What to do if the fulltextsearch index does not work?

[szaimen]

When enabling Fulltextsearch in the AIO interface, it will try to do the fulltextsearch index on the next startup of the Nextcloud container. If that fails, it will try to re-do the index every time the container starts. If that should be the case for you, you can fix the problem by doing the following:

1. report the error log to the upstream project: https://github.com/nextcloud/fulltextsearch/issues (make sure that there is not an open issue about this already)
2. Create the marker file manually, so that it does not repeatedly tries to re-do the index:

# Run via CLI:
sudo docker exec -it nextcloud-aio-nextcloud bash

# Now inside the container
touch "$NEXTCLOUD_DATA_DIR/fts-index.done"
exit

This should skip the attempt to do the indexing in the future.

[RudolfAchter]

In my case indexing seems to restart in an infinite loop. When the container starts it tries to index all files until hitting the memory limit it seems

I try the workaround with creating the fts-index.done file and see if index ever gets finished

[gbertolottiKS]

If we want to start again the index generation, removing the "fts-index.done" file is enough?
Thanks

[szaimen]

okay, then you could also run the command manually. However not sure if this works since I didnt test it.

[gbertolottiKS]

Sorry, to what command do you refer?
What are the steps you think we should do to restart manually the generation?
Thanks for your patience

[szaimen]

all-in-one/Containers/nextcloud/entrypoint.sh

Line 767 in 75fdb28

if php /var/www/html/occ fulltextsearch:test && php /var/www/html/occ fulltextsearch:index "{\"errors\": \"reset\"}" --no-readline; then

[gbertolottiKS]

Thanks; a last question: before launching this command, the containers should be stopped or active?

[szaimen]

Active