[ 
https://issues.apache.org/jira/browse/IGNITE-22877?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Roman Puchkovskiy updated IGNITE-22877:
---------------------------------------
    Description: 
It should be possible to restart an IgniteServer. Such a restart should stop 
the IgniteImpl instance stored in the IgniteServer, then start another 
IgniteImpl and replace the old instance with the new one after the start 
completes.

The method should return a future, like this:

CompletableFuture<Void> restart();

Ignite instance returned to the user from IgniteServer via api() method should 
be a wrapper delegating to a volatile reference to IgniteImpl (which will be 
swapped when restarting). All sub-objects (to any depth) returned by this 
Ignite wrapper (like IgniteTables) should in turn be wrappers around the same 
volatile reference to IgniteImpl. This will allow the instances obtained by the 
user to remain functional after such an in-process restart.

> In-process IgniteServer restart
> -------------------------------
>
>                 Key: IGNITE-22877
>                 URL: https://issues.apache.org/jira/browse/IGNITE-22877
>             Project: Ignite
>          Issue Type: Improvement
>            Reporter: Roman Puchkovskiy
>            Priority: Major
>              Labels: ignite-3
>
> It should be possible to restart an IgniteServer. Such a restart should stop 
> the IgniteImpl instance stored in the IgniteServer, then start another 
> IgniteImpl and replace the old instance with the new one after the start 
> completes.
> The method should return a future, like this:
> CompletableFuture<Void> restart();
> Ignite instance returned to the user from IgniteServer via api() method 
> should be a wrapper delegating to a volatile reference to IgniteImpl (which 
> will be swapped when restarting). All sub-objects (to any depth) returned by 
> this Ignite wrapper (like IgniteTables) should in turn be wrappers around the 
> same volatile reference to IgniteImpl. This will allow the instances obtained 
> by the user to remain functional after such an in-process restart.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

Reply via email to