Will bring up the list of available commands and their description. I use this mainly and hardly log into admin anymore. Even then, I hardly need to use a lot of them. If you see a command, then Google it, so you understand it. The NC forum is okay to find stuff, but I can't stand the moderator. He needs to find something else to do that he finds enjoyable!

 

Hope that gives you a headstart. This is one of the better threads on the unRaid forum, I would follow it if I was you.

I'm happy to help, again I'm not an expert, but I know how frustrating it can be, learned the hard way.

 

 

[Tucubanito07]

2 minutes ago, Geck0 said:

After an upgrade, ideally run this:

 

     

Effectively, updates Nextcloud with files that have been synced, etc. 

Then run this:

   Effectively tells Nextcloud which files are in the system.

Think about it, you restore an older version. Older version doesn't know whats been synced or what files are there. Nextcloud will not show files that you have added recently, because "old Nextcloud" doesn't know they're there.           

(You can also do this if you add files manually to the data folder and want Nextcloud to discover them. If you add them to a specific user folder ie. username geck0, then you can do it like this for that user:

occ files:scan geck0

Otherwise you're updating for all users, which could take a while.

 

These are also likely. You shouldn't / shouldn't need to run them unless you see it as recommended in Admin overview page. You should always check there and your log after an upgrade or restore. NC may be running, but have errors. Don't run any old command to adjust your database willy nilly. You'll break it (possibly)

 

This is also common, make sure you have backups, in case it clears files you may still want.

 

 

I hardly have to use repair mode. Maintenance mode is useful, particularly if an upgrade is stuck in maintenance mode. You can turn it on / off. You can also set this under your config file directly, if you want.

Finally:

occ upgrade will seomtimes get you over the line, if you've got a faulty upgrade. Linuxserver.io disabled this in the GUI and you shouldn't need it. You also shouldn't use it for normal upgrading, if Linuxserver.io enabled upgrading when going to new versions. I haven't used it in ages

 

You can upgrade apps using OCC apps command, which is also my preference. Sometimes the GUI doesn't do it properly after an upgrade and gives you an error message of incompatible app version. I will then remove the app and reinstall via command line.

 

Bottom line, get used to NOT running commands inside the docker, when you can run them from CLI, and don't edit files or database unless you absolutely have to. Modifying a DB or main files will leave you struggling to upgrade or get rid of errors in your logs. If you can't edit it in appdata folder, then ask if you really should be?

 

docker exec -it "nextcloud dockername" occ 

Will bring up the list of available commands and their description. I use this mainly and hardly log into admin anymore. Even then, I hardly need to use a lot of them. If you see a command, then Google it, so you understand it. The NC forum is okay to find stuff, but I can't stand the moderator. He needs to find something else to do that he finds enjoyable!

 

Hope that gives you a headstart. This is one of the better threads on the unRaid forum, I would follow it if I was you.

I'm happy to help, again I'm not an expert, but I know how frustrating it can be, learned the hard way.

 

 

Thanks you for all your help. I dont see any errors. All I see is this on the setup warning page. I am currently running the occ files:scan --all. 

[Geck0]

Does't mean anything without logs or checking your setup. You need to get used to posting more relevant detail.

Use your new found powers to troubleshoot. Teach you to fish

 

1. Check your logs under admin to see what's being reported.
2. Use the occ command to check your setup (look for the right command), it will tell you what's wrong.

Hint: Its one of the first commands in the help list.

[Tucubanito07]

8 minutes ago, Geck0 said:

Does't mean anything without logs or checking your setup. You need to get used to posting more relevant detail.

Use your new found powers to troubleshoot. Teach you to fish

 

1. Check your logs under admin to see what's being reported.
2. Use the occ command to check your setup (look for the right command), it will tell you what's wrong.

Hint: Its one of the first commands in the help list.

I did check the logs and there are no errors or warning. Thats why I said they were no errors other than the one i posted.

 

Also, I have used both commands occ maintenance:data-fingerprint and occ files:scan --all. After running those commands, it looks good and how it used to be.

 

I am going to try these occ maintenance:mode and occ maintenance:repair tomorrow and see if I can get that error to go away. I really appreciate your help. Because of you i was able to get it back up and running. You rock man. Thanks.

Edited 14 hours ago by Tucubanito07

[Geck0]

Remember what I said, read up on what the commands do first.

I wouldn't be running maintenance:repair immediately. Maintenance mode is only useful if you want to block processes whilst maintenance is going on. ie. You're doing something "maintany" and somebody is trying to upload huge files. This will stop that from happening.

If you still have that error message in admin overview. Try something simple, like clear your cookies first and / or get it to rescan. Sometimes rebooting the docker will clear a message like that.

 

Good luck!

[Tucubanito07]

2 minutes ago, Geck0 said:

Remember what I said, read up on what the commands do first.

I wouldn't be running maintenance:repair immediately. Maintenance mode is only useful if you want to block processes whilst maintenance is going on. ie. You're doing something "maintany" and somebody is trying to upload huge files. This will stop that from happening.

If you still have that error message in admin overview. Try something simple, like clear your cookies first and / or get it to rescan. Sometimes rebooting the docker will clear a message like that.

 

Good luck!

I keep getting a 504 gateway when it goes to https://nextcloud.com/settings/ajax/checksetup. I have to look into it. Dude for real, thanks. I believe is my firewall blocking it. I have my network lockdown so most likely that.

Edited 13 hours ago by Tucubanito07