Search function not indexing any new tickets since 3.6.1 upgraded

[zammadusr]

Infos:

• Used Zammad version: 3.6.1
• Used Zammad installation source: yum -y install zammad
• Operating system: CentOS 8
• Browser + version: Chrome, Firefox, Brave

Expected behavior:

Searching for any tickets entered post 3.6.1 are not found. Old tickets are found.

Likewise, reports only show tickets up to the point Zammad was upgraded to 3.6.1

Actual behavior:

Run a report and no results for tickets entered after the 3.6.1 upgrade was run

Steps to reproduce the behavior:

Run a report and no results for tickets entered after the 3.6.1 upgrade was run

I assume I’ve managed to break the search system somehow with the upgrade.

The elasticsearch service is running. Would anyone know what I can check?

When I originally installed Zammad for the first time last year I ran these commands related to the search system:
zammad run rails r “Setting.set(‘es_url’, ‘http://localhost:9200’)”
zammad run rake searchindex:rebuild
zammad run rails r “Setting.set(‘es_index’, Socket.gethostname.downcase + ‘_zammad’)”
zammad run rails r “Setting.set(‘es_attachment_ignore’, [ ‘.png’, ‘.jpg’, ‘.jpeg’, ‘.mpeg’, ‘.mpg’, ‘.mov’, ‘.bin’, ‘.exe’, ‘.box’, ‘.mbox’ ] )”
zammad run rails r “Setting.set(‘es_attachment_max_size_in_mb’, 50)”

Will I re-run these commands?

[zammadusr]

I ran a rebuild:

zammad run rake searchindex:rebuild

and had an error:

drop indexes…done
delete pipeline (pipeline)… done
create indexes…done
create pipeline (pipeline)… rake aborted!
Unable to process put request to elasticsearch URL ‘http://localhost:9200/_ingest/pipeline/zammad561189949055’. Check the response and payload for detailed information:

Response:
#<UserAgent::Result:0x00005632cf2c1020 @success=false, @body="{“error”:{“root_cause”:[{“type”:“parse_exception”,“reason”:“No processor type exists with name [attachment]”,“processor_type”:“foreach”,“suppressed”:[{“type”:“parse_exception”,“reason”:“No processor type exists with name [attachment]”,“processor_type”:“foreach”}]}],“type”:“parse_exception”,“reason”:“No processor type exists with name [attachment]”,“processor_type”:“foreach”,“suppressed”:[{“type”:“parse_exception”,“reason”:“No processor type exists with name [attachment]”,“processor_type”:“foreach”}]},“status”:400}", @data=nil, @code=“400”, @content_type=nil, @error=“Client Error: #<Net::HTTPBadRequest 400 Bad Request readbody=true>!”>

Payload:
{“description”:“Extract zammad-attachment information from arrays”,“processors”:[{“foreach”:{“field”:“article”,“processor”:{“foreach”:{“field”:"_ingest._value.attachment",“processor”:{“attachment”:{“target_field”:"_ingest._value",“field”:"_ingest._value._content",“ignore_failure”:true,“ignore_missing”:true}},“ignore_failure”:true,“ignore_missing”:true}},“ignore_failure”:true,“ignore_missing”:true}},{“foreach”:{“field”:“attachment”,“processor”:{“attachment”:{“target_field”:"_ingest._value",“field”:"_ingest._value._content",“ignore_failure”:true,“ignore_missing”:true}},"

[zammadusr]

The fix in case anyone else comes across this issue:

If you run “yum update” on a CentOS server to update everything, for some reason the elasticsearch ingest-attachment plugin will be lost.

The solution:

1. Re-install the ingest-attachment plugin:
   sudo /usr/share/elasticsearch/bin/elasticsearch-plugin install ingest-attachment

2. restart the elasticsearch service

3. rebuild the index:
   zammad run rake searchindex:rebuild

[MrGeneration]

Just a small side note:
If you update elasticsearch only (or there’s just no update available for Zammad), the attachment plugin will get “invalid”. It needs manual updating.

Usually Zammad does this for you if you update Zammad.