bigdata-developers

[Bigdata-developers] Quorum Problems

[Maximilian B. <br...@su...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Dear All,

I'm very new to blazegraph but I changed to blazegraph cause of the
clustering possibilities. I'm poor of having trouble with this. I want
to cluster three nodes, all three become members in the quorum but only
two join them.
I discover this with zooinspector.

The only difference between the three servers is that one of them
doesn't have DNS is that a problem? Maybe on you can help me.


My config file looks like this (the same on every machine):

## Configure basic environment variables.  Obviously, you must use your
own parameters for LOCATORS and ZK_SERVERS.

## This will not override parameters in the environment.


# Name of the federation of services (controls the Apache River GROUPS).


if [ -z "${FEDNAME}" ]; then

        export FEDNAME=tgRDFCluster

fi


# Path for local storage for this federation of services.


if [ -z "${FED_DIR}" ]; then

        export FED_DIR=/home/tomcat-sesame/blazegraphCluster/data

fi


# Name of the replication cluster to which this HAJournalServer will belong.


if [ -z "${LOGICAL_SERVICE_ID}" ]; then

        export LOGICAL_SERVICE_ID=tgHA-1

fi


# Where to find the Apache River service registrars (can also use
multicast).


if [ -z "${LOCATORS}" ]; then

        #Use for a HA1+ configuration

        #export LOCATORS="jini://localhost/"

        #HA3 example

        export
LOCATORS="jini://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/"

fi


# Where to find the Apache Zookeeper ensemble.


if [ -z "${ZK_SERVERS}" ] ; then

        #Use for single node configuration

        export ZK_SERVERS="localhost:2181"

        #Use for a multiple ZK configuration

        #export ZK_SERVERS="bigdata15:2081,bigdata16:2081,bigdata17:2081"

fi


#Replication Factor (set to one for HA1) configuration


if [ -z "${REPLICATION_FACTOR}" ] ; then

        #Use for a HA1 configuration

        export REPLICATION_FACTOR=3

        #Use for a HA1+ configuration

        #export REPLICATION_FACTOR=3

fi


#Port for the NanoSparqlServer Jetty


if [ -z "${JETTY_PORT}" ] ; then

        export JETTY_PORT=7070

fi


#Group commit (true|false)


if [ -z "${GROUP_COMMIT}" ] ; then

        export GROUP_COMMIT=true


- -- 
Maximilian Brodhun

Abteilung Forschung und Entwicklung
Georg-August-Universität Göttingen
Niedersächsische Staats- und Universitätsbibliothek Göttingen
D-37070 Göttingen

Papendiek 14 (Historisches Gebäude, Raum 2.409)
+49 551 39-4923 (Tel.)

br...@su...

http://www.sub.uni-goettingende/
http://www.rdd.sub.uni-goettingen.de/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
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=7LK/
-----END PGP SIGNATURE-----

Re: [Bigdata-developers] Quorum Problems

[Bryan T. <br...@sy...>]

This could very easily be DNS. Also, java can have long timeouts (30-60
seconds) if reverse DNS is not properly configured.

You can use http://localhost:port/bigdata/status to see the detailed status
(including zookeeper).   This information is also available under the
"status" tab of the workbench.

Thanks,
Bryan

----
Bryan Thompson
Chief Scientist & Founder
SYSTAP, LLC
4501 Tower Road
Greensboro, NC 27410
br...@sy...
http://blazegraph.com
http://blog.bigdata.com <http://bigdata.com>
http://mapgraph.io

Blazegraph™ <http://www.blazegraph.com/> is our ultra high-performance
graph database that supports both RDF/SPARQL and Tinkerpop/Blueprints
APIs.  MapGraph™ <http://www.systap.com/mapgraph> is our disruptive new
technology to use GPUs to accelerate data-parallel graph analytics.

CONFIDENTIALITY NOTICE:  This email and its contents and attachments are
for the sole use of the intended recipient(s) and are confidential or
proprietary to SYSTAP. Any unauthorized review, use, disclosure,
dissemination or copying of this email or its contents or attachments is
prohibited. If you have received this communication in error, please notify
the sender by reply email and permanently delete all copies of the email
and its contents and attachments.

On Tue, Mar 24, 2015 at 9:54 AM, Maximilian Brodhun <
br...@su...> wrote:

