Re: Reg :- Multiple Node Cluster set up in Virtual Box

[@Nandan@]

Hi All,

Thanks for sharing all information.
I am starting to work on this.
Now Problem which I am getting right now is:-
1) How to select Network for Virtual Machine by which I can able to get
different IP for different Virtual Box?
2) As I am using WIFI for HOST machine which is Windows 10, so is there any
internal configuration required or I need to select specific Network
Adapter into Virtual Boxs by which  I will get IP1,IP2,IP3 for
node1,node2,node3 respectively.

Please give me some ideas.
Thanks in advance,
Nandan Priyadarshi


On Tue, Nov 7, 2017 at 8:28 AM, James Briggs  wrote:

> Nandan: The original Datastax training classes (when it was still called
> Riptano)
> used 3 virtualbox Debian instances to setup a Cassandra cluster.
>
> Thanks, James Briggs.
> --
> Cassandra/MySQL DBA. Available in San Jose area or remote.
> cass_top: https://github.com/jamesbriggs/cassandra-top
>
>
> --
> *From:* kurt greaves 
> *To:* User 
> *Sent:* Monday, November 6, 2017 3:08 PM
> *Subject:* Re: Reg :- Multiple Node Cluster set up in Virtual Box
>
> Worth keeping in mind that in 3.6 onwards nodes will not start unless they
> can contact a seed. Not quite SPOF but still problematic. CASSANDRA-13851
> ​
>
>
>

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[James Briggs]

Nandan: The original Datastax training classes (when it was still called 
Riptano)
used 3 virtualbox Debian instances to setup a Cassandra cluster. Thanks, James 
Briggs.
--
Cassandra/MySQL DBA. Available in San Jose area or remote.
cass_top: https://github.com/jamesbriggs/cassandra-top

  From: kurt greaves 
 To: User  
 Sent: Monday, November 6, 2017 3:08 PM
 Subject: Re: Reg :- Multiple Node Cluster set up in Virtual Box
   
Worth keeping in mind that in 3.6 onwards nodes will not start unless they can 
contact a seed. Not quite SPOF but still problematic. CASSANDRA-13851​

   

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[kurt greaves]

Worth keeping in mind that in 3.6 onwards nodes will not start unless they
can contact a seed. Not quite SPOF but still problematic. CASSANDRA-13851
​

What can NOT be done during repairs (2.2.x and 3.0.x)

[Herbert Fischer]

Hello,

I know from DS docs that we must not change the cluster topology while
there are repairs going on, and vice-versa.

Is there anything else that cannot be done while there are repairs
happening? Are schema-changes fine? Is creating new keyspaces or tables
fine?

best,

Herbert

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[Jeff Jirsa]

> On Nov 6, 2017, at 6:52 AM, @Nandan@  wrote:
> 
> Hi Jeff,
> Thanks for you suggestion.
> I have few questions here. 
> 1) It is fine to set up first node and put node1 ip as seed , and then I have 
> to follow to set up node 2,3,4 respectively with same seed node as node1 's 
> ip address. but this will also comes as SPOF as if node1 will close for some 
> time being.. 

Once the cluster is up you can change the seeds. The seeds are not a SPOF for a 
running cluster - they’re used the first time a node joins the ring (and for 
some other stuff you can ignore for now), as long as the cluster isn’t growing 
seeds don’t matter very much.



> 2) is it possible that ,after cluster set up with node1 's IP as seed and 
> cluster name as "ABC Cluster" and later I can change seed nodes like to add 
> node3 and node4 IP address in all 4 nodes. 

Yes, you can change seeds later. You can’t change the cluster name, though.

> 3) if it is possible then , are there nay chances that cluster may get 
> disturbed due to changing seeds. 

Not really, people tend to think seeds are more important than they are. You 
want them to be the same, but if they’re different across nodes for a while 
it’s unlikely to hurt much. 

