[
https://issues.apache.org/jira/browse/HBASE-28187?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ke Han updated HBASE-28187:
---------------------------
Description:
Flush a columnfamily that doesn't exist in the table will cause NPE ERROR in
both shell and the HMaster logs.
h1. Reproduce
Start up HBase 2.5.5 cluster, executing the following commands with hbase shell
in HMaster node will lead to NPE. (Can be reproduced determinstically)
{code:java}
create 'table', {NAME => 'cf1', VERSIONS => 2, COMPRESSION => 'GZ', BLOOMFILTER
=> 'ROWCOL'}, {NAME => 'cf2', VERSIONS => 4, COMPRESSION => 'NONE', BLOOMFILTER
=> 'ROWCOL'}
incr 'table', 'row1', 'cf1:cell', 2
flush 'table', 'cf3'{code}
The shell outputs
{code:java}
hbase:006:0> create 'table', {NAME => 'cf1', VERSIONS => 2, COMPRESSION =>
'GZ', BLOOMFILTER => 'ROWCOL'}, {NAME => 'cf2', VERSIONS => 4, COMPRESSION =>
'NONE', BLOOMFILTER => 'ROWCOL'}
Created table table
Took 2.1238 seconds
=> Hbase::Table - table
hbase:007:0>
hbase:008:0> incr 'table', 'row1', 'cf1:cell', 2
COUNTER VALUE = 2
Took 0.0131 seconds
hbase:009:0>
hbase:010:0> flush 'table', 'cf3'
ERROR: java.io.IOException: java.lang.NullPointerException
at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:479)
at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:124)
at org.apache.hadoop.hbase.ipc.RpcHandler.run(RpcHandler.java:102)
at org.apache.hadoop.hbase.ipc.RpcHandler.run(RpcHandler.java:82)
Caused by:
org.apache.hadoop.hbase.errorhandling.ForeignException$ProxyThrowable:
java.lang.NullPointerException
at
org.apache.hadoop.hbase.procedure.flush.RegionServerFlushTableProcedureManager$FlushTableSubprocedurePool.waitForOutstandingTasks(RegionServerFlushTableProcedureManager.java:274)
at
org.apache.hadoop.hbase.procedure.flush.FlushTableSubprocedure.flushRegions(FlushTableSubprocedure.java:115)
at
org.apache.hadoop.hbase.procedure.flush.FlushTableSubprocedure.acquireBarrier(FlushTableSubprocedure.java:126)
at org.apache.hadoop.hbase.procedure.Subprocedure.call(Subprocedure.java:160)
at org.apache.hadoop.hbase.procedure.Subprocedure.call(Subprocedure.java:46)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:750)
For usage try 'help "flush"'
Took 12.1713 seconds
{code}
According to the _flush (flush.rb)_ command specification, user can flush a
specific column family.
{code:java}
Flush all regions in passed table or pass a region row to
flush an individual region or a region server name whose format
is 'host,port,startcode', to flush all its regions.
You can also flush a single column family for all regions within a table,
or for an specific region only.
For example:
hbase> flush 'TABLENAME'
hbase> flush 'TABLENAME','FAMILYNAME' {code}
In the above case, *cf3* an incorrect input (non-existing column family). If
user tries to flush it, the expected output is:
# HBase rejects this operation
# returns a prompt saying the column family doesn't exist {_}"{_}{_}{+}ERROR:
Unknown CF...{+}".{_}
h1. Root Cause
There's a missing check for the whether the target flushing columnfamily exists.
was:
Flush a columnfamily that doesn't exist in the table will cause NPE ERROR in
both shell and the HMaster logs.
h1. Reproduce
Start up HBase 2.5.5 cluster, executing the following commands with hbase shell
in HMaster node will lead to NPE. (Can be reproduced determinstically)
{code:java}
create 'table', {NAME => 'cf1', VERSIONS => 2, COMPRESSION => 'GZ', BLOOMFILTER
=> 'ROWCOL'}, {NAME => 'cf2', VERSIONS => 4, COMPRESSION => 'NONE', BLOOMFILTER
=> 'ROWCOL'}
incr 'table', 'row1', 'cf1:cell', 2
flush 'table', 'cf3'{code}
The shell outputs
{code:java}
hbase:006:0> create 'table', {NAME => 'cf1', VERSIONS => 2, COMPRESSION =>
'GZ', BLOOMFILTER => 'ROWCOL'}, {NAME => 'cf2', VERSIONS => 4, COMPRESSION =>
'NONE', BLOOMFILTER => 'ROWCOL'}
Created table table
Took 2.1238 seconds
=> Hbase::Table - table
hbase:007:0>
hbase:008:0> incr 'table', 'row1', 'cf1:cell', 2
COUNTER VALUE = 2
Took 0.0131 seconds
hbase:009:0>
hbase:010:0> flush 'table', 'cf3'
ERROR: java.io.IOException: java.lang.NullPointerException
at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:479)
at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:124)
at org.apache.hadoop.hbase.ipc.RpcHandler.run(RpcHandler.java:102)
at org.apache.hadoop.hbase.ipc.RpcHandler.run(RpcHandler.java:82)
Caused by:
org.apache.hadoop.hbase.errorhandling.ForeignException$ProxyThrowable:
java.lang.NullPointerException
at
org.apache.hadoop.hbase.procedure.flush.RegionServerFlushTableProcedureManager$FlushTableSubprocedurePool.waitForOutstandingTasks(RegionServerFlushTableProcedureManager.java:274)
at
org.apache.hadoop.hbase.procedure.flush.FlushTableSubprocedure.flushRegions(FlushTableSubprocedure.java:115)
at
org.apache.hadoop.hbase.procedure.flush.FlushTableSubprocedure.acquireBarrier(FlushTableSubprocedure.java:126)
at org.apache.hadoop.hbase.procedure.Subprocedure.call(Subprocedure.java:160)
at org.apache.hadoop.hbase.procedure.Subprocedure.call(Subprocedure.java:46)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:750)
For usage try 'help "flush"'
Took 12.1713 seconds
{code}
According to the _flush (flush.rb)_ command specification, user can flush a
specific column family.
