You are viewing a plain text version of this content. The canonical link for it is here.
Posted to solr-user@lucene.apache.org by "Katz, Dan" <Da...@fepoc.com> on 2009/10/14 21:48:26 UTC
advice on failover setup
Hi folks,
I'm tasked with designing a failover architecture for our new Solr
server. I've read the Replication section in the docs
(http://wiki.apache.org/solr/SolrReplication) and I need some
clarification/insight. My questions:
1. Is there such a thing as "master/master" replication?
2. If we have one master and one slave server, and the master goes
down, does the slave automatically become the master? What's the process
for brining the server back up and getting the two back in sync? Is it a
manual process always?
3. We're running Solr inside Tomcat on Windows currently. Any
suggestions for a load balancer that will automatically switch to the
alternate server if one goes down?
Thanks in advance,
--
Dan Katz
Lead Web Developer
FEP Operations Center(r)
202.203.2572 (Direct)
dan.katz@fepoc.com
********************************************************************************
Unauthorized interception of this communication could be a
violation of Federal and State Law. This communication and
any files transmitted with it are confidential and may contain
protected health information. This communication is solely
for the use of the person or entity to whom it was addressed.
If you are not the intended recipient, any use, distribution,
printing or acting in reliance on the contents of this message
is strictly prohibited. If you have received this message
in error, please notify the sender and destroy any and all copies.
Thank you.
*******************************************************************************
Re: advice on failover setup
Posted by Yonik Seeley <yo...@lucidimagination.com>.
On Wed, Oct 14, 2009 at 3:48 PM, Katz, Dan <Da...@fepoc.com> wrote:
> I'm tasked with designing a failover architecture for our new Solr
> server. I've read the Replication section in the docs
> (http://wiki.apache.org/solr/SolrReplication) and I need some
> clarification/insight. My questions:
>
> 1. Is there such a thing as "master/master" replication?
Yes, a server can be both a master and a slave.
> 2. If we have one master and one slave server, and the master goes
> down, does the slave automatically become the master?
No. Solr's fault tolerance is currently limited to the search-side of
things (via load balancing across replicas).
> What's the process
> for brining the server back up and getting the two back in sync? Is it a
> manual process always?
Yes, it currently is. It will be automatic in the future, but that
won't help you now :-)
> 3. We're running Solr inside Tomcat on Windows currently. Any
> suggestions for a load balancer that will automatically switch to the
> alternate server if one goes down?
There is a new load balancing client for Java in Solr 1.4
http://lucene.apache.org/solr/api/org/apache/solr/client/solrj/impl/LBHttpSolrServer.html
Perhaps others have recommendations for generic HTTP load balancers.
-Yonik
http://www.lucidimagination.com
> Thanks in advance,
>
> --
> Dan Katz
> Lead Web Developer
> FEP Operations Center(r)
> 202.203.2572 (Direct)
> dan.katz@fepoc.com
>
>
> ********************************************************************************
> Unauthorized interception of this communication could be a
> violation of Federal and State Law. This communication and
> any files transmitted with it are confidential and may contain
> protected health information. This communication is solely
> for the use of the person or entity to whom it was addressed.
> If you are not the intended recipient, any use, distribution,
> printing or acting in reliance on the contents of this message
> is strictly prohibited. If you have received this message
> in error, please notify the sender and destroy any and all copies.
> Thank you.
> *******************************************************************************
>
>
Re: advice on failover setup
Posted by Jason Rutherglen <ja...@gmail.com>.
Don,
I neglected to mention the Solr Katta integration patch, SOLR-1395
That's a great place to start coding wise!
-J
On Wed, Oct 14, 2009 at 4:20 PM, Don Clore <do...@5to1.com> wrote:
> I'm sorry, for clarification, is it the *wiki# pages that are under
> development, or the features (I'm guessing the latter)?
>
> If the latter (ZooKeeperIntegration and KattaIntegration are not available
> yet), is there any sort of guess as to when these features might become
> available?
>
> thanks,
> Don
>
> On Wed, Oct 14, 2009 at 2:13 PM, Jason Rutherglen <
> jason.rutherglen@gmail.com> wrote:
>
>> Dan,
>>
>> For automatic failover there are 2 wiki pages that may be helpful,
>> however both are in the development stage.