>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Dear All,
>
> I'm very new to blazegraph but I changed to blazegraph cause of the
> clustering possibilities. I'm poor of having trouble with this. I want
> to cluster three nodes, all three become members in the quorum but only
> two join them.
> I discover this with zooinspector.
>
> The only difference between the three servers is that one of them
> doesn't have DNS is that a problem? Maybe on you can help me.
>
>
> My config file looks like this (the same on every machine):
>
> ## Configure basic environment variables.  Obviously, you must use your
> own parameters for LOCATORS and ZK_SERVERS.
>
> ## This will not override parameters in the environment.
>
>
> # Name of the federation of services (controls the Apache River GROUPS).
>
>
> if [ -z "${FEDNAME}" ]; then
>
>         export FEDNAME=tgRDFCluster
>
> fi
>
>
> # Path for local storage for this federation of services.
>
>
> if [ -z "${FED_DIR}" ]; then
>
>         export FED_DIR=/home/tomcat-sesame/blazegraphCluster/data
>
> fi
>
>
> # Name of the replication cluster to which this HAJournalServer will
> belong.
>
>
> if [ -z "${LOGICAL_SERVICE_ID}" ]; then
>
>         export LOGICAL_SERVICE_ID=tgHA-1
>
> fi
>
>
> # Where to find the Apache River service registrars (can also use
> multicast).
>
>
> if [ -z "${LOCATORS}" ]; then
>
>         #Use for a HA1+ configuration
>
>         #export LOCATORS="jini://localhost/"
>
>         #HA3 example
>
>         export
> LOCATORS="jini://
> textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/
> "
>
> fi
>
>
> # Where to find the Apache Zookeeper ensemble.
>
>
> if [ -z "${ZK_SERVERS}" ] ; then
>
>         #Use for single node configuration
>
>         export ZK_SERVERS="localhost:2181"
>
>         #Use for a multiple ZK configuration
>
>         #export ZK_SERVERS="bigdata15:2081,bigdata16:2081,bigdata17:2081"
>
> fi
>
>
> #Replication Factor (set to one for HA1) configuration
>
>
> if [ -z "${REPLICATION_FACTOR}" ] ; then
>
>         #Use for a HA1 configuration
>
>         export REPLICATION_FACTOR=3
>
>         #Use for a HA1+ configuration
>
>         #export REPLICATION_FACTOR=3
>
> fi
>
>
> #Port for the NanoSparqlServer Jetty
>
>
> if [ -z "${JETTY_PORT}" ] ; then
>
>         export JETTY_PORT=7070
>
> fi
>
>
> #Group commit (true|false)
>
>
> if [ -z "${GROUP_COMMIT}" ] ; then
>
>         export GROUP_COMMIT=true
>
>
> - --
> Maximilian Brodhun
>
> Abteilung Forschung und Entwicklung
> Georg-August-Universität Göttingen
> Niedersächsische Staats- und Universitätsbibliothek Göttingen
> D-37070 Göttingen
>
> Papendiek 14 (Historisches Gebäude, Raum 2.409)
> +49 551 39-4923 (Tel.)
>
> br...@su...
>
> http://www.sub.uni-goettingende/
> http://www.rdd.sub.uni-goettingen.de/
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1
>
> iQIcBAEBAgAGBQJVEWyPAAoJEBDMOSiH8mYucJMP/0PgYrjY1It0HQ6H65S8Qi0P
> /2mit6rDD7saQwVm0+f29ZVtteE3mWgGHCgsycthJvuVKiM5rJBvsa7cZGfA0DTy
> SY7wS6sx/87J4U38dY7qqPbl846eOHvyIIfxU/IkK7p/UEyJd39V3Nu+TYGnEmzS
> beEX6CijpORpbvivNeCv+Lgcb2yrLB5/AwTdURmH65v/nP7oopgMMz+yazznw+vh
> UOtYtjMT/msgzl8pAr76W/wAvsiihk6DhzdAgBPCwAriMD3JUuAawXvuWNqRdYfZ
> bGCgzdMSMCtGn22St7G2F2BqKLzhW6kY50LV/A0M4DfFzvJ5bZdARYsj1IDwkRL1
> KKv88tYDdN+w+i1H84a20PN5wI9AVsZCbvMFELfrc2wUsdOpXalbiFY6uagwzkzs
> bolgHicuw7o63QvrhBMl2xPf/y7qCkMVjcGPClwhcUzHX3a8s6degjPptga/CZMg
> 0YnUSK5vSSqPquLanfExza4BbrG7LhgmzKPGkPkoqhgURhQaBCZA+A6tw+8NSs9P
> PYrkKwqEYNhug1LIif2BhUlOSwNhshPLmhn0/jt8HWIsnqDckFy4xhaLQYo3tG6+
> bUzLSQCc9z2fdV889Bwn87rHV/II5nk3aJKk1NgTu7C+6txpFj9GtKk1wGkDmf41
> 8TNMmsIgHLA8OQMXd++E
> =7LK/
> -----END PGP SIGNATURE-----
>
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website,
> sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for
> all
> things parallel software development, from weekly thought leadership blogs
> to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> _______________________________________________
> Bigdata-developers mailing list
> Big...@li...
> https://lists.sourceforge.net/lists/listinfo/bigdata-developers
>

Re: [Bigdata-developers] Quorum Problems

[Maximilian B. <br...@su...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thanks for the quick answer.
I'm wondering cause I put all hosts in etc/hosts and I can ping all
servers by using the name not the IP.

I give all servers a higher timeout time but the problem still occurs.
The status tab from blazegraph gives the following response: (maybe an
RMI Problem?)


    Quorum Services

  * http://myhostname.de:7073/bigdata
<http://textgrid-blazequorum.gwdg.de:7073/bigdata> : is not joined,
pipelineOrder=0, writePipelineAddr=/10.254.1.6:3090, service=self,
extendedRunState={server=Running, quorumService=SeekConsensus @ 0,
haReady=-1, haStatus=NotReady,
serviceId=791f97d3-6c12-4470-ac61-97d03c0cd43b, now=1427274241853}
  * Unable to reach service:
Proxy[HAGlue,BasicInvocationHandler[BasicObjectEndpoint[704b0793-0285-4a26-ae9a-904a3fc3b5ee,TcpEndpoint[127.0.1.1:3080]]]]
  * Unable to reach service:
Proxy[HAGlue,BasicInvocationHandler[BasicObjectEndpoint[f0c1c096-03d8-4d49-8093-30b50d605d8b,TcpEndpoint[127.0.1.1:3080]]]]


    Zookeeper

tgHA-1(1 children)
    quorum(4 children)
com.bigdata.quorum.zk.QuorumTokenState{lastValidToken=67,currentToken=-1,replicationFactor=3}
        joined(2 children)
            joined0000000300 (Ephemeral165590815907643397)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
            joined0000000301 (Ephemeral93533224253980679)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
        member(3 children)
            member71ef114a-b872-470a-ac9b-0ff632aa0b59
(Ephemeral93533224253980679)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
            member791f97d3-6c12-4470-ac61-97d03c0cd43b
(Ephemeral237648408135794694)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=791f97d3-6c12-4470-ac61-97d03c0cd43b}
            member8deaf15c-776d-48d9-84d5-2157c56dbe48
(Ephemeral165590815907643397)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
        pipeline(3 children)
            pipeline0000000401 (Ephemeral237648408135794694)
com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=791f97d3-6c12-4470-ac61-97d03c0cd43b,addrSelf=/10.254.1.6:3090}
            pipeline0000000403 (Ephemeral165590815907643397)
com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48,addrSelf=/10.254.1.2:3090}
            pipeline0000000404 (Ephemeral93533224253980679)
com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59,addrSelf=/10.254.1.5:3090}
        votes(1 children)
            0(2 children)
                 vote0000000000 (Ephemeral165590815907643397)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
                vote0000000001 (Ephemeral93533224253980679)
com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}



Am 24.03.2015 um 14:58 schrieb Bryan Thompson:
> This could very easily be DNS. Also, java can have long timeouts (30-60 seconds) if reverse DNS is not
properly configured.
>
> You can use http://localhost:port/bigdata/status to see the detailed
status (including zookeeper).   This information is also available under
the "status" tab of the workbench.
>
> Thanks,
> Bryan
>
> ----
> Bryan Thompson
> Chief Scientist & Founder
> SYSTAP, LLC
> 4501 Tower Road
> Greensboro, NC 27410
> br...@sy... <mailto:br...@sy...>
> http://blazegraph.com
> http://blog.bigdata.com <http://bigdata.com>
> http://mapgraph.io
>
> Blazegraph™ <http://www.blazegraph.com/> is our ultra high-performance
graph database that supports both RDF/SPARQL and Tinkerpop/Blueprints
APIs.  MapGraph™ <http://www.systap.com/mapgraph> is our disruptive new
technology to use GPUs to accelerate data-parallel graph analytics.
>
> CONFIDENTIALITY NOTICE:  This email and its contents and attachments
are for the sole use of the intended recipient(s) and are confidential
or proprietary to SYSTAP. Any unauthorized review, use, disclosure,
dissemination or copying of this email or its contents or attachments is
prohibited. If you have received this communication in error, please
notify the sender by reply email and permanently delete all copies of
the email and its contents and attachments.

