[ 
https://issues.apache.org/jira/browse/IOTDB-1209?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17312113#comment-17312113
 ] 

Houliang Qi commented on IOTDB-1209:
------------------------------------

Hi, [~ericpai], the picture you post can not be seen. The tsfile is created by 
the iotdb-benchmark, but I think you can just write the records as the above 
Step2 shows and then execute the flush command, you will get one new tsfile and 
clean all the data and reproduce the error.

> When loading a tsfile with a large timestamp, the data is incorrect
> -------------------------------------------------------------------
>
>                 Key: IOTDB-1209
>                 URL: https://issues.apache.org/jira/browse/IOTDB-1209
>             Project: Apache IoTDB
>          Issue Type: Bug
>    Affects Versions: 0.11.2
>            Reporter: Houliang Qi
>            Assignee: Eric Pai
>            Priority: Critical
>         Attachments: 1.1.png, 1.2.png, 1.3.png, 1.4.png, 1.5.png, 
> 1678502448000-1-0.tsfile
>
>
> Step 1: 
>  # insert into root.group_1.d_1(timestamp, s_1, s_0, s_2) 
> values(2018-07-18T00:00:00.000+08:00, 18, false, 18.18)
>  # flush
>  # insert into root.group_1.d_1(timestamp, s_1, s_0, s_2) 
> values(2018-07-19T00:00:00.000+08:00, 19, false, 19.18)
>  # flush
> !1.1.png!
>  
> Step2:
> load one tsfile which the tsfile name have a  large timestamp, but its data's 
> timestamp are small. 
>  #  load "../1678502448000-1-0.tsfile" 
>  # !1.2.png!
>  # after load the tsfile, we can see that the root.group_1.d_1 has 3 rows as 
> below. 
>  # !1.3.png!
>  
> Step3:insert another two rows
>  # insert into root.group_1.d_1(timestamp, s_1, s_0, s_2) 
> values(2019-07-19T00:00:00.000+08:00, 2019, false, 19.18)
>  # insert into root.group_1.d_1(timestamp, s_1, s_0, s_2) 
> values(2019-08-19T00:00:00.000+08:00, 2018, false, 19.18)
>  # we can see that the query result  is rowng when we query the 
> root.group_1.d_1
> !1.4.png!
>  
> Step4 : Query with the order by time keyword, we can see from the result that 
> it have two problems:
>  #  it's not ordered by time
>  # time value of time 2018-07-18T00:00:00.000+08:00 (s1, s_0, s2)are (18, 
> false, 18.18 ) which we insert in step one, however, the result show is 
> another vlaue.
>   !1.5.png!
>  
>  
>  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to