>>
>> http://wiki.apache.org/solr/ZooKeeperIntegration
>> http://wiki.apache.org/solr/KattaIntegration
>>
>> -J
>>
>> On Wed, Oct 14, 2009 at 12:48 PM, Katz, Dan <Da...@fepoc.com> wrote:
>> > Hi folks,
>> >
>> > I'm tasked with designing a failover architecture for our new Solr
>> > server. I've read the Replication section in the docs
>> > (http://wiki.apache.org/solr/SolrReplication) and I need some
>> > clarification/insight. My questions:
>> >
>> > 1. Is there such a thing as "master/master" replication?
>> > 2. If we have one master and one slave server, and the master goes
>> > down, does the slave automatically become the master? What's the process
>> > for brining the server back up and getting the two back in sync? Is it a
>> > manual process always?
>> > 3. We're running Solr inside Tomcat on Windows currently. Any
>> > suggestions for a load balancer that will automatically switch to the
>> > alternate server if one goes down?
>> >
>> > Thanks in advance,
>> >
>> > --
>> > Dan Katz
>> > Lead Web Developer
>> > FEP Operations Center(r)
>> > 202.203.2572 (Direct)
>> > dan.katz@fepoc.com
>> >
>> >
>> >
>> ********************************************************************************
>> > Unauthorized interception of this communication could be a
>> > violation of Federal and State Law. This communication and
>> > any files transmitted with it are confidential and may contain
>> > protected health information. This communication is solely
>> > for the use of the person or entity to whom it was addressed.
>> > If you are not the intended recipient, any use, distribution,
>> > printing or acting in reliance on the contents of this message
>> > is strictly prohibited. If you have received this message
>> > in error, please notify the sender and destroy any and all copies.
>> > Thank you.
>> >
>> *******************************************************************************
>> >
>> >
>>
>
Re: advice on failover setup
Posted by Jason Rutherglen <ja...@gmail.com>.
Don,
Sorry, yes the features are under development and also hopefully
the wikis as well. :)
When they become available, well I can say personally I need the
Katta integration working in the next few months. Jason Venner
got it working over at his company. It might be good to
describe your use case to see what is a good fit for you.
-J
On Wed, Oct 14, 2009 at 4:20 PM, Don Clore <do...@5to1.com> wrote:
> I'm sorry, for clarification, is it the *wiki# pages that are under
> development, or the features (I'm guessing the latter)?
>
> If the latter (ZooKeeperIntegration and KattaIntegration are not available
> yet), is there any sort of guess as to when these features might become
> available?
>
> thanks,
> Don
>
> On Wed, Oct 14, 2009 at 2:13 PM, Jason Rutherglen <
> jason.rutherglen@gmail.com> wrote:
>
>> Dan,
>>
>> For automatic failover there are 2 wiki pages that may be helpful,
>> however both are in the development stage.
>>
>> http://wiki.apache.org/solr/ZooKeeperIntegration
>> http://wiki.apache.org/solr/KattaIntegration
>>
>> -J
>>
>> On Wed, Oct 14, 2009 at 12:48 PM, Katz, Dan <Da...@fepoc.com> wrote:
>> > Hi folks,
>> >
>> > I'm tasked with designing a failover architecture for our new Solr
>> > server. I've read the Replication section in the docs
>> > (http://wiki.apache.org/solr/SolrReplication) and I need some
>> > clarification/insight. My questions:
>> >
>> > 1. Is there such a thing as "master/master" replication?
>> > 2. If we have one master and one slave server, and the master goes
>> > down, does the slave automatically become the master? What's the process
>> > for brining the server back up and getting the two back in sync? Is it a
>> > manual process always?
>> > 3. We're running Solr inside Tomcat on Windows currently. Any
>> > suggestions for a load balancer that will automatically switch to the
>> > alternate server if one goes down?
>> >
>> > Thanks in advance,
>> >
>> > --
>> > Dan Katz
>> > Lead Web Developer
>> > FEP Operations Center(r)
>> > 202.203.2572 (Direct)
>> > dan.katz@fepoc.com
>> >
>> >
>> >
>> ********************************************************************************
>> > Unauthorized interception of this communication could be a
>> > violation of Federal and State Law. This communication and
>> > any files transmitted with it are confidential and may contain
>> > protected health information. This communication is solely
>> > for the use of the person or entity to whom it was addressed.