>
>
> On Tue, Mar 24, 2015 at 9:54 AM, Maximilian Brodhun
<br...@su... <mailto:br...@su...>>
wrote:
>
>
> Dear All,
>
> I'm very new to blazegraph but I changed to blazegraph cause of the
> clustering possibilities. I'm poor of having trouble with this. I want
> to cluster three nodes, all three become members in the quorum but only
> two join them.
> I discover this with zooinspector.
>
> The only difference between the three servers is that one of them
> doesn't have DNS is that a problem? Maybe on you can help me.
>
>
> My config file looks like this (the same on every machine):
>
> ## Configure basic environment variables.  Obviously, you must use your
> own parameters for LOCATORS and ZK_SERVERS.
>
> ## This will not override parameters in the environment.
>
>
> # Name of the federation of services (controls the Apache River GROUPS).
>
>
> if [ -z "${FEDNAME}" ]; then
>
>         export FEDNAME=tgRDFCluster
>
> fi
>
>
> # Path for local storage for this federation of services.
>
>
> if [ -z "${FED_DIR}" ]; then
>
>         export FED_DIR=/home/tomcat-sesame/blazegraphCluster/data
>
> fi
>
>
> # Name of the replication cluster to which this HAJournalServer will
belong.
>
>
> if [ -z "${LOGICAL_SERVICE_ID}" ]; then
>
>         export LOGICAL_SERVICE_ID=tgHA-1
>
> fi
>
>
> # Where to find the Apache River service registrars (can also use
> multicast).
>
>
> if [ -z "${LOCATORS}" ]; then
>
>         #Use for a HA1+ configuration
>
>         #export LOCATORS="jini://localhost/"
>
>         #HA3 example
>
>         export
>
LOCATORS="jini://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/
<http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>"
>
> fi
>
>
> # Where to find the Apache Zookeeper ensemble.
>
>
> if [ -z "${ZK_SERVERS}" ] ; then
>
>         #Use for single node configuration
>
>         export ZK_SERVERS="localhost:2181"
>
>         #Use for a multiple ZK configuration
>
>         #export ZK_SERVERS="bigdata15:2081,bigdata16:2081,bigdata17:2081"
>
> fi
>
>
> #Replication Factor (set to one for HA1) configuration
>
>
> if [ -z "${REPLICATION_FACTOR}" ] ; then
>
>         #Use for a HA1 configuration
>
>         export REPLICATION_FACTOR=3
>
>         #Use for a HA1+ configuration
>
>         #export REPLICATION_FACTOR=3
>
> fi
>
>
> #Port for the NanoSparqlServer Jetty
>
>
> if [ -z "${JETTY_PORT}" ] ; then
>
>         export JETTY_PORT=7070
>
> fi
>
>
> #Group commit (true|false)
>
>
> if [ -z "${GROUP_COMMIT}" ] ; then
>
>         export GROUP_COMMIT=true
>
>
>
>
>    
------------------------------------------------------------------------------
>     Dive into the World of Parallel Programming The Go Parallel
Website, sponsored
>     by Intel and developed in partnership with Slashdot Media, is your
hub for all
>     things parallel software development, from weekly thought
leadership blogs to
>     news, videos, case studies, tutorials and more. Take a look and
join the
>     conversation now. http://goparallel.sourceforge.net/
>     _______________________________________________
>     Bigdata-developers mailing list
>     Big...@li...
<mailto:Big...@li...>
>     https://lists.sourceforge.net/lists/listinfo/bigdata-developers
>
>

- -- 
Maximilian Brodhun

Abteilung Forschung und Entwicklung
Georg-August-Universität Göttingen
Niedersächsische Staats- und Universitätsbibliothek Göttingen
D-37070 Göttingen

Papendiek 14 (Historisches Gebäude, Raum 2.409)
+49 551 39-4923 (Tel.)

br...@su...

http://www.sub.uni-goettingende/
http://www.rdd.sub.uni-goettingen.de/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iQIcBAEBAgAGBQJVEn0FAAoJEBDMOSiH8mYu01EP/ibqT/5sn3dcDHRQ0upDEq4r
dJh6IYUkQnCDJJ36lgozVSlEEKHAnLgMW8HcFWwA4ox74v/NJtvnvMz+mjDuExQ1
xZbMQHGW19pXSJoiI2rWbi/j7cwdU46EUlAnAxf96UN/P4Srg3OPcqrvGo9Y6YC9
5Z+WmLMBZ+kVG9++Vhe0JCoqe6L8NkjVk/wvOwM5Qdh4Et99HLUA0tFBwsK+8cQK
EBrkjDsGmsvtJQGNVygN4EWDP0GfQPXU7XzPExOhKO/mPzFTRuyd0FAzXlSWWnqR
TA77bM9rqBCo3jmWyFme84gpggVeUWMNFyLEdJDKYIU3JU/wcf+5JxHqD5xMGSQ6
Y2e4c9MNEkkernT8XedUdcttnNBvZqN4M5AILNr19uD5zz+RKpgsg+fNRvAhXvz2
HFevFWG7+5M87b781z6gjcVvyPSsIOvNKdNljzWYSjOwO5f1hAEz0f2LuYJF9vzf
HaCQ7Gt1tCyGUn/UPteGhUOq6xWOqHZHAgwz0eGGMpxI/4vKNgPq8huObPXvJieV
gu3/RVupNsyNwyczeMDg6OSYyz791EEKfxkKhJ9rN3DpPbAdenCE4/7udpTR7R8c
AJjBttHQ2tXAFS1Tvp5ViM9lEFXT315yLhhWkHhtMt1RVygYPrVr8p3nZmpZ67iW
IwYYhOIpyrLoai1CY6Mm
=JnID
-----END PGP SIGNATURE-----

Re: [Bigdata-developers] Quorum Problems

[Bryan T. <br...@sy...>]

I see several different manners in which ip addresses are being expressed
(host name, 10.x.x.x network, 127.x.x,x network).  This would appear to be
the status for the one host that can not reach the others.  While it has
discovered zookeeper and obtained R I proxies for two other services (those
127.0.1.1 addresses look odd) the services are not responding at those
addresses.  This could be an ip configuration issue (I would try
consistently using the private network ip addresses thoughout) or a
firewall issue.

 Sometimes you can windup with bad proxy caches when bouncing the services
and making configuration changes.  If this happens the easiest thing to do
is bring down all 3 services, wait a minute or two for zookeeper to
definitively notice that the services are dead and expire the ephemeral
znodes.  Also, make sure that the services really are down using jps to
list java processes and use netstat to make sure that lingering ports have
been released.  The check the logs for each service on restart. Once you
have a proper configuration, these steps are not necessary but sometimes
they can help while attempting to coverage on a correct is installation.

