[
https://issues.apache.org/jira/browse/IOTDB-4637?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17616881#comment-17616881
]
Yuan Tian commented on IOTDB-4637:
----------------------------------
It can be seen that the root cause is that IoTDB has opened too many files. In
current query processing, we will keep all the related files opened until this
query finished and this will be optimized in the future which can eliminate
this error.
Before we do that optimization, you can try 2 methods:
# modify the system max allowed opened file handles
# set compaction to true and compaction can reduce the file numbers.
!image-2022-10-13-16-42-11-268.png!
> [0.13.3]In the case of a large amount of data, perform an aggregate query to
> report NPE
> ---------------------------------------------------------------------------------------
>
> Key: IOTDB-4637
> URL: https://issues.apache.org/jira/browse/IOTDB-4637
> Project: Apache IoTDB
> Issue Type: Bug
> Components: Core/Query
> Affects Versions: 0.13.3
> Reporter: xiaozhihong
> Assignee: Yuan Tian
> Priority: Major
> Attachments: image-2022-10-13-16-42-03-326.png,
> image-2022-10-13-16-42-11-268.png, log_all.log
>
>
> In Version V0.13.3, I prepared 208G data files: 1 storage group, 1 device,
> 500 timeseries.
> After started server, then enter CLI, executed aggregate sql.
> For a while, repored NPE message.
> {code:java}
> IoTDB> select count(s_0) from root.test.g_0.d_0 ;
> Msg: 500: [INTERNAL_SERVER_ERROR(500)] Exception occurred: "select count(s_0)
> from root.test.g_0.d_0". executeStatement failed.
> java.lang.NullPointerException {code}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
