2

I am currently using hbase-indexer to index hbase in solr. When I execute foolowing command to check the indexer, 

hbase-indexer$ bin/hbase-indexer list-indexers --zookeeper 127.0.0.1:2181

The result is said that: 

myindexer
+ Lifecycle state: ACTIVE 
+ Incremental indexing state: SUBSCRIBE_AND_CONSUME
+ Batch indexing state: INACTIVE
+ SEP subscription ID: Indexer_myindexer
+ SEP subscription timestamp: 2017-01-24T13:15:48.614+09:00
+ Connection type: solr
+ Connection params:
  + solr.zk = localhost:2181/solr
  + solr.collection = tagcollect
+ Indexer config:
    222 bytes, use -dump to see content
+ Indexer component factory:     
com.ngdata.hbaseindexer.conf.DefaultIndexerComponentFactory
+ Additional batch index CLI arguments:
  (none)
+ Default additional batch index CLI arguments:
  (none)
+ Processes
  + 1 running processes
  + 0 failed processes

I think hbase-indexer works well as shown above, because it is displayed as + 1 running processes.(Prior to this, I've already executed hbase-indexer daemon by the command : ~$ bin/hbase-indexer server )

For test, I've insert data in Hbase through put command and checked the data was inserted.

But, solr qry said following that: (No Record)

I wish your knowledge and experience associated with this to be shared. Thank you. 

{
"responseHeader":{
"zkConnected":true,
"status":0,
"QTime":7,
"params":{
  "q":"*:*",
  "indent":"on",
  "wt":"json",
  "_":"1485246329559"}},
"response":{"numFound":0,"start":0,"maxScore":0.0,"docs":[]
}}
Chris Joo
  • 577
  • 10
  • 24

1 Answers1

1

We encountered same issue.

As You are saying sever instance has good health, below are reasons which it wont work.

  • Firstly, If 'Write ahead log'(WAL) is disabled (may be for write performance reasons) then your puts wont create solr documents.

Hbase NRT indexer works on WAL. if its disabled then it wont create solr documents.

  • Second reason may be mophiline configurations if they are not correct then it wont create solr documents

However, I'd suggest to write a custom mapreduce programs(or spark jobs as well) to index solr documents by reading hbase data (if not Real time, that means when ever your put data in to hbase immeditely it wont reflect, after mapreduce solr indexer runs solr documents will be created)

Ram Ghadiyaram
  • 28,239
  • 13
  • 95
  • 121
  • Thank you so much, Ram... How can I check or change HBase WAL setting? I find only 'hbase.master.logcleaner.ttl' and 'hbase.master.logcleaner.plugins'... Second reason seems not applied to me. If I could not solve this issue, I should consider a more intuitive programming approach like spark job or mapreduce programs as you suggest.... – Chris Joo Jan 24 '17 at 13:35
  • Still you are not able to identify the property and its value whether its disabled or enabled....If you are using java/mapreduce client, you can try like this `final Configuration conf = HBaseConfiguration.create(); conf.writeXml(System.out); ` It will print in XML format with all the configuration including this wal property as well...Best of luck – Ram Ghadiyaram Jan 24 '17 at 14:14