apache / apache/pulsar-client-go
Error comment/behavior of default negative backoff policy
- Dominant language
- Go
- Stars
- 745
- Forks
- 389
- Avg merge
- 3d 20h
- Merged PRs (30d)
- 3
Description
#### Expected behavior
int the `pulsar/negative_backoff_policy.go` file, the Next function declare two variables
```go
minNackTimeMs := int64(1000 * 30) and comment "30 sec"
maxNackTimeMs := 1000 * 60 * 10 and comment "10 min"
```
But the Next function which is called in `pulsar/consumer_partition.go` `reconnectToBroker`(or maybe other places),
and we can see that the return value of used in `time.Sleep()` which means its unit should be nanosecond(same as time.Duration).
So if we want a "30 sec" or "10 min", than we should declare below:
```go
minNackTimeMs := int64(30 * 1000 * 1000 * 1000) // 30sec
maxNackTimeMs := int64(10 * 60 * 1000 * 1000 * 1000) // 10min
```
#### Actual behavior
Tell us what happens instead
#### Steps to reproduce
How can we reproduce the issue
#### System configuration
**Pulsar version**: x.y
Contributor guide
Research direction
Inspect pulsar/negative_backoff_policy.go and the Next function, then trace its callers in pulsar/consumer_partition.go, including reconnectToBroker. Verify the returned duration's unit before it reaches time.Sleep; done means the configured 30-second and 10-minute backoffs, comments, and observed behavior agree.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- go
- Domain
- backend, distributed-systems
- Issue type
- Bug
- Difficulty
- 2/5
- Estimated time
- 1-3 hours
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 42/100