On Wednesday, March 25, 2015, Maximilian Brodhun <
br...@su...> wrote:

>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Thanks for the quick answer.
> I'm wondering cause I put all hosts in etc/hosts and I can ping all
> servers by using the name not the IP.
>
> I give all servers a higher timeout time but the problem still occurs. The
> status tab from blazegraph gives the following response: (maybe an RMI
> Problem?)
>
>
>     Quorum Services
>
>   * http://myhostname.de:7073/bigdata
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata> : is not joined,
> pipelineOrder=0, writePipelineAddr=/10.254.1.6:3090, service=self,
> extendedRunState={server=Running, quorumService=SeekConsensus @ 0,
> haReady=-1, haStatus=NotReady,
> serviceId=791f97d3-6c12-4470-ac61-97d03c0cd43b, now=1427274241853}
>   * Unable to reach service:
> Proxy[HAGlue,BasicInvocationHandler[BasicObjectEndpoint[704b0793-0285-4a26-ae9a-904a3fc3b5ee,TcpEndpoint[127.0.1.1:3080
> ]]]]
>   * Unable to reach service:
> Proxy[HAGlue,BasicInvocationHandler[BasicObjectEndpoint[f0c1c096-03d8-4d49-8093-30b50d605d8b,TcpEndpoint[127.0.1.1:3080
> ]]]]
>
>
>     Zookeeper
>
> tgHA-1(1 children)
>     quorum(4 children)
> com.bigdata.quorum.zk.QuorumTokenState{lastValidToken=67,currentToken=-1,replicationFactor=3}
>         joined(2 children)
>             joined0000000300 (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
>             joined0000000301 (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
>         member(3 children)
>             member71ef114a-b872-470a-ac9b-0ff632aa0b59
> (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
>             member791f97d3-6c12-4470-ac61-97d03c0cd43b
> (Ephemeral237648408135794694)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=791f97d3-6c12-4470-ac61-97d03c0cd43b}
>             member8deaf15c-776d-48d9-84d5-2157c56dbe48
> (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
>         pipeline(3 children)
>             pipeline0000000401 (Ephemeral237648408135794694)
> com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=791f97d3-6c12-4470-ac61-97d03c0cd43b,addrSelf=/
> 10.254.1.6:3090}
>             pipeline0000000403 (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48,addrSelf=/
> 10.254.1.2:3090}
>             pipeline0000000404 (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59,addrSelf=/
> 10.254.1.5:3090}
>         votes(1 children)
>             0(2 children)
>                  vote0000000000 (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
>                 vote0000000001 (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
>
>
>
> Am 24.03.2015 um 14:58 schrieb Bryan Thompson:
> > This could very easily be DNS. Also, java can have long timeouts (30-60
> seconds) if reverse DNS is not properly configured.
> >
> > You can use http://localhost:port/bigdata/status to see the detailed
> status (including zookeeper).   This information is also available under
> the "status" tab of the workbench.
> >
> > Thanks,
> > Bryan
> >
> > ----
> > Bryan Thompson
> > Chief Scientist & Founder
> > SYSTAP, LLC
> > 4501 Tower Road
> > Greensboro, NC 27410
> > br...@sy... <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> > http://blazegraph.com
> > http://blog.bigdata.com <http://bigdata.com> <http://bigdata.com>
> > http://mapgraph.io
> >
> > Blazegraph™ <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> is our ultra high-performance graph database that supports both RDF/SPARQL
> and Tinkerpop/Blueprints APIs.  MapGraph™ <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> is our disruptive new technology to use
> GPUs to accelerate data-parallel graph analytics.
> >
> > CONFIDENTIALITY NOTICE:  This email and its contents and attachments are
> for the sole use of the intended recipient(s) and are confidential or
> proprietary to SYSTAP. Any unauthorized review, use, disclosure,
> dissemination or copying of this email or its contents or attachments is
> prohibited. If you have received this communication in error, please notify
> the sender by reply email and permanently delete all copies of the email
> and its contents and attachments.
> >
> >
> > On Tue, Mar 24, 2015 at 9:54 AM, Maximilian Brodhun <
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>> wrote:
> >
> >
> > Dear All,
> >
> > I'm very new to blazegraph but I changed to blazegraph cause of the
> > clustering possibilities. I'm poor of having trouble with this. I want
> > to cluster three nodes, all three become members in the quorum but only
> > two join them.
> > I discover this with zooinspector.
> >
> > The only difference between the three servers is that one of them
> > doesn't have DNS is that a problem? Maybe on you can help me.
> >
> >
> > My config file looks like this (the same on every machine):
> >
> > ## Configure basic environment variables.  Obviously, you must use your
> > own parameters for LOCATORS and ZK_SERVERS.
> >
> > ## This will not override parameters in the environment.
> >
> >
> > # Name of the federation of services (controls the Apache River GROUPS).
> >
> >
> > if [ -z "${FEDNAME}" ]; then
> >
> >         export FEDNAME=tgRDFCluster
> >
> > fi
> >
> >
> > # Path for local storage for this federation of services.
> >
> >
> > if [ -z "${FED_DIR}" ]; then
> >
> >         export FED_DIR=/home/tomcat-sesame/blazegraphCluster/data
> >
> > fi
> >
> >
> > # Name of the replication cluster to which this HAJournalServer will
> belong.
> >
> >
> > if [ -z "${LOGICAL_SERVICE_ID}" ]; then
> >
> >         export LOGICAL_SERVICE_ID=tgHA-1
> >
> > fi
> >
> >
> > # Where to find the Apache River service registrars (can also use
> > multicast).
> >
> >
> > if [ -z "${LOCATORS}" ]; then
> >
> >         #Use for a HA1+ configuration
> >
> >         #export LOCATORS="jini://localhost/"
> >
> >         #HA3 example
> >
> >         export
> > LOCATORS="jini://
> textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> "
> >
> > fi
> >
> >
> > # Where to find the Apache Zookeeper ensemble.
> >
> >
> > if [ -z "${ZK_SERVERS}" ] ; then
> >
> >         #Use for single node configuration
> >
> >         export ZK_SERVERS="localhost:2181"
> >
> >         #Use for a multiple ZK configuration
> >
> >         #export ZK_SERVERS="bigdata15:2081,bigdata16:2081,bigdata17:2081"
> >
> > fi
> >
> >
> > #Replication Factor (set to one for HA1) configuration
> >
> >
> > if [ -z "${REPLICATION_FACTOR}" ] ; then
> >
> >         #Use for a HA1 configuration
> >
> >         export REPLICATION_FACTOR=3
> >
> >         #Use for a HA1+ configuration
> >
> >         #export REPLICATION_FACTOR=3
> >
> > fi
> >
> >
> > #Port for the NanoSparqlServer Jetty
> >
> >
> > if [ -z "${JETTY_PORT}" ] ; then
> >
> >         export JETTY_PORT=7070
> >
> > fi
> >
> >
> > #Group commit (true|false)
> >
> >
> > if [ -z "${GROUP_COMMIT}" ] ; then
> >
> >         export GROUP_COMMIT=true
> >
> >
> >
> >
> >
> ------------------------------------------------------------------------------
> >     Dive into the World of Parallel Programming The Go Parallel Website,
> sponsored
> >     by Intel and developed in partnership with Slashdot Media, is your
> hub for all
> >     things parallel software development, from weekly thought leadership
> blogs to
> >     news, videos, case studies, tutorials and more. Take a look and join
> the
> >     conversation now. http://goparallel.sourceforge.net/
> >     _______________________________________________
> >     Bigdata-developers mailing list
> >     Big...@li...
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> >     https://lists.sourceforge.net/lists/listinfo/bigdata-developers
> >
> >
>
> - --
> Maximilian Brodhun
>
> Abteilung Forschung und Entwicklung
> Georg-August-Universität Göttingen
> Niedersächsische Staats- und Universitätsbibliothek Göttingen
> D-37070 Göttingen
>
> Papendiek 14 (Historisches Gebäude, Raum 2.409)
> +49 551 39-4923 (Tel.)
>
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
>
> http://www.sub.uni-goettingende/
> http://www.rdd.sub.uni-goettingen.de/
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1
>
> iQIcBAEBAgAGBQJVEn0FAAoJEBDMOSiH8mYu01EP/ibqT/5sn3dcDHRQ0upDEq4r
> dJh6IYUkQnCDJJ36lgozVSlEEKHAnLgMW8HcFWwA4ox74v/NJtvnvMz+mjDuExQ1
> xZbMQHGW19pXSJoiI2rWbi/j7cwdU46EUlAnAxf96UN/P4Srg3OPcqrvGo9Y6YC9
> 5Z+WmLMBZ+kVG9++Vhe0JCoqe6L8NkjVk/wvOwM5Qdh4Et99HLUA0tFBwsK+8cQK
> EBrkjDsGmsvtJQGNVygN4EWDP0GfQPXU7XzPExOhKO/mPzFTRuyd0FAzXlSWWnqR
> TA77bM9rqBCo3jmWyFme84gpggVeUWMNFyLEdJDKYIU3JU/wcf+5JxHqD5xMGSQ6
> Y2e4c9MNEkkernT8XedUdcttnNBvZqN4M5AILNr19uD5zz+RKpgsg+fNRvAhXvz2
> HFevFWG7+5M87b781z6gjcVvyPSsIOvNKdNljzWYSjOwO5f1hAEz0f2LuYJF9vzf
> HaCQ7Gt1tCyGUn/UPteGhUOq6xWOqHZHAgwz0eGGMpxI/4vKNgPq8huObPXvJieV
> gu3/RVupNsyNwyczeMDg6OSYyz791EEKfxkKhJ9rN3DpPbAdenCE4/7udpTR7R8c
> AJjBttHQ2tXAFS1Tvp5ViM9lEFXT315yLhhWkHhtMt1RVygYPrVr8p3nZmpZ67iW
> IwYYhOIpyrLoai1CY6Mm
> =JnID
> -----END PGP SIGNATURE-----
>
>

