-
Notifications
You must be signed in to change notification settings - Fork 6.9k
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
Add profile event in hedged requests and change timeout from sec to ms. #21886
Conversation
… requests and better process receiving packets
@@ -4,8 +4,8 @@ | |||
<default> | |||
<load_balancing>in_order</load_balancing> | |||
<max_parallel_replicas>2</max_parallel_replicas> | |||
<hedged_connection_timeout>100</hedged_connection_timeout> | |||
<receive_data_timeout>2</receive_data_timeout> | |||
<hedged_connection_timeout_ms>100</hedged_connection_timeout_ms> |
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.
This one should be increased, right?
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.
Connection timeout was already in ms.
Many integration tests from test_distributed_respect_user_timeouts are flaky under tsan, can this be related? |
I see the same problem on master, I think it's not related to this PR. |
OK, let's merge then. |
…nge timeout from sec to ms.
Backport #21886 to 21.3: Add profile event in hedged requests and change timeout from sec to ms.
I hereby agree to the terms of the CLA available at: https://yandex.ru/legal/cla/?lang=en
Changelog category (leave one):
Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):
Add profile event HedgedRequestsChangeReplica, change read data timeout from sec to ms.
Detailed description / Documentation draft:
HedgedRequestsChangeReplica event counts the total number of times when a change replica timeout has been alarmed. Change
read_data_timeout
timeout from sec to ms, add_ms
ending to the timeouts. Add better processing the case whenmax_parallel_replicas
> 1 but there is no way to sample data in this query.