>> > If you are not the intended recipient, any use, distribution,
>> > printing or acting in reliance on the contents of this message
>> > is strictly prohibited. If you have received this message
>> > in error, please notify the sender and destroy any and all copies.
>> > Thank you.
>> >
>> *******************************************************************************
>> >
>> >
>>
>
Re: advice on failover setup
Posted by Don Clore <do...@5to1.com>.
I'm sorry, for clarification, is it the *wiki# pages that are under
development, or the features (I'm guessing the latter)?
If the latter (ZooKeeperIntegration and KattaIntegration are not available
yet), is there any sort of guess as to when these features might become
available?
thanks,
Don
On Wed, Oct 14, 2009 at 2:13 PM, Jason Rutherglen <
jason.rutherglen@gmail.com> wrote:
> Dan,
>
> For automatic failover there are 2 wiki pages that may be helpful,
> however both are in the development stage.
>
> http://wiki.apache.org/solr/ZooKeeperIntegration
> http://wiki.apache.org/solr/KattaIntegration
>
> -J
>
> On Wed, Oct 14, 2009 at 12:48 PM, Katz, Dan <Da...@fepoc.com> wrote:
> > Hi folks,
> >
> > I'm tasked with designing a failover architecture for our new Solr
> > server. I've read the Replication section in the docs
> > (http://wiki.apache.org/solr/SolrReplication) and I need some
> > clarification/insight. My questions:
> >
> > 1. Is there such a thing as "master/master" replication?
> > 2. If we have one master and one slave server, and the master goes
> > down, does the slave automatically become the master? What's the process
> > for brining the server back up and getting the two back in sync? Is it a
> > manual process always?
> > 3. We're running Solr inside Tomcat on Windows currently. Any
> > suggestions for a load balancer that will automatically switch to the
> > alternate server if one goes down?
> >
> > Thanks in advance,
> >
> > --
> > Dan Katz
> > Lead Web Developer
> > FEP Operations Center(r)
> > 202.203.2572 (Direct)
> > dan.katz@fepoc.com
> >
> >
> >
> ********************************************************************************
> > Unauthorized interception of this communication could be a
> > violation of Federal and State Law. This communication and
> > any files transmitted with it are confidential and may contain
> > protected health information. This communication is solely
> > for the use of the person or entity to whom it was addressed.
> > If you are not the intended recipient, any use, distribution,
> > printing or acting in reliance on the contents of this message
> > is strictly prohibited. If you have received this message
> > in error, please notify the sender and destroy any and all copies.
> > Thank you.
> >
> *******************************************************************************
> >
> >
>
Re: advice on failover setup
Posted by Jason Rutherglen <ja...@gmail.com>.
Dan,
For automatic failover there are 2 wiki pages that may be helpful,
however both are in the development stage.
http://wiki.apache.org/solr/ZooKeeperIntegration
http://wiki.apache.org/solr/KattaIntegration
-J
On Wed, Oct 14, 2009 at 12:48 PM, Katz, Dan <Da...@fepoc.com> wrote:
> Hi folks,
>
> I'm tasked with designing a failover architecture for our new Solr
> server. I've read the Replication section in the docs
> (http://wiki.apache.org/solr/SolrReplication) and I need some
> clarification/insight. My questions:
>
> 1. Is there such a thing as "master/master" replication?
> 2. If we have one master and one slave server, and the master goes
> down, does the slave automatically become the master? What's the process
> for brining the server back up and getting the two back in sync? Is it a
> manual process always?
> 3. We're running Solr inside Tomcat on Windows currently. Any
> suggestions for a load balancer that will automatically switch to the
> alternate server if one goes down?
>
> Thanks in advance,
>
> --
> Dan Katz
> Lead Web Developer
> FEP Operations Center(r)
> 202.203.2572 (Direct)
> dan.katz@fepoc.com
>
>
> ********************************************************************************
> Unauthorized interception of this communication could be a
> violation of Federal and State Law. This communication and
> any files transmitted with it are confidential and may contain
> protected health information. This communication is solely
> for the use of the person or entity to whom it was addressed.
> If you are not the intended recipient, any use, distribution,
> printing or acting in reliance on the contents of this message
> is strictly prohibited. If you have received this message
> in error, please notify the sender and destroy any and all copies.
> Thank you.
> *******************************************************************************
>
>