-- 
----
Bryan Thompson
Chief Scientist & Founder
SYSTAP, LLC
4501 Tower Road
Greensboro, NC 27410
br...@sy...
http://blazegraph.com
http://blog.bigdata.com <http://bigdata.com>
http://mapgraph.io

Blazegraph™ <http://www.blazegraph.com/> is our ultra high-performance
graph database that supports both RDF/SPARQL and Tinkerpop/Blueprints
APIs.  MapGraph™ <http://www.systap.com/mapgraph> is our disruptive new
technology to use GPUs to accelerate data-parallel graph analytics.

CONFIDENTIALITY NOTICE:  This email and its contents and attachments are
for the sole use of the intended recipient(s) and are confidential or
proprietary to SYSTAP. Any unauthorized review, use, disclosure,
dissemination or copying of this email or its contents or attachments is
prohibited. If you have received this communication in error, please notify
the sender by reply email and permanently delete all copies of the email
and its contents and attachments.

Re: [Bigdata-developers] Quorum Problems

[Bryan T. <br...@sy...>]

Sounds like you have been making progress.  If the leader is not ready, you
might have inconsistent data in the journal root blocks on the servers.
Try failing the leader and see if the other two meet.  If they do, then
redeploy the leader (remove the service directory, redeploy and restart the
service).j. In general, once you figure out the configuration of the
HAJournalServer, OS, and network I would do a from scratch deploy to make
sure that everything (deployment configuration and data) is the same on
each node.  If you have things setup correctly, the cluster should come
online smoothly.

The cluster is available for read and write as long as a majority (2 out of
3) of the nodes are available.  If the cluster goes below a majority then
you can not read or write on it.  This is because it can no longer be
decided which services have a consensus around the state of the database.
However, it is always possible to open the backing journal file in a
non-clustered mode. The file format is identical.

You can read all about the HA features at the HAJournalServer page of the
wiki.

http://wiki.bigdata.com/wiki/index.php/HAJournalServer

There is also a complete description of the architecture for standalone,
HA, and scale-out linked from the white papers section of our blog.

http://blog.bigdata.com/
http://www.blazegraph.com/whitepapers/bigdata_architecture_whitepaper.pdf

Thanks,
Bryan

On Thursday, March 26, 2015, Maximilian Brodhun <
br...@su...> wrote:

