Concurrent SG queries causes QueryTimeoutException

Hi all,
I manage a Couchbase cluster infrastructure with 2 nodes Data/Query/Index.

Recently, we’re observing an issue related to tons of queries sent by a Community SG in a specific timewindow (usually every day from 7AM to 9AM and from 7PM to 9PM).

The application uses the SG to enrich data and to set permissions on the documents with Channels and Roles.
The concurrent queries are the following:

SELECT meta(<bucket>).xattrs._sync.access.role:User as value FROM <bucket> USE INDEX (sg_access_x1) WHERE any op in object_pairs(meta(<bucket>).xattrs._sync.access) satisfies op.name = $userName end;

After 1m15s the queries end in timeout, causing the application BE disruption, since it doesn’t get a result.

We also tried to increase the SG query timeout, but we observed even more queries stressing out the entire cluster. Hence we set it again to the default 75s value.

Any idea on why is this happening?
Is there any related known issue or is just something misconfigured at the application level that is triggering too much queries?

3 posts - 2 participants

Read full topic