cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Gaurav (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (CASSANDRA-9207) COPY FROM command does not restore all records
Date Wed, 29 Apr 2015 09:22:06 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-9207?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14519006#comment-14519006
] 

Gaurav edited comment on CASSANDRA-9207 at 4/29/15 9:21 AM:
------------------------------------------------------------

csv file of the table which cause the issue.
Initially this file had 411 rows and after copying this file had 310 rows.

Notice:
row with the identifier 
push,1.4292e+12,20c57add-b7ca-471a-97b1-0f0f62591709
has been duplicated.

Ideally this should be different since, 1.4292e+12 is unix epoch time and upon expanding it
will be different for all 3 entries.

while restoring, only this hard-string is read and found duplicate. Resulting in some record
missing.  When i copied the content of target (table with 310 records) then on searching with

push,1.4292e+12,20c57add-b7ca-471a-97b1-0f0f62591709 record, i found only 1 entry. Which confirms
the issue.



was (Author: guptakvgaurav):
csv file of the table which cause the issue.
Initially this file had 411 rows and after copying this file had 310 rows.


> COPY FROM command does not restore all records
> ----------------------------------------------
>
>                 Key: CASSANDRA-9207
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-9207
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core, Tools
>         Environment: centOs, cassandra 2.1.1
>            Reporter: Gaurav
>              Labels: COPY, cqlsh
>             Fix For: 2.1.5
>
>         Attachments: cron_task.csv
>
>
> Steps to reproduce:
> 1. run COPY TO command on keyspace A. (on my environment problem occurred on only 1 table
out of 19, only thing that io noticed is this table has >1500 records and other tables
has <1000 records)
> 2. pull the csv file and place it on machine (say machine B)where  other database resides.
(In my case this was another instance of amazon machine)
> 3. Now, run the COPY TO command on machine B. (both keyspaces, one on machine A and one
on machine B has same schema.)
> Observation:
> 1. when COPY TO command is run for table having records > 1500. Command gave following
output:
> Processing 1000 records
> 1573 records copied.
> but when i tried to verify it by running below mentioned commands, i received only 273
records.
> 1. SELECT COUNT(*) FROM table-name;
> 2. SELECT * FROM table-name;
> Note: please let me know if other information needs to be shared with you.
> Also, is there any other way to take the back-up of keyspace and restoring it on other
machine.



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

Mime
View raw message