> Please clarify these doubts. 
> 
> 
>> On Nov 6, 2017 10:44 PM, "Jeff Jirsa"  wrote:
>> Looks like official docs for first-time-setup are pretty lacking.
>> 
>> One node at a time:
>> - Install the deb package: 
>> http://cassandra.apache.org/doc/latest/getting_started/installing.html 
>> - Then configure - 
>> http://cassandra.apache.org/doc/latest/getting_started/configuring.html
>>   - Pick a cluster name
>>   - Set the listen_address (and maybe broadcast_address)
>>   - Put the IP of the first node as the seed.
>> - Start the node
>> - Wait 2 minutes and then proceed to the next one.
>> 
>> 
>> 
>>  
>> 
>>> On Mon, Nov 6, 2017 at 6:33 AM, @Nandan@  
>>> wrote:
>>> Hi Varun ,
>>> I tried CCM , but as for practice and for deep learning , finally I 
>>> understood that CCM is not a good way to go along.
>>> Like my goal is to learn about configuration aspects as well as to know in 
>>> details about administration parts. 
>>> So I am trying to do configure all 4 virtual boxs as 4 nodes. 
>>> Thanks for reply. Hope we will work on this .
>>> 
>>> 
>>> Thanks,
>>> Nandan Priyadarshi
>>> 
 On Nov 6, 2017 10:29 PM, "Varun Barala"  wrote:
 you can try CCM
 https://academy.datastax.com/planet-cassandra/getting-started-with-ccm-cassandra-cluster-manager
 
 Thanks
 
> On Mon, Nov 6, 2017 at 10:12 PM, @Nandan@ 
>  wrote:
> Hi Users ,
>  Just seeking some perfect guidelines to set up multi-node cluster  
> single Data Center in single host machine. 
> I am currently using windows 10 as host machine and installed Oracle 
> virtual box in which I created 4 virtual machines and all had Ubuntu 
> 16.04 
> I would like to configure a flexible robust no SPOF  data center.
> So please let me know how do I start and what steps, I have to follow to 
> configure this multi node cluster?
> My goal is to create 4 node cluster now and later based on learning 
> experiences I will remove 1 node and add 2 more nodes to check everything 
> should be working perfectly.
> 
> Just hope to get some step by step guidelines from all of you.
> 
> Thanks in advance and best regards,
> Nandan Priyadarshi 
 
>> 

3.0.6 - CorruptSSTableException

[Riccardo Ferrari]

Hi list,

It happened that one of our EC2 instance of our cluster got rebooted.
Unfortunately when back Cassandra 3.0.6 failed to restart complaining about:

ERROR [NonPeriodicTasks:1] 2017-11-04 03:44:20,019 LogTransaction.java:204
- Unable to delete //system/local/ma-292-big-Data.db as it does
not exist
ERROR [CompactionExecutor:1] 2017-11-04 03:44:20,136
CassandraDaemon.java:195 - Exception in thread
Thread[CompactionExecutor:1,1,main]
org.apache.cassandra.io.sstable.CorruptSSTableException: Corrupted:
//system/local/ma-294-big-Data.db
at
org.apache.cassandra.io.sstable.format.big.BigTableScanner$KeyScanningIterator.computeNext(BigTableScanner.java:351)
~[apache-cassandra-3.0.6.jar:3.0.6]

I tried to run an sstablescrub system local with no luck as it seems to get
stuck in an endless loop:

Scrubbing
BigTableReader(path='//system/local/ma-293-big-Data.db') (5862
bytes)
WARNING: Error reading row (stacktrace follows):
Retrying from row index; data is 5855 bytes starting at 7
WARNING: Retry failed too. Skipping to next row (retry's stacktrace follows)
WARNING: Error reading row (stacktrace follows):
WARNING: Row starting at position 0 is unreadable; skipping to next
WARNING: Data file row position 7 differs from index file row position -1
WARNING: Error reading row (stacktrace follows):

I do have a snapshot taken 3 hours before the crash.
Is it safe to restore that snapshot and run a repair?
Is there anything else I can try to recover ?