{code:java}
sh all regions in passed table or pass a region row to
flush an individual region or a region server name whose format
is 'host,port,startcode', to flush all its regions.
You can also flush a single column family for all regions within a table,
or for an specific region only.
For example:
hbase> flush 'TABLENAME'
hbase> flush 'TABLENAME','FAMILYNAME' {code}
In the above case, *cf3* an incorrect input (non-existing column family). If
user tries to flush it, the expected output is:
# HBase rejects this operation
# returns a prompt saying the column family doesn't exist {_}"{_}{_}{+}ERROR:
Unknown CF...{+}".{_}
h1. Root Cause
There's a missing check for the whether the target flushing columnfamily exists.
> NPE when flushing a non-existing column family
> ----------------------------------------------
>
> Key: HBASE-28187
> URL: https://issues.apache.org/jira/browse/HBASE-28187
> Project: HBase
> Issue Type: Bug
> Reporter: Ke Han
> Priority: Major
>
> Flush a columnfamily that doesn't exist in the table will cause NPE ERROR in
> both shell and the HMaster logs.
> h1. Reproduce
> Start up HBase 2.5.5 cluster, executing the following commands with hbase
> shell in HMaster node will lead to NPE. (Can be reproduced determinstically)
> {code:java}
> create 'table', {NAME => 'cf1', VERSIONS => 2, COMPRESSION => 'GZ',
> BLOOMFILTER => 'ROWCOL'}, {NAME => 'cf2', VERSIONS => 4, COMPRESSION =>
> 'NONE', BLOOMFILTER => 'ROWCOL'}
> incr 'table', 'row1', 'cf1:cell', 2
> flush 'table', 'cf3'{code}
> The shell outputs
> {code:java}
> hbase:006:0> create 'table', {NAME => 'cf1', VERSIONS => 2, COMPRESSION =>
> 'GZ', BLOOMFILTER => 'ROWCOL'}, {NAME => 'cf2', VERSIONS => 4, COMPRESSION =>
> 'NONE', BLOOMFILTER => 'ROWCOL'}
> Created table table
> Took 2.1238 seconds
>
> => Hbase::Table - table
> hbase:007:0>
> hbase:008:0> incr 'table', 'row1', 'cf1:cell', 2
> COUNTER VALUE = 2
> Took 0.0131 seconds
>
> hbase:009:0>
> hbase:010:0> flush 'table', 'cf3'
> ERROR: java.io.IOException: java.lang.NullPointerException
> at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:479)
> at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:124)
> at org.apache.hadoop.hbase.ipc.RpcHandler.run(RpcHandler.java:102)
> at org.apache.hadoop.hbase.ipc.RpcHandler.run(RpcHandler.java:82)
> Caused by:
> org.apache.hadoop.hbase.errorhandling.ForeignException$ProxyThrowable:
> java.lang.NullPointerException
> at
> org.apache.hadoop.hbase.procedure.flush.RegionServerFlushTableProcedureManager$FlushTableSubprocedurePool.waitForOutstandingTasks(RegionServerFlushTableProcedureManager.java:274)
> at
> org.apache.hadoop.hbase.procedure.flush.FlushTableSubprocedure.flushRegions(FlushTableSubprocedure.java:115)
> at
> org.apache.hadoop.hbase.procedure.flush.FlushTableSubprocedure.acquireBarrier(FlushTableSubprocedure.java:126)
> at org.apache.hadoop.hbase.procedure.Subprocedure.call(Subprocedure.java:160)
> at org.apache.hadoop.hbase.procedure.Subprocedure.call(Subprocedure.java:46)
> at java.util.concurrent.FutureTask.run(FutureTask.java:266)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:750)
> For usage try 'help "flush"'
> Took 12.1713 seconds
> {code}
>
> According to the _flush (flush.rb)_ command specification, user can flush a
> specific column family.
> {code:java}
> Flush all regions in passed table or pass a region row to
> flush an individual region or a region server name whose format
> is 'host,port,startcode', to flush all its regions.
> You can also flush a single column family for all regions within a table,
> or for an specific region only.
> For example:
> hbase> flush 'TABLENAME'
> hbase> flush 'TABLENAME','FAMILYNAME' {code}
> In the above case, *cf3* an incorrect input (non-existing column family). If
> user tries to flush it, the expected output is:
> # HBase rejects this operation
> # returns a prompt saying the column family doesn't exist
> {_}"{_}{_}{+}ERROR: Unknown CF...{+}".{_}
> h1. Root Cause
> There's a missing check for the whether the target flushing columnfamily
> exists.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
