Skip to content

Commit 01e7b31

Browse files
committed
PB-2395: explain that bad retry may lead to dropped requests
1 parent 590782a commit 01e7b31

1 file changed

Lines changed: 2 additions & 1 deletion

File tree

get-started/retry.md

Lines changed: 2 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -9,7 +9,8 @@ Any request you make has the possibility to fail for a variety of reasons. It
99
is your responsibility to decide what to do when this happens. When encountering
1010
transient errors, you will typically want to retry the request. This requires
1111
some care to do correctly as incorrect retry strategies may cause many requests
12-
to be (re)sent in a short amount of time. To mitigate this
12+
to be (re)sent in a short amount of time, which may eventually cause them to be
13+
throttled or otherwise permanently dropped. To mitigate this
1314
[thundering herd problem](https://en.wikipedia.org/wiki/Thundering_herd_problem),
1415
we recommend you implement retries with jittered truncated exponential backoff.
1516

0 commit comments

Comments
 (0)