[
https://issues.apache.org/jira/browse/IGNITE-10117?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Alexei Scherbakov updated IGNITE-10117:
---------------------------------------
Description:
This is because
org.apache.ignite.internal.processors.cache.persistence.IgniteCacheDatabaseSharedManager#reserveHistoryForExchange
is called before
org.apache.ignite.internal.processors.cache.persistence.IgniteCacheDatabaseSharedManager#beforeExchange,
which restores correct partition state.
{noformat}
public void testHistory() throws Exception {
IgniteEx crd = startGrid(0);
startGrid(1);
crd.cluster().active(true);
awaitPartitionMapExchange();
int part = 0;
List<Integer> keys = loadDataToPartition(part, DEFAULT_CACHE_NAME, 100,
0, 1);
forceCheckpoint(); // Prevent IGNITE-10088
stopGrid(1);
awaitPartitionMapExchange();
List<Integer> keys1 = loadDataToPartition(part, DEFAULT_CACHE_NAME,
100, 100, 1);
stopAllGrids();
crd = startGrid(0);
startGrid(1);
crd.cluster().active(true);
awaitPartitionMapExchange(); // grid0 will not provide history.
}
{noformat}
was:
This is because
org.apache.ignite.internal.processors.cache.persistence.IgniteCacheDatabaseSharedManager#reserveHistoryForExchange
is called before
org.apache.ignite.internal.processors.cache.persistence.IgniteCacheDatabaseSharedManager#beforeExchange,
which restores correct partition state.
{noformat}
public void testHistory() throws Exception {
IgniteEx crd = startGrid(0);
startGrid(1);
crd.cluster().active(true);
awaitPartitionMapExchange();
int part = 0;
List<Integer> keys = loadDataToPartition(part, DEFAULT_CACHE_NAME, 100,
0, 1);
forceCheckpoint(); // Prevent IGNITE-10088
stopAllGrids();
awaitPartitionMapExchange();
List<Integer> keys1 = loadDataToPartition(part, DEFAULT_CACHE_NAME,
100, 100, 1);
crd = startGrid(0);
startGrid(1);
crd.cluster().active(true);
awaitPartitionMapExchange(); // grid0 will not provide history.
}
{noformat}
> Node is mistakenly excluded from history suppliers preventing historical
> rebalance.
> -----------------------------------------------------------------------------------
>
> Key: IGNITE-10117
> URL: https://issues.apache.org/jira/browse/IGNITE-10117
> Project: Ignite
> Issue Type: Bug
> Reporter: Alexei Scherbakov
> Priority: Major
> Fix For: 2.8
>
>
> This is because
> org.apache.ignite.internal.processors.cache.persistence.IgniteCacheDatabaseSharedManager#reserveHistoryForExchange
> is called before
> org.apache.ignite.internal.processors.cache.persistence.IgniteCacheDatabaseSharedManager#beforeExchange,
> which restores correct partition state.
> {noformat}
> public void testHistory() throws Exception {
> IgniteEx crd = startGrid(0);
> startGrid(1);
> crd.cluster().active(true);
> awaitPartitionMapExchange();
> int part = 0;
> List<Integer> keys = loadDataToPartition(part, DEFAULT_CACHE_NAME,
> 100, 0, 1);
> forceCheckpoint(); // Prevent IGNITE-10088
> stopGrid(1);
> awaitPartitionMapExchange();
> List<Integer> keys1 = loadDataToPartition(part, DEFAULT_CACHE_NAME,
> 100, 100, 1);
> stopAllGrids();
>
> crd = startGrid(0);
> startGrid(1);
> crd.cluster().active(true);
> awaitPartitionMapExchange(); // grid0 will not provide history.
> }
> {noformat}
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
