[
https://issues.apache.org/jira/browse/IGNITE-1012?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Pavel Tupitsyn closed IGNITE-1012.
-----------------------------------
Assignee: Vladimir Ozerov (was: Pavel Tupitsyn)
> Not all entries are returned by scan query in case of concurrent partition
> exchange.
> ------------------------------------------------------------------------------------
>
> Key: IGNITE-1012
> URL: https://issues.apache.org/jira/browse/IGNITE-1012
> Project: Ignite
> Issue Type: Bug
> Components: cache
> Affects Versions: sprint-4
> Reporter: Vladimir Ozerov
> Assignee: Vladimir Ozerov
> Priority: Critical
>
> The problem can be reproduced using two nodes with partitioned cache and the
> following code snippet:
> {code}
> Ignite grid = Ignition.ignite("grid-0");
> IgniteCache cache = grid.cache("cache");
> Affinity aff = grid.affinity("cache");
> // Populate.
> Set<Integer> data = new HashSet<>();
> for (int i = 0; i < 100; i++) {
> cache.put(i, i);
> data.add(i);
> }
> // Check.
> for (int part = 0; part < aff.partitions(); part++) {
> Set<Integer> partData = new HashSet<>();
> for (Integer key : data) {
> if (aff.partition(key) == part)
> partData.add(key);
> }
> ScanQuery qry = new ScanQuery(part);
> List<Cache.Entry<Integer, Integer>> qryData =
> cache.query(qry).getAll();
> for (Cache.Entry<Integer, Integer> entry : qryData)
> partData.remove(entry.getKey());
> assert partData.isEmpty() : "Expected partition keys: " +
> partData;
> }
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