Any help much appreciated,
regards

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[@Nandan@]

Hi Jeff,
Thanks for you suggestion.
I have few questions here.
1) It is fine to set up first node and put node1 ip as seed , and then I
have to follow to set up node 2,3,4 respectively with same seed node as
node1 's ip address. but this will also comes as SPOF as if node1 will
close for some time being..
2) is it possible that ,after cluster set up with node1 's IP as seed and
cluster name as "ABC Cluster" and later I can change seed nodes like to add
node3 and node4 IP address in all 4 nodes.
3) if it is possible then , are there nay chances that cluster may get
disturbed due to changing seeds.
Please clarify these doubts.

Thanks and Best Regards,
Nandan Priyadarshi


On Nov 6, 2017 10:44 PM, "Jeff Jirsa"  wrote:

> Looks like official docs for first-time-setup are pretty lacking.
>
> One node at a time:
> - Install the deb package: http://cassandra.apache.org/doc/latest/getting_
> started/installing.html
> - Then configure - http://cassandra.apache.org/doc/latest/getting_started/
> configuring.html
>   - Pick a cluster name
>   - Set the listen_address (and maybe broadcast_address)
>   - Put the IP of the first node as the seed.
> - Start the node
> - Wait 2 minutes and then proceed to the next one.
>
>
>
>
>
> On Mon, Nov 6, 2017 at 6:33 AM, @Nandan@ 
> wrote:
>
>> Hi Varun ,
>> I tried CCM , but as for practice and for deep learning , finally I
>> understood that CCM is not a good way to go along.
>> Like my goal is to learn about configuration aspects as well as to know
>> in details about administration parts.
>> So I am trying to do configure all 4 virtual boxs as 4 nodes.
>> Thanks for reply. Hope we will work on this .
>>
>>
>> Thanks,
>> Nandan Priyadarshi
>>
>> On Nov 6, 2017 10:29 PM, "Varun Barala"  wrote:
>>
>>> you can try *CCM*
>>> https://academy.datastax.com/planet-cassandra/getting-starte
>>> d-with-ccm-cassandra-cluster-manager
>>>
>>> Thanks
>>>
>>> On Mon, Nov 6, 2017 at 10:12 PM, @Nandan@ >> m> wrote:
>>>
 Hi Users ,
  Just seeking some perfect guidelines to set up multi-node cluster
 single Data Center in single host machine.
 I am currently using windows 10 as host machine and installed Oracle
 virtual box in which I created 4 virtual machines and all had Ubuntu 16.04
 I would like to configure a flexible robust no SPOF  data center.
 So please let me know how do I start and what steps, I have to follow
 to configure this multi node cluster?
 My goal is to create 4 node cluster now and later based on learning
 experiences I will remove 1 node and add 2 more nodes to check everything
 should be working perfectly.

 Just hope to get some step by step guidelines from all of you.

 Thanks in advance and best regards,
 Nandan Priyadarshi

>>>
>>>
>

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[Jürgen Albersdorfer]

You might want to have a look onto this:
https://www.digitalocean.com/community/tutorials/how-to-run-a-multi-node-cluster-database-with-cassandra-on-ubuntu-14-04

It worked great for me, but I have to admit that my cluster is running, but not 
yet in production.

Von: @Nandan@ [mailto:nandanpriyadarshi...@gmail.com]
Gesendet: Montag, 6. November 2017 15:34
An: user 
Betreff: Re: Reg :- Multiple Node Cluster set up in Virtual Box

Hi Varun ,
I tried CCM , but as for practice and for deep learning , finally I understood 
that CCM is not a good way to go along.
Like my goal is to learn about configuration aspects as well as to know in 
details about administration parts.
So I am trying to do configure all 4 virtual boxs as 4 nodes.
Thanks for reply. Hope we will work on this .


