hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Qiang Tian <tian...@gmail.com>
Subject Re: YCSB load data stucked
Date Thu, 23 Oct 2014 03:06:22 GMT
update:

the problem exists with below repo:
 https://github.com/ramkrish86/ycsb/tree/ycsb_visibility

Andrew's repo is fine:
 https://github.com/apurtell/YCSB/tree/new_hbase_client

On Wed, Oct 22, 2014 at 2:59 PM, Qiang Tian <tianq01@gmail.com> wrote:

> with command "bin/ycsb load hbase -P workloads/workloada -p threadcount=10
> -p columnfamily=f1 -p recordcount=1000  -s", I just see 10 rows in hbase
> shell...I tried it in both standalone and a real 3 node cluster, the same
> result.  a quick look at core/src/main/java/com/yahoo/ycsb/Client.java, the
> parameter looks to be right..
>
> anything wrong?
>
> thanks..
>
>
> On Wed, Oct 22, 2014 at 11:42 AM, Qiang Tian <tianq01@gmail.com> wrote:
>
>> Thanks Ted,
>> I also got Mapkeeper error, comment the module works around it(
>> https://github.com/brianfrankcooper/YCSB/issues/152)
>>
>> On Wed, Oct 22, 2014 at 10:23 AM, Ted Yu <yuzhihong@gmail.com> wrote:
>>
>>> Once you clone ycsb, you should build it with your choice of 0.98
>>>
>>> Here's thread where Andrew mentioned his ycsb repo:
>>>
>>>
>>> http://search-hadoop.com/m/DHED4NaxYb1/andrew+purtell+ycsb+2014&subj=Re+Performance+oddity+between+AWS+instance+sizes
>>>
>>> Cheers
>>>
>>> On Tue, Oct 21, 2014 at 7:15 PM, Qiang Tian <tianq01@gmail.com> wrote:
>>>
>>> > Thanks Ted.
>>> > do you mean I should rebuild the ycsb? could you point me Andrew's
>>> repo?
>>> >
>>> >
>>> > On Tue, Oct 21, 2014 at 5:37 PM, Ted Yu <yuzhihong@gmail.com> wrote:
>>> >
>>> > > Cycling bits:
>>> > > http://search-hadoop.com/m/DHED4N0syk1
>>> > >
>>> > > Andrew has his ycsb repo as well.
>>> > >
>>> > > Cheers
>>> > >
>>> > > On Oct 21, 2014, at 2:28 AM, Qiang Tian <tianq01@gmail.com> wrote:
>>> > >
>>> > > > Hi Guys,
>>> > > > I am running YCSB 0.1.4 against hbase 0.98.5,
>>> > > >
>>> > > > "bin/ycsb load hbase -P workloads/workloada -p columnfamily=f1
-p
>>> > > > recordcount=1000 -p threadcount=4 -s | tee -a workloada.dat"
>>> stucked as
>>> > > > below:
>>> > > >
>>> > > > 10 sec: 0 operations;
>>> > > > 20 sec: 0 operations;
>>> > > > 30 sec: 0 operations;
>>> > > > 40 sec: 0 operations;
>>> > > > 50 sec: 0 operations;
>>> > > > ...
>>> > > >
>>> > > > stacktrace shows YCSB is locating meta region:
>>> > > >
>>> > > > at java/lang/Object.wait(Object.java:201)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/zookeeper/ZooKeeperNodeTracker.blockUntilAvailable(ZooKeeperNodeTracker.java:122)
>>> > > > (entered lock:
>>> > > >
>>> org/apache/hadoop/hbase/zookeeper/RootRegionTracker@0x00000000E01A02B0
>>> > ,
>>> > > > entry count: 1)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/zookeeper/RootRegionTracker.waitRootRegionLocation(RootRegionTracker.java:73)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegion(HConnectionManager.java:786)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegion(HConnectionManager.java:766)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegionInMeta(HConnectionManager.java:895)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegion(HConnectionManager.java:797)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegion(HConnectionManager.java:766)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegionInMeta(HConnectionManager.java:895)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegion(HConnectionManager.java:801)
>>> > > > at
>>> > > >
>>> > >
>>> >
>>> org/apache/hadoop/hbase/client/HConnectionManager$HConnectionImplementation.locateRegion(HConnectionManager.java:766)
>>> > > > at org/apache/hadoop/hbase/client/HTable.<init>(HTable.java:189)
>>> > > > at org/apache/hadoop/hbase/client/HTable.<init>(HTable.java:163)
>>> > > > at com/yahoo/ycsb/db/HBaseClient.getHTable(HBaseClient.java:111)
>>> > > > (entered lock: java/lang/Object@0x00000000E006AD20, entry count:
>>> 1)
>>> > > > at com/yahoo/ycsb/db/HBaseClient.update(HBaseClient.java:295)
>>> > > > at com/yahoo/ycsb/db/HBaseClient.insert(HBaseClient.java:350)
>>> > > > at com/yahoo/ycsb/DBWrapper.insert(DBWrapper.java:145)
>>> > > > at
>>> > com/yahoo/ycsb/workloads/CoreWorkload.doInsert(CoreWorkload.java:461)
>>> > > > at com/yahoo/ycsb/ClientThread.run(Client.java:269)
>>> > > >
>>> > > >
>>> > > > however looking at the RS log, the meta region is up successfully,
>>> > hbase
>>> > > > shell works fine. I can also see it in zk cli:
>>> > > >
>>> > > > [zk: localhost:2181(CONNECTED) 2] get /hbase/meta-region-server
>>> > > > ? regionserver:60020??|~md"PBUF
>>> > > >
>>> > > >  ihc028 ??  ?????)
>>> > > > cZxid = 0xc57
>>> > > > ctime = Tue Oct 21 16:23:41 CST 2014
>>> > > > mZxid = 0xc57
>>> > > > mtime = Tue Oct 21 16:23:41 CST 2014
>>> > > > pZxid = 0xc57
>>> > > > cversion = 0
>>> > > > dataVersion = 0
>>> > > > aclVersion = 0
>>> > > > ephemeralOwner = 0x0
>>> > > > dataLength = 71
>>> > > > numChildren = 0
>>> > > >
>>> > > >
>>> > > > is it a configure issue? anything wrong with my steps?
>>> > > > 1. on hbase master node, download ycsb-0.1.4.tar.gz
>>> > > > 2. copy all jars in hbase lib dir to hbase-binding/lib
>>> > > > 3. copy hbase-site.xml to hbase-binding/conf
>>> > > > 4. create table in hbase shell: create 'usertable','f1','f2','f3'
>>> > > > 5. load data via above command.
>>> > > >
>>> > > > thanks.
>>> > >
>>> >
>>>
>>
>>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message