Error "Search index out of date. Updating..." and restart of the docker-container every 3 Minutes after upgrading from 2.7.2 to 2.8.2

[Kunibert-007]

Description

After a successful SQLite-upgrade from 2.7.2 to 2.8.2 the webserver fails to start. The docker-container does a restart every 3 Minutes.
I see paperless generates a new index but it restart without any error in the log. Waited 1 hour but it is in a loop:

Paperless-ngx docker container starting...
Creating directory scratch directory /tmp/paperless
Adjusting permissions of paperless files. This may take a while.
Waiting for Redis...
Connected to Redis broker.
Apply database migrations...
Operations to perform:
  Apply all migrations: account, admin, auditlog, auth, authtoken, contenttypes, django_celery_results, documents, guardian, paperless, paperless_mail, sessions, socialaccount
Running migrations:
  No migrations to apply.
Running Django checks
System check identified no issues (0 silenced).
Search index out of date. Updating...
Paperless-ngx docker container starting...
Creating directory scratch directory /tmp/paperless
Adjusting permissions of paperless files. This may take a while.
Waiting for Redis...
Connected to Redis broker.
Apply database migrations...
Operations to perform:
  Apply all migrations: account, admin, auditlog, auth, authtoken, contenttypes, django_celery_results, documents, guardian, paperless, paperless_mail, sessions, socialaccount
Running migrations:
  No migrations to apply.
Running Django checks
System check identified no issues (0 silenced).
Search index out of date. Updating...

Steps to reproduce

Update from 2.7.2 to 2.8.2 with embedded SQLite

Webserver logs

see above

Browser logs

No response

Paperless-ngx version

2.8.2

Host OS

Synology Docker

Installation method

Docker - official image

Browser

No response

Configuration changes

No response

Other

No response

Please confirm the following

• I believe this issue is a bug that affects all users of Paperless-ngx, not something specific to my installation.
• I have already searched for relevant existing issues and discussions before opening this report.
• I have updated the title field above with a concise description.

[shamoon]

Since the docker Image isn’t broken widely I think this is more like the related to your installation So I’ve converted this to a discussion.

There’s not all that much to go off here but perhaps the container is running out of resources, e.g. memory you could try creating the search index manually

[stumpylog]

https://docs.paperless-ngx.com/administration/#index

[Kunibert-007]

That doesn't work because of the restart every few minutes...

[stumpylog]

You can also remove the index directory, as it will be created again if needed

[Kunibert-007]

You mean I should delete this directory and this file?

[Kunibert-007]

I deleted both but the same effect. The container starts, runs for about 3 Minutes and then it restart.

Paperless-ngx docker container starting...
Creating directory /usr/src/paperless/data/index
Creating directory scratch directory /tmp/paperless
Adjusting permissions of paperless files. This may take a while.
Waiting for Redis...
Connected to Redis broker.
Apply database migrations...
Operations to perform:
Apply all migrations: account, admin, auditlog, auth, authtoken, contenttypes, django_celery_results, documents, guardian, paperless, paperless_mail, sessions, socialaccount
Running migrations:
No migrations to apply.
Running Django checks
System check identified no issues (0 silenced).
Search index out of date. Updating...

Same with version 2.8.1.. I came from 2.7.2., there are no problems. Before updating I used the sanitizer and the reorg of the index for a clean base under 2.7.2..

[sw-carlos-cristobal]

I was able to resolve my issue.

Somehow permissions in the container were incorrect and the permissions for the .index_version had the script unable up update the file. Could the redirection be swallowing the error message?

paperless-ngx/docker/docker-prepare.sh

Line 89 in d4d0604

echo ${index_version} | tee "${index_version_file}" >/dev/null

I reset all permissions from within my docker shell to change the owner to paperless. Then I ran the reindexing command (making sure I was acting as paperless user) and it succeeded.

paperless-ngx/docker/docker-prepare.sh

Line 88 in d4d0604

python3 manage.py document_index reindex --no-progress-bar

Upon restarting of the docker container it started successfully.

In order to give myself time to do all of the above I just commented out the search_index check from here:

paperless-ngx/docker/docker-prepare.sh

Line 112 in d4d0604

search_index

I imagine I could have probably just fixed permissions and restart the container, but I figured I'd document all the steps I took.

[Kunibert-007]

That`s to much for my basic knowhow, can't fix it like you do.
And I wonder the problem is only in Version 2.8.x, not in 2.7.2.

[jtrumpio]

I too am having this issue. My install's been great for the last year, then after updating today, this. Will try above recommendations and report back.

[Kunibert-007]

Same problem with 2.8.3 but 2.7.2 works perfect.
Deleted the index-directory and it was automatic recreated but same effect - after 2-3 minutes with reindex it fails and the container restarts.

I gave the rights on the folder to everyone:

[Kunibert-007]

I see a difference between version 2.7.2 and 2.8.x.
In 2.8.x there is "Search index out of date" and the check starts (and fails):

In 2.7.2 there is no check (so nothing fails):

[jtrumpio]

I let it sit overight in the boot loop and it was fine this morning when I woke up.

[Kunibert-007]

Will test it, too. I let it in the loop now :-)

[Kunibert-007]

140 restarts and 5 hours later - it has finished! And seems to work now 👍

[matt9595-matt]

I resolved this issue. For me, it was my Postgres parameter 'idle_in_transaction_session_timeout' was set to 5min. Once I set the parameter to zero it worked just fine.

[dazilla3]

All your tricks are dont fix the problems in my case. I havent any idea what to do now.
The idle_in_transaction_session_timeout Parameter wasnt working.
The loop over night didnt helped to.

I think my last working paperless-ngx was at the 5. may because that was the last watchtower updated which pulled a new image because i was working with latest tag. When i try now to work with 2.8.1 or 2.8.2 its the same result.

[braberli]

Same for me: Latest docker image(> 2.8.x) do not work with exactly the same issue from the original post.
Using a 2.7.2 image with the same config works as expected.

Before downgrading I checked all file and directory permissions. For testing purposes I even "chmodded" them to 777.

I didn't change anything around paperless in the last two months.