[
https://issues.apache.org/jira/browse/RATIS-982?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
runzhiwang updated RATIS-982:
-----------------------------
Description:
This happens in test, but it maybe also happen in production.
For example, leader is s3 and follower is s4.
1. kill s4, and restart s4.
{code:java}
2020-06-19T07:03:18.1000860Z 2020-06-19 07:03:18,095 [Thread-6194] INFO
ratis.MiniRaftCluster (MiniRaftCluster.java:killServer(458)) - killServer s4
2020-06-19T07:03:18.1001826Z 2020-06-19 07:03:18,095 [Thread-6194] INFO
ratis.MiniRaftCluster (MiniRaftCluster.java:newRaftServer(330)) -
newRaftServer: s4, group-5BD7E8A01610:[s3:0.0.0.0:43375, s4:0.0.0.0:33719,
s0:0.0.0.0:34867, s1:0.0.0.0:33783, s2:0.0.0.0:40473], format? false
{code}
2. s4 start and set configuration from storage at
[setRaftConf(raftConf.getLogEntryIndex(), raftConf)
|https://github.com/apache/incubator-ratis/blob/master/ratis-server/src/main/java/org/apache/ratis/server/impl/ServerState.java#L170]
and s4 will change to RUNNING at
[lifeCycle.transition(RUNNING)|https://github.com/apache/incubator-ratis/blob/master/ratis-server/src/main/java/org/apache/ratis/server/impl/RaftServerImpl.java#L213]
2020-06-19T07:03:18.1345896Z 2020-06-19 07:03:18,127 [pool-16-thread-1] INFO
impl.RaftServerImpl (ServerState.java:setRaftConf(356)) -
s4@group-5BD7E8A01610: set configuration 0: [s3:0.0.0.0:43375,
s4:0.0.0.0:33719, s0:0.0.0.0:34867, s1:0.0.0.0:33783, s2:0.0.0.0:40473],
old=null at 0
3. s3 send append entry request to s4, and s4 change to RUNNING at
[lifeCycle.compareAndTransition(STARTING,
RUNNING)|https://github.com/apache/incubator-ratis/blob/master/ratis-server/src/main/java/org/apache/ratis/server/impl/RaftServerImpl.java#L1003]
4. If change to RUNNING in step3 happens before step2, then step2 will throw
exception.
was:
This happens in test, but it maybe also happen in production.
For example, leader is s3 and follower is s4.
1. kill s4, and restart s4.
{code:java}
2020-06-19T07:03:18.1000860Z 2020-06-19 07:03:18,095 [Thread-6194] INFO
ratis.MiniRaftCluster (MiniRaftCluster.java:killServer(458)) - killServer s4
2020-06-19T07:03:18.1001826Z 2020-06-19 07:03:18,095 [Thread-6194] INFO
ratis.MiniRaftCluster (MiniRaftCluster.java:newRaftServer(330)) -
newRaftServer: s4, group-5BD7E8A01610:[s3:0.0.0.0:43375, s4:0.0.0.0:33719,
s0:0.0.0.0:34867, s1:0.0.0.0:33783, s2:0.0.0.0:40473], format? false
{code}
2. s4 set configuration from storage at
setRaftConf(raftConf.getLogEntryIndex(), raftConf)
2020-06-19T07:03:18.1345896Z 2020-06-19 07:03:18,127 [pool-16-thread-1] INFO
impl.RaftServerImpl (ServerState.java:setRaftConf(356)) -
s4@group-5BD7E8A01610: set configuration 0: [s3:0.0.0.0:43375,
s4:0.0.0.0:33719, s0:0.0.0.0:34867, s1:0.0.0.0:33783, s2:0.0.0.0:40473],
old=null at 0
> RaftServerImpl failed to change from RUNNING to RUNNING
> -------------------------------------------------------
>
> Key: RATIS-982
> URL: https://issues.apache.org/jira/browse/RATIS-982
> Project: Ratis
> Issue Type: Bug
> Reporter: runzhiwang
> Assignee: runzhiwang
> Priority: Major
>
> This happens in test, but it maybe also happen in production.
> For example, leader is s3 and follower is s4.
> 1. kill s4, and restart s4.
> {code:java}
> 2020-06-19T07:03:18.1000860Z 2020-06-19 07:03:18,095 [Thread-6194] INFO
> ratis.MiniRaftCluster (MiniRaftCluster.java:killServer(458)) - killServer s4
> 2020-06-19T07:03:18.1001826Z 2020-06-19 07:03:18,095 [Thread-6194] INFO
> ratis.MiniRaftCluster (MiniRaftCluster.java:newRaftServer(330)) -
> newRaftServer: s4, group-5BD7E8A01610:[s3:0.0.0.0:43375, s4:0.0.0.0:33719,
> s0:0.0.0.0:34867, s1:0.0.0.0:33783, s2:0.0.0.0:40473], format? false
> {code}
> 2. s4 start and set configuration from storage at
> [setRaftConf(raftConf.getLogEntryIndex(), raftConf)
> |https://github.com/apache/incubator-ratis/blob/master/ratis-server/src/main/java/org/apache/ratis/server/impl/ServerState.java#L170]
> and s4 will change to RUNNING at
> [lifeCycle.transition(RUNNING)|https://github.com/apache/incubator-ratis/blob/master/ratis-server/src/main/java/org/apache/ratis/server/impl/RaftServerImpl.java#L213]
> 2020-06-19T07:03:18.1345896Z 2020-06-19 07:03:18,127 [pool-16-thread-1] INFO
> impl.RaftServerImpl (ServerState.java:setRaftConf(356)) -
> s4@group-5BD7E8A01610: set configuration 0: [s3:0.0.0.0:43375,
> s4:0.0.0.0:33719, s0:0.0.0.0:34867, s1:0.0.0.0:33783, s2:0.0.0.0:40473],
> old=null at 0
> 3. s3 send append entry request to s4, and s4 change to RUNNING at
> [lifeCycle.compareAndTransition(STARTING,
> RUNNING)|https://github.com/apache/incubator-ratis/blob/master/ratis-server/src/main/java/org/apache/ratis/server/impl/RaftServerImpl.java#L1003]
> 4. If change to RUNNING in step3 happens before step2, then step2 will throw
> exception.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
