[
https://issues.apache.org/jira/browse/DRILL-7956?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hanoch Yang updated DRILL-7956:
-------------------------------
Description:
I used 1.18.0 to install the drill locally and installed an hbase storage
plugin. When I executed the query continuously, when the fifth query was
submitted, the REST API started to become unresponsive and the web page could
not be accessed. After a few minutes, everything returned to normal. However,
during this period of time, the REST API was unavailable, and the logs and
system monitoring did not find any abnormalities. Why?
Every sigle query will spent 7min to execute
This is hbase config:
{
"type": "hbase",
"config": {
"hbase.zookeeper.quorum": "ip,ip,ip,ip",
"hbase.zookeeper.property.clientPort": "2181"
},
"size.calculator.enabled": false,
"enabled": true
}
This is the query statement:
select * from `hbase`.`qqq` where CONVERT_FROM(`qqq`.c1.ip,'UTF8') =
'10.38.12.94' and CONVERT_FROM(`qqq`.c1.`time`,'UTF8') >= '1621785600000' and
CONVERT_FROM(`qqq`.c1.`time`,'UTF8') <= '1624464000000'
was:
I used 1.18.0 to install the drill locally and installed an hbase storage
plugin. When I executed the query continuously, when the fifth query was
submitted, the REST API started to become unresponsive and the web page could
not be accessed. After a few minutes, everything returned to normal. However,
during this period of time, the REST API was unavailable, and the logs and
system monitoring did not find any abnormalities. Why?
This is the query statement:
select * from `hbase`.`qqq` where CONVERT_FROM(`qqq`.c1.ip,'UTF8') =
'10.38.12.94' and CONVERT_FROM(`qqq`.c1.`time`,'UTF8') >= '1621785600000' and
CONVERT_FROM(`qqq`.c1.`time`,'UTF8') <= '1624464000000'
> The web does not respond when there are too many queries
> --------------------------------------------------------
>
> Key: DRILL-7956
> URL: https://issues.apache.org/jira/browse/DRILL-7956
> Project: Apache Drill
> Issue Type: Bug
> Components: Client - HTTP
> Affects Versions: 1.18.0
> Reporter: Hanoch Yang
> Priority: Major
>
> I used 1.18.0 to install the drill locally and installed an hbase storage
> plugin. When I executed the query continuously, when the fifth query was
> submitted, the REST API started to become unresponsive and the web page could
> not be accessed. After a few minutes, everything returned to normal. However,
> during this period of time, the REST API was unavailable, and the logs and
> system monitoring did not find any abnormalities. Why?
> Every sigle query will spent 7min to execute
> This is hbase config:
>
> {
> "type": "hbase",
> "config": {
> "hbase.zookeeper.quorum": "ip,ip,ip,ip",
> "hbase.zookeeper.property.clientPort": "2181"
> },
> "size.calculator.enabled": false,
> "enabled": true
> }
> This is the query statement:
> select * from `hbase`.`qqq` where CONVERT_FROM(`qqq`.c1.ip,'UTF8') =
> '10.38.12.94' and CONVERT_FROM(`qqq`.c1.`time`,'UTF8') >= '1621785600000' and
> CONVERT_FROM(`qqq`.c1.`time`,'UTF8') <= '1624464000000'
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