>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Dear Bryan,
>
> I'm lucky to say that now all nodes are joined the quorum. Thanks a lot
> for your help!
> I have a problem that the leader has the status "notReady", but I fill
> figure out why.
>
>
> A last question I have is, will it become possible to still have write
> access if one node will be down?
>
> Best regards,
>
> Max
>
>
> Am 25.03.2015 um 16:57 schrieb Bryan Thompson:
> >
> > Ok. I encourage you to revert the files and just use environment
> overrides. this is more consistent and systematic.  Shut down and remove
> the installs.  See if it comes up with the documented overrides. If not,
> send me the list of exported overrides and the status output.
> >
> > Make sure the services are shutdown. Use jps to list java processes.
> Then deploy and start with those overrides.
> >
> > You could also remove the deployments to get as close as possible to a
> blank slate.
> >
> > Check your firewalls.
> >
> > Bryan
> >
> > On Mar 25, 2015 11:44 AM, "Maximilian Brodhun" <
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>> wrote:
> >
> >
> > Oh, I'm sorry, this was an older configuration. In the actual (also not
> working configuration) the port 3080 wasn't put there.
> >
> > Thanks again for your quick answer.
> >
> > Max
> >
> > Am 25.03.2015 um 16:39 schrieb Bryan Thompson:
> >
> > > One of your join locators is wrong. It has port 3080.
> >
> > > You can achieve these overrides just by exporting the relevant
> variables as defined on the HAJournalServer page of the wiki.
> >
> > > Bryan
> >
> > > On Mar 25, 2015 11:32 AM, "Maximilian Brodhun" <
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>> wrote:
> >
> >
> > > Sorry for bothering you so much.
> >
> > > Thanks for your help, but I have still problems. The 127.0.1.1 looks
> also odd to me.
> > > Maybe it is helpful to tell you, what I have done so far.
> >
> > > I get the code for Version 1.5.1 by using git on one server. (Having
> no other instances of blazegraph running)
> > > After this I edit the config file
> src/resources/HAJournal/HAJournal.config regarding this lines (marked red,
> the others are like before) and copy it to the other server to having the
> same configuration at all servers.
> >
> >
> > > BIG DATA SECTION:
> >
> > >   * private static fedname =
> ConfigMath.getProperty("FEDNAME","tgRDFCluster");
> > >   * private static rmiPort =
> Integer.parseInt(ConfigMath.getProperty("RMI_PORT","3080"));
> > >   * private static haPort =
> Integer.parseInt(ConfigMath.getProperty("HA_PORT","3090"));
> > >   * private static replicationFactor =
> Integer.parseInt(ConfigMath.getProperty("REPLICATION_FACTOR","3"));
> > >   * private static logicalServiceId =
> ConfigMath.getProperty("LOGICAL_SERVICE_ID","tgHA-1");
> > >   * private static serviceId = null;
> > >   * private static fedDir = new
> File(ConfigMath.getProperty("FED_DIR","."),fedname);
> > >   * private static serviceDir = new
> File(fedDir,logicalServiceId+File.separator+"HAJournalServer");
> > >   * private static dataDir = new
> File(ConfigMath.getProperty("DATA_DIR",""+serviceDir));
> > >   * private static haLogDir = new
> File(ConfigMath.getProperty("HALOG_DIR",""+serviceDir+File.separator+"tgHALog"));
> > >   * private static snapshotDir = new
> File(ConfigMath.getProperty("SNAPSHOT_DIR",""+serviceDir+File.separator+"snapshot"));
> > >   * private static snapshotPolicy = new
> DefaultSnapshotPolicy(200/*hhmm*/,20/*percent*/);
> > >   * private static restorePolicy = new
> DefaultRestorePolicy(ConfigMath.d2ms(7));
> > >   * static private groups =
> ConfigMath.getGroups(ConfigMath.getProperty("GROUPS",bigdata.fedname));
> > >   * static private locators =
> ConfigMath.getLocators(ConfigMath.getProperty("LOCATORS","jini://hostname1/,jini://hostname2/,jini://hostname3/"));
> > >   * static private leaseTimeout = ConfigMath.s2ms(100); // 20
> > >   * static private sessionTimeout = (int)ConfigMath.s2ms(100); // 5
> > >   * private static namespace = "tg";
> >
> >
> > > ZOOKEEPER SECTION:
> >
> > >   * servers =
> ConfigMath.getProperty("ZK_SERVERS","hostname1:2081,hostname2:2081,hostname2:2081");
> >
> > > Doing the ant deploy-artifact on all server and edit the script in
> dist/bigdata/bin/config.sh like I send you before:
> >
> > >   * export FEDNAME=tgRDFCluster
> > >   * export FED_DIR=/home/"username"/blazegraphCluster/data
> > >   * export LOGICAL_SERVICE_ID=tgHA-1
> > >   * export
> LOCATORS="jini:/hostname1/,jini://hostname2/,jini://hostname3:3080/"
> > >   * export ZK_SERVERS="localhost:2181"
> > >   * export REPLICATION_FACTOR=3
> > >   * export JETTY_PORT=7072
> > >   * export GROUP_COMMIT=true
> >
> >
> > > For having all hosts knows each other and make it possible to
> communicate I free the ports in the firewall and edit all etc/hosts files:
> >
> > >   * 127.0.0.1       localhost
> > >   * 127.0.1.1       hostname.domain  shortname (without domain)
> > >   *
> > >   * # The following lines are desirable for IPv6 capable hosts
> > >   * ::1     localhost ip6-localhost ip6-loopback
> > >   * ff02::1 ip6-allnodes
> > >   * ff02::2 ip6-allrouters
> >
> >
> > > Hope you aren't bugged about this mail.
> >
> > > best regards,
> >
> > > Maximilian
> >
> > > Am 25.03.2015 um 12:01 schrieb Bryan Thompson:
> > > > I see several different manners in which ip addresses are being
> expressed (host name, 10.x.x.x network, 127.x.x,x network).  This would
> appear to be the status for the one host that can not reach the others.
> While it has discovered zookeeper and obtained R I proxies for two other
> services (those 127.0.1.1 addresses look odd) the services are not
> responding at those addresses.  This could be an ip configuration issue (I
> would try consistently using the private network ip addresses thoughout) or
> a firewall issue.
> >
> > > >  Sometimes you can windup with bad proxy caches when bouncing the
> services and making configuration changes.  If this happens the easiest
> thing to do is bring down all 3 services, wait a minute or two for
> zookeeper to definitively notice that the services are dead and expire the
> ephemeral znodes.  Also, make sure that the services really are down using
> jps to list java processes and use netstat to make sure that lingering
> ports have been released.  The check the logs for each service on restart.
> Once you have a proper configuration, these steps are not necessary but
> sometimes they can help while attempting to coverage on a correct is
> installation.
> >
> > > > On Wednesday, March 25, 2015, Maximilian Brodhun <
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>> wrote:
> >
> >
> > > > Thanks for the quick answer.
> > > > I'm wondering cause I put all hosts in etc/hosts and I can ping all
> servers by using the name not the IP.
> >
> > > > I give all servers a higher timeout time but the problem still
> occurs. The status tab from blazegraph gives the following response: (maybe
> an RMI Problem?)
> >
> >
> > > >     Quorum Services
> >
> > > >   * http://myhostname.de:7073/bigdata
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata>
> <http://textgrid-blazequorum.gwdg.de:7073/bigdata> : is not joined,
> pipelineOrder=0, writePipelineAddr=/10.254.1.6:3090
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090>, service=self,
> extendedRunState={server=Running, quorumService=SeekConsensus @ 0,
> haReady=-1, haStatus=NotReady,
> serviceId=791f97d3-6c12-4470-ac61-97d03c0cd43b, now=1427274241853}
> > > >   * Unable to reach service:
> Proxy[HAGlue,BasicInvocationHandler[BasicObjectEndpoint[704b0793-0285-4a26-ae9a-904a3fc3b5ee,TcpEndpoint[127.0.1.1:3080
> ]]]]
> > > >   * Unable to reach service:
> Proxy[HAGlue,BasicInvocationHandler[BasicObjectEndpoint[f0c1c096-03d8-4d49-8093-30b50d605d8b,TcpEndpoint[127.0.1.1:3080
> ]]]]
> >
> >
> > > >     Zookeeper
> >
> > > > tgHA-1(1 children)
> > > >     quorum(4 children)
> com.bigdata.quorum.zk.QuorumTokenState{lastValidToken=67,currentToken=-1,replicationFactor=3}
> > > >         joined(2 children)
> > > >             joined0000000300 (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
> > > >             joined0000000301 (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
> > > >         member(3 children)
> > > >             member71ef114a-b872-470a-ac9b-0ff632aa0b59
> (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
> > > >             member791f97d3-6c12-4470-ac61-97d03c0cd43b
> (Ephemeral237648408135794694)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=791f97d3-6c12-4470-ac61-97d03c0cd43b}
> > > >             member8deaf15c-776d-48d9-84d5-2157c56dbe48
> (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
> > > >         pipeline(3 children)
> > > >             pipeline0000000401 (Ephemeral237648408135794694)
> com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=791f97d3-6c12-4470-ac61-97d03c0cd43b,addrSelf=/
> 10.254.1.6:3090 <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> <http://10.254.1.6:3090> <http://10.254.1.6:3090> <http://10.254.1.6:3090>
> }
> > > >             pipeline0000000403 (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48,addrSelf=/
> 10.254.1.2:3090 <http://10.254.1.2:3090> <http://10.254.1.2:3090>
> <http://10.254.1.2:3090> <http://10.254.1.2:3090> <http://10.254.1.2:3090>
> <http://10.254.1.2:3090> <http://10.254.1.2:3090> <http://10.254.1.2:3090>
> <http://10.254.1.2:3090> <http://10.254.1.2:3090> <http://10.254.1.2:3090>
> <http://10.254.1.2:3090> <http://10.254.1.2:3090> <http://10.254.1.2:3090>
> }
> > > >             pipeline0000000404 (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumPipelineState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59,addrSelf=/
> 10.254.1.5:3090 <http://10.254.1.5:3090> <http://10.254.1.5:3090>
> <http://10.254.1.5:3090> <http://10.254.1.5:3090> <http://10.254.1.5:3090>
> <http://10.254.1.5:3090> <http://10.254.1.5:3090> <http://10.254.1.5:3090>
> <http://10.254.1.5:3090> <http://10.254.1.5:3090> <http://10.254.1.5:3090>
> <http://10.254.1.5:3090> <http://10.254.1.5:3090> <http://10.254.1.5:3090>
> }
> > > >         votes(1 children)
> > > >             0(2 children)
> > > >                  vote0000000000 (Ephemeral165590815907643397)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=8deaf15c-776d-48d9-84d5-2157c56dbe48}
> > > >                 vote0000000001 (Ephemeral93533224253980679)
> com.bigdata.quorum.zk.QuorumServiceState{serviceUUID=71ef114a-b872-470a-ac9b-0ff632aa0b59}
> >
> >
> >
> > > > Am 24.03.2015 um 14:58 schrieb Bryan Thompson:
> > > > > This could very easily be DNS. Also, java can have long timeouts
> (30-60 seconds) if reverse DNS is not properly configured.
> >
> > > > > You can use http://localhost:port/bigdata/status to see the
> detailed status (including zookeeper).   This information is also available
> under the "status" tab of the workbench.
> >
> > > > > Thanks,
> > > > > Bryan
> >
> > > > > ----
> > > > > Bryan Thompson
> > > > > Chief Scientist & Founder
> > > > > SYSTAP, LLC
> > > > > 4501 Tower Road
> > > > > Greensboro, NC 27410
> > > > > br...@sy...
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> > > > > http://blazegraph.com
> > > > > http://blog.bigdata.com <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com>
> > > > > http://mapgraph.io
> >
> > > > > Blazegraph™ <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> is our ultra high-performance graph database
> that supports both RDF/SPARQL and Tinkerpop/Blueprints APIs.  MapGraph™
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph> is our
> disruptive new technology to use GPUs to accelerate data-parallel graph
> analytics.
> >
> > > > > CONFIDENTIALITY NOTICE:  This email and its contents and
> attachments are for the sole use of the intended recipient(s) and are
> confidential or proprietary to SYSTAP. Any unauthorized review, use,
> disclosure, dissemination or copying of this email or its contents or
> attachments is prohibited. If you have received this communication in
> error, please notify the sender by reply email and permanently delete all
> copies of the email and its contents and attachments.
> >
> >
> > > > > On Tue, Mar 24, 2015 at 9:54 AM, Maximilian Brodhun <
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <mailto:br...@su...>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
> <javascript:_e(%7B%7D,'cvml','br...@su...');>> wrote:
> >
> >
> > > > > Dear All,
> >
> > > > > I'm very new to blazegraph but I changed to blazegraph cause of the
> > > > > clustering possibilities. I'm poor of having trouble with this. I
> want
> > > > > to cluster three nodes, all three become members in the quorum but
> only
> > > > > two join them.
> > > > > I discover this with zooinspector.
> >
> > > > > The only difference between the three servers is that one of them
> > > > > doesn't have DNS is that a problem? Maybe on you can help me.
> >
> >
> > > > > My config file looks like this (the same on every machine):
> >
> > > > > ## Configure basic environment variables.  Obviously, you must use
> your
> > > > > own parameters for LOCATORS and ZK_SERVERS.
> >
> > > > > ## This will not override parameters in the environment.
> >
> >
> > > > > # Name of the federation of services (controls the Apache River
> GROUPS).
> >
> >
> > > > > if [ -z "${FEDNAME}" ]; then
> >
> > > > >         export FEDNAME=tgRDFCluster
> >
> > > > > fi
> >
> >