Thanks,
Nandan Priyadarshi

On Nov 6, 2017 10:29 PM, "Varun Barala" 
> wrote:
you can try CCM
https://academy.datastax.com/planet-cassandra/getting-started-with-ccm-cassandra-cluster-manager

Thanks

On Mon, Nov 6, 2017 at 10:12 PM, @Nandan@ 
> wrote:
Hi Users ,
 Just seeking some perfect guidelines to set up multi-node cluster  single Data 
Center in single host machine.
I am currently using windows 10 as host machine and installed Oracle virtual 
box in which I created 4 virtual machines and all had Ubuntu 16.04
I would like to configure a flexible robust no SPOF  data center.
So please let me know how do I start and what steps, I have to follow to 
configure this multi node cluster?
My goal is to create 4 node cluster now and later based on learning experiences 
I will remove 1 node and add 2 more nodes to check everything should be working 
perfectly.

Just hope to get some step by step guidelines from all of you.

Thanks in advance and best regards,
Nandan Priyadarshi

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[Jeff Jirsa]

Looks like official docs for first-time-setup are pretty lacking.

One node at a time:
- Install the deb package:
http://cassandra.apache.org/doc/latest/getting_started/installing.html
- Then configure -
http://cassandra.apache.org/doc/latest/getting_started/configuring.html
  - Pick a cluster name
  - Set the listen_address (and maybe broadcast_address)
  - Put the IP of the first node as the seed.
- Start the node
- Wait 2 minutes and then proceed to the next one.





On Mon, Nov 6, 2017 at 6:33 AM, @Nandan@ 
wrote:

> Hi Varun ,
> I tried CCM , but as for practice and for deep learning , finally I
> understood that CCM is not a good way to go along.
> Like my goal is to learn about configuration aspects as well as to know in
> details about administration parts.
> So I am trying to do configure all 4 virtual boxs as 4 nodes.
> Thanks for reply. Hope we will work on this .
>
>
> Thanks,
> Nandan Priyadarshi
>
> On Nov 6, 2017 10:29 PM, "Varun Barala"  wrote:
>
>> you can try *CCM*
>> https://academy.datastax.com/planet-cassandra/getting-starte
>> d-with-ccm-cassandra-cluster-manager
>>
>> Thanks
>>
>> On Mon, Nov 6, 2017 at 10:12 PM, @Nandan@ > > wrote:
>>
>>> Hi Users ,
>>>  Just seeking some perfect guidelines to set up multi-node cluster
>>> single Data Center in single host machine.
>>> I am currently using windows 10 as host machine and installed Oracle
>>> virtual box in which I created 4 virtual machines and all had Ubuntu 16.04
>>> I would like to configure a flexible robust no SPOF  data center.
>>> So please let me know how do I start and what steps, I have to follow to
>>> configure this multi node cluster?
>>> My goal is to create 4 node cluster now and later based on learning
>>> experiences I will remove 1 node and add 2 more nodes to check everything
>>> should be working perfectly.
>>>
>>> Just hope to get some step by step guidelines from all of you.
>>>
>>> Thanks in advance and best regards,
>>> Nandan Priyadarshi
>>>
>>
>>

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[@Nandan@]

Hi Varun ,
I tried CCM , but as for practice and for deep learning , finally I
understood that CCM is not a good way to go along.
Like my goal is to learn about configuration aspects as well as to know in
details about administration parts.
So I am trying to do configure all 4 virtual boxs as 4 nodes.
Thanks for reply. Hope we will work on this .


Thanks,
Nandan Priyadarshi

On Nov 6, 2017 10:29 PM, "Varun Barala"  wrote:

> you can try *CCM*
> https://academy.datastax.com/planet-cassandra/getting-
> started-with-ccm-cassandra-cluster-manager
>
> Thanks
>
> On Mon, Nov 6, 2017 at 10:12 PM, @Nandan@ 
> wrote:
>
>> Hi Users ,
>>  Just seeking some perfect guidelines to set up multi-node cluster
>> single Data Center in single host machine.
>> I am currently using windows 10 as host machine and installed Oracle
>> virtual box in which I created 4 virtual machines and all had Ubuntu 16.04
>> I would like to configure a flexible robust no SPOF  data center.
>> So please let me know how do I start and what steps, I have to follow to
>> configure this multi node cluster?
>> My goal is to create 4 node cluster now and later based on learning
>> experiences I will remove 1 node and add 2 more nodes to check everything
>> should be working perfectly.
>>
>> Just hope to get some step by step guidelines from all of you.
>>
>> Thanks in advance and best regards,
>> Nandan Priyadarshi
>>
>
>

