Query regarding HttpClient retry mechanism #26
Comments
|
Looks like a bug. I will take a look at it. |
jsabin
pushed a commit
that referenced
this issue
Jul 30, 2015
… being incremented each time execute() was called. The intent was to increase the retry value once for each execution. So if retry is 3, it would be executed once and then retry 3 times.
|
Fixed in version #d59498fcb3bb7a527f3b644edd51fc089016efab. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
HttpClient has a retry mechanism.
Can I ask why the internal retry counter is incremented on every call to execute
https://github.com/kairosdb/kairosdb-client/blob/master/src/main/java/org/kairosdb/client/HttpClient.java#L85
After testing the client last night - I see that after a few thousand successful requests, when the server (on my laptop) went away (died), there were delays of more than 40 minutes between "exceptions"
when the server recovered some time later requests were successful again, but when it went away again, the retry mechanism was still causing delays of more than an hour.
Why the ++operator at line 85?
is it meant for some kind of exponential backoff?
what if it were reset to the initial value after a successful "1st attempt" submission instead of successful submissions altering the internal state every time?
happy to provide a patch for any of this.
The text was updated successfully, but these errors were encountered: