asterixdb-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From mingda li <limingda1...@gmail.com>
Subject Re: Using NCService to build AsterixDB on cluster
Date Sat, 24 Sep 2016 02:43:29 GMT
Good news!!!
Thanks Michael, Ian and Till.
I talked with Ian and find the reason and solve it.
It can work now with the cc.conf modified according to Michael's suggestion.
When I open the http://cc:19002/admin/cluster, it returns
{"State":"ACTIVE","Metadata_Node":"red4","partition_0":"red4","partition_1":"red3","partition_2":"red2","partition_3":"red"}.
SO it can work for all the nodes!

Thanks again:-)



On Fri, Sep 23, 2016 at 6:51 PM, Till Westmann <tillw@apache.org> wrote:

> Hi Mingda,
> have you started the NCService processes manually on the cluster nodes?
> Maybe you only have an NCService running on one node?
> Cheers,
> Till
> On 23 Sep 2016, at 18:18, mingda li wrote:
>
> I have tried to modify the cc.conf to another form as following. But when
> I run all NCs and the CC. It actually started only the nc/red and cc.
> Has anyone met the problem before while using NCService to configure
> AsterixDB?
>
> [nc4/red4]
>
> txnlogdir=/tmp/asterix/red4/txnlog
>
> coredumpdir=/tmp/asterix/red4/coredump
>
> iodevices=/tmp/asterix/red4
>
> address=131.179.64.23
>
> app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>
> storagedir=storage
>
> command=asterixnc
>
> [nc3/red3]
>
> txnlogdir=/tmp/asterix/red3/txnlog
>
> coredumpdir=/tmp/asterix/red3/coredump
>
> iodevices=/tmp/asterix/red3
>
> address=131.179.64.22
>
> app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>
> storagedir=storage
>
> command=asterixnc
>
> [nc2/red2]
>
> txnlogdir=/tmp/asterix/red2/txnlog
>
> coredumpdir=/tmp/asterix/red2/coredump
>
> iodevices=/tmp/asterix/red2
>
> address=131.179.64.21
>
> app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>
> storagedir=storage
>
> command=asterixnc
>
> [nc/red]
>
> txnlogdir=/tmp/asterix/red/txnlog
>
> coredumpdir=/tmp/asterix/red/coredump
>
> iodevices=/tmp/asterix/red
>
> address=131.179.64.20
>
>
> [nc]
>
> app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>
> storagedir=storage
>
> command=asterixnc
>
>
> [cc]
>
> cluster.address = 131.179.64.20
>
> http.port = 12345
>
> On Fri, Sep 23, 2016 at 10:26 AM, mingda li <limingda1993@gmail.com>
> wrote:
>
>> Dear Blow,
>>
>> Thanks, I have tried to modify the cc.conf file according to your advice
>> to the following. But when I open the http://masterNode:19002/ad
>> min/cluster, I found that it returns {"State":"UNUSABLE","M
>> etadata_Node":"red4","partition_3":"red"}.
>> And the asterixDB on  http://masterNode:19001 cannot work. Could you
>> tell me if I have make another mistake?
>>
>> Thanks,
>> Mingda
>>
>> *cc.conf:*
>>
>> *[nc/red4]*
>>
>> *txnlogdir=/tmp/asterix/red4/txnlog*
>>
>> *coredumpdir=/tmp/asterix/red4/coredump*
>>
>> *iodevices=/tmp/asterix/red4*
>>
>> *address=131.179.64.23*
>>
>> *[nc/red3]*
>>
>> *txnlogdir=/tmp/asterix/red3/txnlog*
>>
>> *coredumpdir=/tmp/asterix/red3/coredump*
>>
>> *iodevices=/tmp/asterix/red3*
>>
>> *address=131.179.64.22*
>>
>> *[nc/red2]*
>>
>> *txnlogdir=/tmp/asterix/red2/txnlog*
>>
>> *coredumpdir=/tmp/asterix/red2/coredump*
>>
>> *iodevices=/tmp/asterix/red2*
>>
>> *address=131.179.64.21*
>>
>> *[nc/red]*
>>
>> *txnlogdir=/tmp/asterix/red/txnlog*
>>
>> *coredumpdir=/tmp/asterix/red/coredump*
>>
>> *iodevices=/tmp/asterix/red*
>>
>> *address=131.179.64.20*
>>
>>
>> *[nc]*
>>
>> *app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint*
>>
>> *storagedir=storage*
>>
>> *command=asterixnc*
>>
>>
>> *[cc]*
>>
>> *cluster.address = 131.179.64.20*
>>
>> *http.port = 12345*
>>
>>
>> On Thu, Sep 22, 2016 at 8:17 PM, Michael Blow <mblow.apache@gmail.com>
>> wrote:
>>
>>> There is just nc and nc/<node> sections, you should not have nc1, nc2,
>>> etc. also, each no must be unique name, you can not name them all 'red'
>>>
>>> Hope it helps,
>>>
>>> Thanks,
>>>
>>> -MDB
>>>
>>> On Thu, Sep 22, 2016 at 8:33 PM mingda li <limingda1993@gmail.com>
>>> wrote:
>>>
>>>> Dear all,
>>>>
>>>> These days, I am building the AsterixDB on cluster following the
>>>> instruction here:
>>>> https://ci.apache.org/projects/asterixdb/ncservice.html
>>>>
>>>> There are one master node with one nc and three cluster nodes with one
>>>> nc each machine. The cc.conf is as following. And after I use
>>>> $BASEDIR/bin/asterixncservice >red-service.log 2>&1 & to start
three
>>>> cluster nodes' AsterixDB and the master node's AsterixDB,   I run $BASEDIR/bin/asterixcc
>>>> -config-file cc.conf > cc.log 2>&1 & to start the CC.  There
is no
>>>> error message and I can open the http://master:19001 and run aql. But
>>>> when I open the http://master:19002/admin/cluster, there is only one
>>>> nc here ({"State":"ACTIVE","Metadata_Node":"red","partition_0":"red"}).
>>>> I couldn't find the ncs on other machines. Does anyone know why?
>>>>
>>>> Thanks,
>>>> Mingda
>>>>
>>>> cc.conf file is as following:
>>>> [nc4/red] txnlogdir=/tmp/asterix/red/txnlog
>>>> coredumpdir=/tmp/asterix/red/coredump iodevices=/tmp/asterix/red
>>>> [nc4] app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>>>> storagedir=storage address=131.179.64.23 command=asterixnc
>>>> [nc3/red] txnlogdir=/tmp/asterix/red/txnlog
>>>> coredumpdir=/tmp/asterix/red/coredump iodevices=/tmp/asterix/red
>>>> [nc3] app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>>>> storagedir=storage address=131.179.64.22 command=asterixnc
>>>> [nc2/red] txnlogdir=/tmp/asterix/red/txnlog
>>>> coredumpdir=/tmp/asterix/red/coredump iodevices=/tmp/asterix/red
>>>> [nc2] app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>>>> storagedir=storage address=131.179.64.21 command=asterixnc
>>>> [nc/red] txnlogdir=/tmp/asterix/red/txnlog
>>>> coredumpdir=/tmp/asterix/red/coredump iodevices=/tmp/asterix/red
>>>> [nc] app.class=org.apache.asterix.hyracks.bootstrap.NCApplicationEntryPoint
>>>> storagedir=storage address=131.179.64.20 command=asterixnc
>>>> [cc] cluster.address = 131.179.64.20 http.port = 12345
>>>>
>>>>
>>>>
>>
>

Mime
View raw message