Replies: 4 comments
-
query.client.timeout# Default value: 5m Configures how long the cluster runs without contact from the client application, such as the CLI, before it abandons and cancels its work. |
Beta Was this translation helpful? Give feedback.
-
In addition to what is said above I'll say the Trino protocol doesn't need persistent connections so there's no "session" to speak of in that sense. When a query is not executing there's no active connection to the cluster. Similarly if you are using a client and not actively fetching more rows there's no ACTIVE connection (other than for connection reuse purposes for example). |
Beta Was this translation helpful? Give feedback.
-
As trino does not need persistent connections (no session), how does authentication works, it repeats for each query? |
Beta Was this translation helpful? Give feedback.
-
The HTTP connection is usually re-used so not need for repeated authentications (in most client implementations). If you use |
Beta Was this translation helpful? Give feedback.
-
Hello,
Is there any setting related to session idle timeout in trino?
If the session is maintained without disconnecting during data fetch, how long will it be maintained?
I'm looking for a setting that is equivalent to
hive.server2.idle.session.timeout
of Hive.Beta Was this translation helpful? Give feedback.
All reactions