Skip to content

Implement exponential backoff for the retry strategy #620

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

Merged
merged 5 commits into from
Aug 26, 2021
Merged

Implement exponential backoff for the retry strategy #620

Changes from 1 commit
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
241b40a
Implement exponential backoff for the retry strategy
brenuart Aug 24, 2021
09339fa
Output status messages on console during test to help troubleshooting
brenuart Aug 24, 2021
fde1889
Update README with some words about the "new" retry strategy
brenuart Aug 24, 2021
8aeef5b
Limit time waiting for the "retry lock" + stop retrying if the thread…
brenuart Aug 26, 2021
bb036c7
Merge branch 'main' into gh619-async-retry-perf
brenuart Aug 26, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Update README with some words about the "new" retry strategy
  • Loading branch information
@brenuart
brenuart committed Aug 24, 2021
commit fde188978a3dfe8bba2dc59940587bee1b5881d5
3 changes: 2 additions & 1 deletion README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -663,7 +663,8 @@ The behaviour of the appender when the RingBuffer is controlled by the `appendTi
| `> 0` | retry during the specified amount of time |


Logging threads waiting for space in the RingBuffer wake up periodically at a frequency defined by `appendRetryFrequency` (default `5ms`). You may increase this frequency for faster reaction time at the expense of higher CPU usage.
Logging threads waiting for space in the RingBuffer wake up periodically at a frequency starting at `1ns` and increasing exponentially up to `appendRetryFrequency` (default `5ms`).
Only one thread is allowed to retry at a time. If a thread is already retrying, additional threads are waiting on a lock until the first is finished. This strategy should help to limit CPU consumption while providing good enough latency and throughput when the ring buffer is at (or close) to its maximal capacity.

When the appender drops an event, it emits a warning status message every `droppedWarnFrequency` consecutive dropped events. Another status message is emitted when the drop period is over and a first event is succesfully enqueued reporting the total number of events that were dropped.

Expand Down