admin-shell doesnt work bigtable-hbase-2x-hadoop

[igorbernstein2]

create 'test', 'cf'
returns a ERROR: Not implemented error

With jruby configured to dump stacktraces via .jrubrc:

log.exceptions=true
log.exceptions=true
log.backtraces=true
log.callers=true
errno.backtrace=true

I see:

2021-03-03T13:10:45.563-05:00 [main] INFO TraceType : Backtrace generated:
                            getHbck at org/apache/hadoop/hbase/client/Connection.java:210
                            invoke0 at jdk/internal/reflect/NativeMethodAccessorImpl.java:-2
                             invoke at jdk/internal/reflect/NativeMethodAccessorImpl.java:62
                             invoke at jdk/internal/reflect/DelegatingMethodAccessorImpl.java:43
                             invoke at java/lang/reflect/Method.java:566
  invokeDirectWithExceptionHandling at org/jruby/javasupport/JavaMethod.java:438
                       invokeDirect at org/jruby/javasupport/JavaMethod.java:302
                         initialize at /Users/igorbernstein/Downloads/hbase-2.2.6-client/lib/ruby/hbase/admin.rb:40
                        newInstance at org/jruby/RubyClass.java:1001
                               call at org/jruby/RubyClass$INVOKER$i$newInstance.gen:-1
                              admin at /Users/igorbernstein/Downloads/hbase-2.2.6-client/lib/ruby/hbase/hbase.rb:50
                              admin at /Users/igorbernstein/Downloads/hbase-2.2.6-client/lib/ruby/shell.rb:87
                              admin at /Users/igorbernstein/Downloads/hbase-2.2.6-client/lib/ruby/shell/commands.rb:77
                            command at /Users/igorbernstein/Downloads/hbase-2.2.6-client/lib/ruby/shell/commands/create.rb:67

Which points to this line:
https://github.com/apache/hbase/blob/rel/2.2.6/hbase-shell/src/main/ruby/hbase/admin.rb#L40

Which calls getHbck(), which we dont currently implement in BigtableConnection.

It seems like the feature was introduced after hbase 2.2.0 in this PR:
apache/hbase#425

To reproduce:

1. download hbase-2.2.6-client-bin.tar.gz
2. Add bigtable-hbase-2.x-hadoop-1.19.0.jar into the lib/
3. Update conf/hbase-site.xml to include:

<property>
    <name>hbase.client.connection.impl</name>
    <value>com.google.cloud.bigtable.hbase2_x.BigtableConnection</value>
  </property>
  <property>
    <name>google.bigtable.project.id</name>
    <value>MY-PROJECT</value>
  </property>
  <property>
    <name>google.bigtable.namespace.warnings</name>
    <value>true</value>
  </property>
  <property>
    <name>google.bigtable.instance.id</name>
    <value>MY-INSTANCE</value>
  </property>
  <property>
    <name>hbase.client.async.connection.impl</name>
    <value>org.apache.hadoop.hbase.client.BigtableAsyncConnection</value>
  </property>
  <property>
    <name>hbase.client.registry.impl</name>
    <value>org.apache.hadoop.hbase.client.BigtableAsyncRegistry</value>
  </property>

4. Add .jrubyrc with

log.exceptions=true
log.exceptions=true
log.backtraces=true
log.callers=true
errno.backtrace=true

5. run ./bin/hbase shell
6. try creating a table: create 'test', 'cf'

[igorbernstein2]

I think this can be fixed by updating BigtableConnection to return a no-op implementation Hbck. But I need to do a bit more research as to what Hbck does and what are the reprecussions

[igorbernstein2]

Also, there might be other blockers for hbase-shell, but this is the first hurdle