> > > > > # Path for local storage for this federation of services.
> >
> >
> > > > > if [ -z "${FED_DIR}" ]; then
> >
> > > > >         export FED_DIR=/home/tomcat-sesame/blazegraphCluster/data
> >
> > > > > fi
> >
> >
> > > > > # Name of the replication cluster to which this HAJournalServer
> will belong.
> >
> >
> > > > > if [ -z "${LOGICAL_SERVICE_ID}" ]; then
> >
> > > > >         export LOGICAL_SERVICE_ID=tgHA-1
> >
> > > > > fi
> >
> >
> > > > > # Where to find the Apache River service registrars (can also use
> > > > > multicast).
> >
> >
> > > > > if [ -z "${LOCATORS}" ]; then
> >
> > > > >         #Use for a HA1+ configuration
> >
> > > > >         #export LOCATORS="jini://localhost/"
> >
> > > > >         #HA3 example
> >
> > > > >         export
> > > > > LOCATORS="jini://
> textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> <http://textgrid-test1.gwdg.de/,jini://textgrid-test1.gwdg.de/,jini://141.5.102.206/>
> "
> >
> > > > > fi
> >
> >
> > > > > # Where to find the Apache Zookeeper ensemble.
> >
> >
> > > > > if [ -z "${ZK_SERVERS}" ] ; then
> >
> > > > >         #Use for single node configuration
> >
> > > > >         export ZK_SERVERS="localhost:2181"
> >
> > > > >         #Use for a multiple ZK configuration
> >
> > > > >         #export
> ZK_SERVERS="bigdata15:2081,bigdata16:2081,bigdata17:2081"
> >
> > > > > fi
> >
> >
> > > > > #Replication Factor (set to one for HA1) configuration
> >
> >
> > > > > if [ -z "${REPLICATION_FACTOR}" ] ; then
> >
> > > > >         #Use for a HA1 configuration
> >
> > > > >         export REPLICATION_FACTOR=3
> >
> > > > >         #Use for a HA1+ configuration
> >
> > > > >         #export REPLICATION_FACTOR=3
> >
> > > > > fi
> >
> >
> > > > > #Port for the NanoSparqlServer Jetty
> >
> >
> > > > > if [ -z "${JETTY_PORT}" ] ; then
> >
> > > > >         export JETTY_PORT=7070
> >
> > > > > fi
> >
> >
> > > > > #Group commit (true|false)
> >
> >
> > > > > if [ -z "${GROUP_COMMIT}" ] ; then
> >
> > > > >         export GROUP_COMMIT=true
> >
> >
> >
> >
> > > > >
> ------------------------------------------------------------------------------
> > > > >     Dive into the World of Parallel Programming The Go Parallel
> Website, sponsored
> > > > >     by Intel and developed in partnership with Slashdot Media, is
> your hub for all
> > > > >     things parallel software development, from weekly thought
> leadership blogs to
> > > > >     news, videos, case studies, tutorials and more. Take a look
> and join the
> > > > >     conversation now. http://goparallel.sourceforge.net/
> > > > >     _______________________________________________
> > > > >     Bigdata-developers mailing list
> > > > >     Big...@li...
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <mailto:Big...@li...>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> <javascript:_e(%7B%7D,'cvml','Big...@li...');>
> > > > >
> https://lists.sourceforge.net/lists/listinfo/bigdata-developers
> >
> >
> >
> >
> >
> >
> > > > --
> > > > ----
> > > > Bryan Thompson
> > > > Chief Scientist & Founder
> > > > SYSTAP, LLC
> > > > 4501 Tower Road
> > > > Greensboro, NC 27410
> > > > br...@sy... <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> <mailto:br...@sy...>
> <javascript:_e(%7B%7D,'cvml','br...@sy...');>
> > > > http://blazegraph.com
> > > > http://blog.bigdata.com <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com> <http://bigdata.com>
> <http://bigdata.com> <http://bigdata.com> <http://bigdata.com>
> > > > http://mapgraph.io
> >
> > > > Blazegraph™ <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> <http://www.blazegraph.com/>
> <http://www.blazegraph.com/> is our ultra high-performance graph database
> that supports both RDF/SPARQL and Tinkerpop/Blueprints APIs.  MapGraph™
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph>
> <http://www.systap.com/mapgraph> <http://www.systap.com/mapgraph> is our
> disruptive new technology to use GPUs to accelerate data-parallel graph
> analytics.
> >
> > > > CONFIDENTIALITY NOTICE:  This email and its contents and attachments
> are for the sole use of the intended recipient(s) and are confidential or
> proprietary to SYSTAP. Any unauthorized review, use, disclosure,
> dissemination or copying of this email or its contents or attachments is
> prohibited. If you have received this communication in error, please notify
> the sender by reply email and permanently delete all copies of the email
> and its contents and attachments.
> >
> >
> >
> >
> >
> >
>
> - --
> Maximilian Brodhun
>
> Abteilung Forschung und Entwicklung
> Georg-August-Universität Göttingen
> Niedersächsische Staats- und Universitätsbibliothek Göttingen
> D-37070 Göttingen
>
> Papendiek 14 (Historisches Gebäude, Raum 2.409)
> +49 551 39-4923 (Tel.)
>
> br...@su...
> <javascript:_e(%7B%7D,'cvml','br...@su...');>
>
> http://www.sub.uni-goettingende/
> http://www.rdd.sub.uni-goettingen.de/
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1
>
> iQIcBAEBAgAGBQJVE+FGAAoJEBDMOSiH8mYu1AQP/3t4YqHm+PAL+jGTf3wH8qry
> UUy5kM1eTYRKrZ0+il6KLFaGsewkAjfy1ordWXq+WO80RmZo4yqepnO+pkUF8LvY
> GE+6vUnzRM6ny+JznmPfIIrWND/qUA75alUvYCD2aPoF/MDosDpHYU9tZxQfUPIm
> 74Y2QyLsZvkRppSan7S0m3EOX0ZYbUCN+FfwwseEAr3bVo2l+DToTQyWazRltxnq
> cj3p2iUUrxP6ojDzn7nVhZJCrnxvGluAgYSHsbD0TVRD1rKs32yZwRg1RtKs2/xh
> 105jV8in+LKIfOJ8nmlVtk0UWPPOecxHiMMpbHC4VlL6HHDJPvR0lzfiVKR90lp8
> 5sh4FAXgvDqsDOLJCdn2x7DyMrUWuNpkbcGFoHdZlrdHDE+soewmLCCdPxMwVJM2
> nvvMuoA3hZ1bAiRjksNyQ0KzFfO31DNSCYDVfYTJw34vCz36xwZW+3dwnhh9oibT
> cDfNQCu97b/UIjobMXvP+JMNtFiDdJmCIKFwXpzsF0HyLYS+5x++qDWzzNyY8wFR
> Xj3sWx+N06mvaWiOnsn2i0j/lSjWPZS7eCKT/113wCoc083AhxXRhDY2bOUHWbz5
> BjOccRkNtbmHl1e3IY4loIhJOh9xQznOXOYYTa4BZwWWnh1fYWQ5GmV5tafU7Ayy
> TmImZmmEQTC05LncxLFt
> =x6f1
> -----END PGP SIGNATURE-----
>
>

-- 
----
Bryan Thompson
Chief Scientist & Founder
SYSTAP, LLC
4501 Tower Road
Greensboro, NC 27410
br...@sy...
http://blazegraph.com
http://blog.bigdata.com <http://bigdata.com>
http://mapgraph.io

Blazegraph™ <http://www.blazegraph.com/> is our ultra high-performance
graph database that supports both RDF/SPARQL and Tinkerpop/Blueprints
APIs.  MapGraph™ <http://www.systap.com/mapgraph> is our disruptive new
technology to use GPUs to accelerate data-parallel graph analytics.

CONFIDENTIALITY NOTICE:  This email and its contents and attachments are
for the sole use of the intended recipient(s) and are confidential or
proprietary to SYSTAP. Any unauthorized review, use, disclosure,
dissemination or copying of this email or its contents or attachments is
prohibited. If you have received this communication in error, please notify
the sender by reply email and permanently delete all copies of the email
and its contents and attachments.