Details
-
Improvement
-
Status: Resolved
-
Blocker
-
Resolution: Fixed
-
None
Description
Currently any ozone client request can spend a huge amount of time in retries and ozone client can retry its requests very aggressively. The waiting time can thus be very high before a client request fails. Further aggressive retries by ratis client used by ozone can bog down a ratis pipeline leader. The Jira aims to make changes to the current retry behavior in Ozone client.
Attachments
Attachments
Issue Links
- blocks
-
HDDS-3994 Write object when met exception can be slower than before
- Resolved
- contains
-
HDDS-3136 retry timeout is large while writing key
- Resolved
-
HDDS-2702 Client failed to recover from ratis AlreadyClosedException exception
- Resolved
-
HDDS-2306 Fix TestWatchForCommit failure
- Resolved
-
HDDS-2963 Use RequestDependentRetry Policy along with ExceptionDependentRetry Policy in OzoneClient
- Resolved
-
HDDS-3163 write Key is hung when write delay is injected in datanode dir
- Resolved
-
HDDS-3268 CommitWatcher#watchForCommit does not timeout
- Resolved
-
HDDS-2917 Add a different retry policy for watch requests
- Resolved
- depends upon
-
RATIS-835 Include exception based attempt count in raft client request
- Resolved
-
RATIS-851 Raft Client should not change leader on ResourceUnavailableException
- Resolved
-
RATIS-853 Unordered Client request should not sleep when NotLeaderException provides leader information
- Resolved
-
RATIS-876 Introduce max timeout in RequestTypeDependentRetryPolicy
- Resolved
-
RATIS-882 Implement ExponentialBackoffRetry
- Resolved
- duplicates
-
HDDS-3228 Enforce Exponential backoff retry policy for raft client in cases of system overload
- Resolved
- links to
1.
|
retry timeout is large while writing key | Resolved | Unassigned |