1) Pool timeout
If so, why I see error message with 5sec timeout, but database.yml has 10sec?
root@9e15d658a00c:/opt/zammad# cat config/database.yml
production:
adapter: postgresql
database: zammad_production
pool: 50
timeout: 10000
→
E, [2020-01-09T02:23:47.111488 #1-70186290246240] ERROR -- : thread_client 46949339110080 exited with error #<ActiveRecord::ConnectionTimeoutError: could not obtain a connection from the pool within 5.000 seconds (waited 8.287 seconds); all pooled connections were in use>
?
2) Overviews count
Thanks for the advice. We reduced overview, will see the load at Mon
3) Search Index rebuild
Yesterday we launched index rebuild rake searchindex:rebuild and got that error after 2-3 hours of rebuilding
/opt/zammad/app/models/concerns/has_search_index_backend.rb:145:in `rescue in block in search_index_reload'
/opt/zammad/app/models/concerns/has_search_index_backend.rb:139:in `block in search_index_reload'
/opt/zammad/app/models/concerns/has_search_index_backend.rb:134:in `each'
/opt/zammad/app/models/concerns/has_search_index_backend.rb:134:in `search_index_reload'
/opt/zammad/lib/tasks/search_index_es.rake:143:in `block (3 levels) in <main>'
/opt/zammad/lib/tasks/search_index_es.rake:139:in `each'
/opt/zammad/lib/tasks/search_index_es.rake:139:in `block (2 levels) in <main>'
/opt/zammad/lib/tasks/search_index_es.rake:153:in `block (2 levels) in <main>'
/usr/local/bundle/gems/rake-12.3.2/exe/rake:27:in `<top (required)>'
Caused by:
Unable to process POST request to elasticsearch URL 'http://zammad-elasticsearch:9200/zammad_production/Ticket/40846?pipeline=zammad813516161384'. Check the response and payload for detailed information:
Response:
#<UserAgent::Result:0x000055c63c055e08 @success=false, @body=nil, @data=nil, @code=0, @content_type=nil, @error="#<SocketError: Failed to open TCP connection to zammad-elasticsearch:9200 (getaddrinfo: Temporary failure in name resolution)>">
But I am completely sure that we don’t have any name resolution issues. To confirm that I run 2 tests. First is
while true; do echo $(date) - $(getent ahosts zammad-elasticsearch) >> 1.log; sleep 1; done;
and second is
while true; do nc zammad-elasticsearch 9200 -zv >> 1.log 2>&1; sleep 1; done;
Both of them were working for more than 12 hours without any issue.
4) asset for ‘customer’ error
Also, I could say that problem #4 from my initial message happens too often I guess. 5-10 times in a minute. I’m talking about
>E, [2020-01-07T03:46:44.883545 #1-47276871602100] ERROR -- : Unable to get asset for 'customer': #<NameError: uninitialized constant Customer>
and related issue topic ERROR -- : Unable to get asset for 'customer': #<NameError: uninitialized constant Customer> - #5 by chesty
How can I try to trace the problem? I guess that we have an error client or user somehow in the DB, maybe with an undefined mandatory field or smt. Just my assumption
5) additional worker
And we still have to launch additional task jobs:work to get jobs done, but this issue we had on 2.8 too. Not related with the upgrade process.