Write queue increasing

Hi,

We have upgraded our cluster to the latest Elastic 8 version but have hit a strange bug twice in two weeks; The write queue is filling up and write tasks are very slow to complete.

This monitoring screenshots clearly shows that the problem started on April 8th.

Each indexing task only contains few documents. I can't understand were the bottleneck is and why the tasks are so slow to process (cluster is indexing at less than 10 doc/s).

$ docker-compose exec app curl 'http://es:9200/_cat/thread_pool/resh?v=true&h=node_name,active,queue,rejected,completed&s=node_name'
node_name                 active queue rejected completed
global.admin       0     0        0         0
global.data01      0     0        0  66940588
global.data02      0     0        0  66557132
global.data03      0     0        0  66746973

$ docker-compose exec app curl 'http://es:9200/_cat/thread_pool/write?v=true&h=node_name,active,queue,rejected,completed&s=node_name'
node_name                 active queue rejected completed
global.admin       0     0        0   1159665
global.data01     16  2444        0    693213
global.data02      0     0        0    674059
global.data03      3     0        0    683521

Thanks

2 Likes

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.