[
https://issues.apache.org/jira/browse/HBASE-17800?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Enis Soztutar updated HBASE-17800:
----------------------------------
Attachment: hbase-17800-v5.patch
[~xiaobingo] I was able to find out and fix a couple of problems that you were
debugging, and also did a couple of other changes that were needed.
On top of v3 patch
- Moved to a retry_executor which is needed due to how retry timer and
eventbase works in folly (see comments in patch).
- Some gymnastics with scheduling the tasks in the retry executors
- Moved to using exception_wrapper inside {{Future}}'s and other places to
carry the exception pointer type information.
- Added a lot more tests, including positive and negative tests and timeout
test.
- Cleanup logging a bit
This version passes the tests, and should be ready to commit. Makes the whole
retrying really work.
> [C++] handle exceptions in client RPC
> -------------------------------------
>
> Key: HBASE-17800
> URL: https://issues.apache.org/jira/browse/HBASE-17800
> Project: HBase
> Issue Type: Sub-task
> Reporter: Xiaobing Zhou
> Assignee: Xiaobing Zhou
> Attachments: HBASE-17800-HBASE-14850.000.patch,
> HBASE-17800-HBASE-14850.001.patch, HBASE-17800-HBASE-14850.002.patch,
> HBASE-17800-HBASE-14850.003.patch, hbase-17800-v5.patch
>
>
> Exceptions are ignored in current client RPC. They should be handled properly
> to be consumed by RPC retry or propagated up to APIs.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
