You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@drill.apache.org by "Ramana Inukonda Nagaraj (JIRA)" <ji...@apache.org> on 2015/04/14 23:53:59 UTC
[jira] [Comment Edited] (DRILL-2792) Killing the drillbit which is
the foreman results in direct memory being held on
[ https://issues.apache.org/jira/browse/DRILL-2792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14494980#comment-14494980 ]
Ramana Inukonda Nagaraj edited comment on DRILL-2792 at 4/14/15 9:53 PM:
-------------------------------------------------------------------------
In the case where foreman is killed direct memory use looks like this:
Before cancellation:
{code}
1/6 select * from sys.memory;
+------------+------------+--------------+------------+----------------+------------+
| hostname | user_port | heap_current | heap_max | direct_current | direct_max |
+------------+------------+--------------+------------+----------------+------------+
| atsqa6c64.qa.lab | 31010 | 184728000 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c58.qa.lab | 31010 | 179768952 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c62.qa.lab | 31010 | 228472528 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c63.qa.lab | 31010 | 325782456 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c61.qa.lab | 31010 | 133183360 | 4151836672 | 6048768 | 34359738368 |
| atsqa6c57.qa.lab | 31010 | 323590544 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c60.qa.lab | 31010 | 179155280 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c59.qa.lab | 31010 | 187145648 | 4151836672 | 2000000 | 34359738368 |
+------------+------------+--------------+------------+----------------+------------+
8 rows selected (2.327 seconds)
{code}
After cancellation:
{code}
0: jdbc:drill:> select * from sys.memory;
+------------+------------+--------------+------------+----------------+------------+
| hostname | user_port | heap_current | heap_max | direct_current | direct_max |
+------------+------------+--------------+------------+----------------+------------+
| atsqa6c63.qa.lab | 31010 | 514486248 | 4151836672 | 1839362458 | 34359738368 |
| atsqa6c57.qa.lab | 31010 | 382623688 | 4151836672 | 1849218886 | 34359738368 |
| atsqa6c60.qa.lab | 31010 | 381417200 | 4151836672 | 1850727912 | 34359738368 |
| atsqa6c64.qa.lab | 31010 | 488955320 | 4151836672 | 1855641859 | 34359738368 |
| atsqa6c59.qa.lab | 31010 | 426557624 | 4151836672 | 1822002015 | 34359738368 |
| atsqa6c62.qa.lab | 31010 | 532419120 | 4151836672 | 1876725045 | 34359738368 |
| atsqa6c58.qa.lab | 31010 | 333926896 | 4151836672 | 1848092666 | 34359738368 |
+------------+------------+--------------+------------+----------------+------------+
7 rows selected (0.151 seconds)
{code}
was (Author: inramana):
In the case where foreman is killed direct memory use looks like this:
Before cancellation:
{code}
1/6 select * from sys.memory;
+------------+------------+--------------+------------+----------------+------------+
| hostname | user_port | heap_current | heap_max | direct_current | direct_max |
+------------+------------+--------------+------------+----------------+------------+
| atsqa6c64.qa.lab | 31010 | 184728000 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c58.qa.lab | 31010 | 179768952 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c62.qa.lab | 31010 | 228472528 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c63.qa.lab | 31010 | 325782456 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c61.qa.lab | 31010 | 133183360 | 4151836672 | 6048768 | 34359738368 |
| atsqa6c57.qa.lab | 31010 | 323590544 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c60.qa.lab | 31010 | 179155280 | 4151836672 | 2000000 | 34359738368 |
| atsqa6c59.qa.lab | 31010 | 187145648 | 4151836672 | 2000000 | 34359738368 |
+------------+------------+--------------+------------+----------------+------------+
8 rows selected (2.327 seconds)
{code}
After cancellation:
0: jdbc:drill:> select * from sys.memory;
+------------+------------+--------------+------------+----------------+------------+
| hostname | user_port | heap_current | heap_max | direct_current | direct_max |
+------------+------------+--------------+------------+----------------+------------+
| atsqa6c63.qa.lab | 31010 | 514486248 | 4151836672 | 1839362458 | 34359738368 |
| atsqa6c57.qa.lab | 31010 | 382623688 | 4151836672 | 1849218886 | 34359738368 |
| atsqa6c60.qa.lab | 31010 | 381417200 | 4151836672 | 1850727912 | 34359738368 |
| atsqa6c64.qa.lab | 31010 | 488955320 | 4151836672 | 1855641859 | 34359738368 |
| atsqa6c59.qa.lab | 31010 | 426557624 | 4151836672 | 1822002015 | 34359738368 |
| atsqa6c62.qa.lab | 31010 | 532419120 | 4151836672 | 1876725045 | 34359738368 |
| atsqa6c58.qa.lab | 31010 | 333926896 | 4151836672 | 1848092666 | 34359738368 |
+------------+------------+--------------+------------+----------------+------------+
7 rows selected (0.151 seconds)
> Killing the drillbit which is the foreman results in direct memory being held on
> --------------------------------------------------------------------------------
>
> Key: DRILL-2792
> URL: https://issues.apache.org/jira/browse/DRILL-2792
> Project: Apache Drill
> Issue Type: Bug
> Components: Execution - Flow
> Reporter: Ramana Inukonda Nagaraj
> Assignee: Chris Westin
>
> Killed one of the drillbits which is the foreman for the query-
> Profiles page reports that query has cancelled.
> Due to bug Drill-2778 sqlline hangs. However after killing sqlline the current direct memory used does not go down to pre query levels.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)