-
Notifications
You must be signed in to change notification settings - Fork 2.6k
Hitless handshake #3735
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: ps_add_fail_over_events_handling
Are you sure you want to change the base?
Hitless handshake #3735
Conversation
… tests for maintenance_events.py file
…tion pool - this should be a separate PR
… Refactored the maintenance events tests not to be multithreaded - we don't need it for those tests.
…ot processed in in Moving state. Tests are updated
…ply them during connect
Signed-off-by: Elena Kolevska <[email protected]> Cleanup Signed-off-by: Elena Kolevska <[email protected]>
Signed-off-by: Elena Kolevska <[email protected]>
…isting ones more generic
Signed-off-by: Elena Kolevska <[email protected]>
Signed-off-by: Elena Kolevska <[email protected]>
Signed-off-by: Elena Kolevska <[email protected]>
Signed-off-by: Elena Kolevska <[email protected]>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull Request Overview
This PR implements handshake functionality for hitless upgrades and updates the server maintenance message format to include sequence IDs for better event tracking.
- Adds endpoint type detection and CLIENT MAINT_NOTIFICATIONS command for proper handshake setup
- Updates maintenance event message parsing to handle new format with sequence IDs
- Changes default behavior to enable maintenance events by default
Reviewed Changes
Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.
Show a summary per file
File | Description |
---|---|
redis/maintenance_events.py |
Adds EndpointType class and endpoint detection logic, enables maintenance events by default |
redis/connection.py |
Implements handshake via CLIENT MAINT_NOTIFICATIONS command during connection setup |
redis/_parsers/base.py |
Updates message parsing to handle new format with sequence IDs and adds error handling |
tests/test_maintenance_events.py |
Adds comprehensive tests for endpoint type detection and configuration validation |
tests/test_maintenance_events_handling.py |
Updates mock responses to match new message format with sequence IDs |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
return False | ||
|
||
|
||
def _get_resolved_ip_from_connection( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it would be better to expose this function on the connection object.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Done.
…e better retry_on_error handling on connection initialization.
8d7cc00
to
10ded34
Compare
* Typos in commands.py * Correct URL in vectorset commands.py
…onPolicyInterface(redis#3732)
…#3512) * Fix async safety when Redis client is used as an async context manager When the async Redis client is used as an async context manager and called from different corotuines, one coroutine can exit, shutting down the client's connection pool, while another coroutine is attempting to use a connection. This results in a connection error, such as: redis.exceptions.ConnectionError: Connection closed by server. Additional locking in `ConnectionPool` resolves the problem but introduces extreme latency due to the locking. Instead, this PR implements a shielded counter that increments as callers enter the async context manager and decrements when they exit. The client then closes its connection pool only after all active contexts exit. Performance is on par with use of the client without a context manager.
Signed-off-by: Elena Kolevska <[email protected]>
Pull Request check-list
Please make sure to review and check all of these items:
NOTE: these things are not required to open a PR and can be done
afterwards / while the PR is open.
Description of change