mapreduce.DefaultVisibilityExpressionResolver: Error scanning 'labels' table #4
Comments
|
did you get the workaround? I was also getting the same. |
|
How to fix this? |
|
Damn, can't remember right now... ! Apparently, you're not the only one: @jean-marc, have you seen this before ? On Fri, Oct 7, 2016 at 10:42 PM, Andrew Zhang [email protected]
|
|
Do you have the full stack trace? 2016-10-08 1:00 GMT-04:00 David Lauzon [email protected]:
|
|
ya, as follows: |
|
Trying again to tag Jean-Marc into discussion: @jmspaggi |
|
Hi @startprogress, does it work in spite of the error message? If not, could you post a step-by-step to reproduce your problem? |
|
I am afraid you have the wrong Jean-Marc jm On Sat, Oct 8, 2016 at 11:33 AM, David Lauzon [email protected]
|
|
Worked ;) Is this stacktrace causing any issue? What HBase version are you trying? As David pointed before on JIRA, if cell level security is not used, this "ERROR" can be considered as an INFO or a WARNING... Is HBase stilll working well after that? JMS |
|
I've run a simple test which throws the error, but works anyhow: Importing the CSV: cat <<EOF >> /tmp/simple.csv
a,b,c
1,2,4
5,6,8
EOF
hbase shell create 'simpletable', 'cf'
hbase org.apache.hadoop.hbase.mapreduce.ImportTsv -Dimporttsv.columns=HBASE_ROW_KEY,cf:b,cf:c '-Dimporttsv.separator=,' simpletable /tmp/simple.csvValidating that the data is there: @startprogress: could you confirm on your side? |
|
@davidonlaptop But with my own data(about 140GB), after the course of Importtsv, the hbase table is still empty. So I think some other errors might exist and they were not logged thus hard to be found out. |
|
Nothing else on the logs? What does your csv file looks like? 2016-10-08 22:00 GMT-04:00 Andrew Zhang [email protected]:
|
|
@jmspaggi my file is tsv file, each field is separated by tab. |
|
@jmspaggi Maybe this is the real problem |
|
This is "just" a timeout so MR should retry. Your table is completely JMS 2016-10-08 22:59 GMT-04:00 Andrew Zhang [email protected]:
|
|
Also, you could try with a subset of your data first, like the first 10 lines. |
|
@davidonlaptop |
|
@jmspaggi |
|
Oh, interesting! So you are saying that doing ImportTsv with namespace JMS 2016-10-10 5:34 GMT-04:00 Andrew Zhang [email protected]:
|
|
@jmspaggi Those parameters are calculated from the size of the file and the configuration of my cluster. |
|
@startprogress : FYI. I just pushed a new Dockerfile with all the latest versions (HBase 1.2.3, OpenJDK 8, Ubuntu 16.04) if you want to give it a try. Maybe, it can solve your issue. |
|
@startprogress what is the value of tableName? Any chance to try with 2016-10-11 9:54 GMT-04:00 David Lauzon [email protected]:
|
|
@jmspaggi tableName is changed to 't1' instead of 'n1:t1', the command worked well now. |
|
@davidonlaptop Ok, thx for ur help. |
|
@startprogress What I'm wondering is that, can you give "n1:t1" instead of 2016-10-12 5:13 GMT-04:00 Andrew Zhang [email protected]:
|
|
@jmspaggi |
HBase complains about a missing table when importing data using ImportTsv.
The text was updated successfully, but these errors were encountered: