You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Version: What redis-py and what redis version is the issue happening on?
redis-py 5.2.0 with redis 7.2.4
Platform: What platform / version?
Python 3.10.12 on Ubuntu 22.04
Description:
Hey everyone,
I have an application that uses a Redis PubSub created with a connection from a connection pool (with a health_check_interval defined). Then, at a high rate (multiple times per second), it verifies if there are messages to be read (by calling the get_message() method, with a timeout), making that connection constantly in use. However, there are many times when there are no messages to be read, and so the can_read in the try_read function returns False. What I am observing is that there are health check commands emitted to Redis for this connection because the next_health_check is never updated, since there are no messages to be read.
Is this the expected behavior? I think it is strange to be doing health checks with this connection because it is constantly in use, despite there being no messages to be read.
Best regards,
Jorge
The text was updated successfully, but these errors were encountered:
jmspereira
changed the title
Health Checks being Performed when not Needed
Health checks being performed when not needed
Nov 8, 2024
Version: What redis-py and what redis version is the issue happening on?
redis-py 5.2.0 with redis 7.2.4
Platform: What platform / version?
Python 3.10.12 on Ubuntu 22.04
Description:
Hey everyone,
I have an application that uses a Redis PubSub created with a connection from a connection pool (with a health_check_interval defined). Then, at a high rate (multiple times per second), it verifies if there are messages to be read (by calling the get_message() method, with a timeout), making that connection constantly in use. However, there are many times when there are no messages to be read, and so the can_read in the try_read function returns False. What I am observing is that there are health check commands emitted to Redis for this connection because the next_health_check is never updated, since there are no messages to be read.
Is this the expected behavior? I think it is strange to be doing health checks with this connection because it is constantly in use, despite there being no messages to be read.
Best regards,
Jorge
The text was updated successfully, but these errors were encountered: