> On April 6, 2017, 8:07 p.m., Hao Hao wrote:
> > sentry-provider/sentry-provider-db/src/main/java/org/apache/sentry/service/thrift/SentryService.java
> > Lines 310 (patched)
> > <https://reviews.apache.org/r/58221/diff/1/?file=1685541#file1685541line325>
> >
> >     what happens if the submitted task has not been shut down and someone 
> > calls startHMSFollower? Is it safe to allow so?

I think so. While old HMSFollwer executor is shutting down (including the old 
HMSFollwer), the start will cause the new HMSFollwer executor and new 
HMSFollwer to be created and the task scheduled. They are seperate instances. 

This scenario is reasonable. For example, user may stop and then start before 
shutting down finishes. We either allow start before shutting down finishes or 
make start do nothing before shutting down finishes.

In stop(), "status = Status.NOT_STARTED;" is called before stopping HMSFollwer, 
which allows start() to call runServer() and then start HMSFollwer. So the 
current code does allow this scenario.

We can test this case.


- Na


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/58221/#review171244
-----------------------------------------------------------


On April 6, 2017, 10:40 p.m., Na Li wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/58221/
> -----------------------------------------------------------
> 
> (Updated April 6, 2017, 10:40 p.m.)
> 
> 
> Review request for sentry, Alexander Kolbasov, Hao Hao, kalyan kumar 
> kalvagadda, and Sergio Pena.
> 
> 
> Repository: sentry
> 
> 
> Description
> -------
> 
> SENTRY-1649 move HMS follower to runServer
> 
> 
> Diffs
> -----
> 
>   
> sentry-provider/sentry-provider-db/src/main/java/org/apache/sentry/service/thrift/SentryService.java
>  132db635f639cef91ce675d34d717e6125a0a4d1 
> 
> 
> Diff: https://reviews.apache.org/r/58221/diff/2/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Na Li
> 
>

Reply via email to