ignite-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Pavel Konstantinov (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (IGNITE-737) ClusterGroup.forDataNodes() returns empty cluster group for daemon node
Date Mon, 20 Apr 2015 04:21:58 GMT

    [ https://issues.apache.org/jira/browse/IGNITE-737?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14502315#comment-14502315
] 

Pavel Konstantinov edited comment on IGNITE-737 at 4/20/15 4:21 AM:
--------------------------------------------------------------------

2. Ask for node in case when we clear local cache
3. 'cache -clear' shows different size in cache list table {code}
+----+----------------------+-------------+----------------------------------------+---------------+
| 5  | dcp_1(@c5)           | PARTITIONED | [f896e2da-526a-4ce2-958f-c197593e0f00] | min:
22033    |
|    |                      |             |                                        | avg:
22033.00 |
|    |                      |             |                                        | max:
22033    |
+----+----------------------+-------------+----------------------------------------+---------------+
{code}
and in results table
{code}
Choose cache number ('c' to cancel) [c]: 5
Cleared cache with name: dcp_1
+======================================================+
|  Node ID8(@)  | Cache Size Before | Cache Size After |
+======================================================+
| F896E2DA(@n1) | 46751             | 0                |
+------------------------------------------------------+
{code}
The reason is that the first table shows size taken only from one node(incorrect) and the
second entire the cache(correct)
4. we should sum size for local cache in first table if node is not specified for 'cache -clear'


was (Author: pkonstantinov):
2. Ask for node in case when we clear local cache
3. 'cache -clear' shows different size in cache list table {code}
+----+----------------------+-------------+----------------------------------------+---------------+
| 5  | dcp_1(@c5)           | PARTITIONED | [f896e2da-526a-4ce2-958f-c197593e0f00] | min:
22033    |
|    |                      |             |                                        | avg:
22033.00 |
|    |                      |             |                                        | max:
22033    |
+----+----------------------+-------------+----------------------------------------+---------------+
{code}
and in results table
{code}
Choose cache number ('c' to cancel) [c]: 5
Cleared cache with name: dcp_1
+======================================================+
|  Node ID8(@)  | Cache Size Before | Cache Size After |
+======================================================+
| F896E2DA(@n1) | 46751             | 0                |
+------------------------------------------------------+
{code}
The reason is that the first table shows size taken only from one node and the second entire
the cache
4. we should sum size for local cache in first table if node is not specified for 'cache -clear'

> ClusterGroup.forDataNodes() returns empty cluster group for daemon node
> -----------------------------------------------------------------------
>
>                 Key: IGNITE-737
>                 URL: https://issues.apache.org/jira/browse/IGNITE-737
>             Project: Ignite
>          Issue Type: Bug
>    Affects Versions: sprint-3
>            Reporter: Alexey Kuznetsov
>            Assignee: Pavel Konstantinov
>             Fix For: sprint-4
>
>         Attachments: ignite-737-test.patch, ignite-737.patch
>
>
> To reproduce:
> Start one ore several nodes with cache "A".
> Start daemon node.
> On daemon node execute ignite.cluster.forDataNodes("A").
> Returned projection should contain remote nodes with cache "A".
> Appropriate tests should also be added.
> Also it will be useful to check all cluster.forXXX() methods on daemon node.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message