Skip to content
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.

Sign up for GitHub

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

Jump to bottom

[Feature Request] Unified gRPC retry logic #27

Open
bergundy opened this issue Mar 14, 2022 · 0 comments
Open

[Feature Request] Unified gRPC retry logic #27

bergundy opened this issue Mar 14, 2022 · 0 comments
Labels
enhancement New feature or request

Comments

@bergundy
Copy link
Member

Each SDK's gRPC retry logic is a bit different and should be unified.
@bergundy bergundy added the enhancement New feature or request label Mar 14, 2022
@bergundy bergundy mentioned this issue Aug 3, 2022
Closed
@Quinn-With-Two-Ns Quinn-With-Two-Ns mentioned this issue May 23, 2023
Closed
@lorensr lorensr mentioned this issue Aug 21, 2023
Closed
@bergundy bergundy mentioned this issue Sep 11, 2023
Closed
5 tasks
This was referenced Feb 27, 2024
Merged
Merged
chronos-tachyon added a commit to temporalio/sdk-core that referenced this issue Feb 27, 2024
@chronos-tachyon
Slow down retries so we keep retrying for more than 10 seconds. (#696)
89816bf 
= What was changed

* Increase backoff factor to 1.7 (from 1.5) for non-poll RPCs

= Why?

The current backoff factor causes clients using the Core SDK to exhaust their retries within 6.011 to 8.966 seconds, depending on random jitter. We want to ensure that all clients keep retrying for at least 10 seconds. With the new backoff factor, retries will not be exhausted until 11.878 to 17.767 seconds after the initial failure.

This advances progress on temporalio/features#27 (aka SDK-118 in internal Jira).
chronos-tachyon added a commit to temporalio/sdk-typescript that referenced this issue Feb 27, 2024
@chronos-tachyon
Bring gRPC retry options in line with other SDKs. (#1368)
b38c70c 
## What was changed

* Raise initial interval to 100ms (from 20ms)
* Drop max interval to 5s (from 10s)
* Drop backoff factor to 1.7 (from 2.0)
* Raise jitter factor to 0.2 (from 0.1)

## Why?

The current retry options for the TypeScript SDK are a little wonky, as the existing initial interval is *by far* the most aggressive among all Temporal SDKs.  This brings the default parameters in line with the changes already made to the Java SDK and in flight for the Core SDK.

This also fixes a wart: the TypeScript SDK can exhaust all retries in 9.200 seconds, if the random jitters for each retry align in just the right way, and we want all SDKs to keep retrying for at least 10 seconds after the initial failure.

This advances progress on temporalio/features#27 (aka SDK-118 in internal Jira).
@sync-by-unito sync-by-unito bot closed this as completed Jul 12, 2024
@josh-berry josh-berry reopened this Jul 13, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@bergundy @josh-berry