Re: Reg :- Multiple Node Cluster set up in Virtual Box

[Varun Barala]

you can try *CCM*
https://academy.datastax.com/planet-cassandra/getting-started-with-ccm-cassandra-cluster-manager

Thanks

On Mon, Nov 6, 2017 at 10:12 PM, @Nandan@ 
wrote:

> Hi Users ,
>  Just seeking some perfect guidelines to set up multi-node cluster  single
> Data Center in single host machine.
> I am currently using windows 10 as host machine and installed Oracle
> virtual box in which I created 4 virtual machines and all had Ubuntu 16.04
> I would like to configure a flexible robust no SPOF  data center.
> So please let me know how do I start and what steps, I have to follow to
> configure this multi node cluster?
> My goal is to create 4 node cluster now and later based on learning
> experiences I will remove 1 node and add 2 more nodes to check everything
> should be working perfectly.
>
> Just hope to get some step by step guidelines from all of you.
>
> Thanks in advance and best regards,
> Nandan Priyadarshi
>

Reg :- Multiple Node Cluster set up in Virtual Box

[@Nandan@]

Hi Users ,
 Just seeking some perfect guidelines to set up multi-node cluster  single
Data Center in single host machine.
I am currently using windows 10 as host machine and installed Oracle
virtual box in which I created 4 virtual machines and all had Ubuntu 16.04
I would like to configure a flexible robust no SPOF  data center.
So please let me know how do I start and what steps, I have to follow to
configure this multi node cluster?
My goal is to create 4 node cluster now and later based on learning
experiences I will remove 1 node and add 2 more nodes to check everything
should be working perfectly.

Just hope to get some step by step guidelines from all of you.

Thanks in advance and best regards,
Nandan Priyadarshi

Re: poolingOptions not serializable?

[Andrea Giordano]

Without pooling options I have no errors and all works correctly (with a
light throughput ).

Trying to raise it, flink gave me a pool busy error about Cassandra So I
used pooling options. Now when I start the program I have the problem
described here
El 6 nov. 2017 9:48, "Nicolas Guyomar"  escribió:

> Hi Andrea,
>
> Do you have the error using the builder ?
>
> PoolingOptions poolingOptions = new PoolingOptions();
> poolingOptions
> .setMaxRequestsPerConnection(HostDistance.LOCAL, 32768)
> .setMaxRequestsPerConnection(HostDistance.REMOTE, 1);
>
>
> Builder builder = Cluster.builder();
> builder.addContactPoint(CASSANDRA_ADDRESS);
> builder.withPort(CASSANDRA_PORT);
> builder.withPoolingOptions(poolingOptions);
>
>
> sinkBuilderNormalStream
> .setQuery("INSERT INTO keyspace_local.values_by_sensors_users"
> + " (user, sensor, timestamp, rdf_stream, observed_value, value)"
> + " VALUES (?, ?, ?, ?, ?, ?);")
> .setClusterBuilder(builder)
> .build();
>
>
> On 4 November 2017 at 19:27, Andrea Giordano <
> andrea.giordano@gmail.com> wrote:
>
>> Hi,
>> I’m using datastax driver to use Cassandra as sink for some data streams
>> with Apache Flink:
>> I have a problem executing my application raising an error about the full
>> queue. I discovered that the default value is 256, probably too low for my
>> load, so I have raised it using poolingOptions setting
>> maxRequestsPerConnection as suggested here: http://docs.datastax.com
>> /en/developer/java-driver/3.1/manual/pooling/.
>>
>> Unfortunately with the following code I obtain the following error when I
>> launch it:
>>
>> The implementation of the ClusterBuilder is not serializable.
>> The object probably contains or references non serializable fields.
>>
>>
>> My code:
>>
>>
>> PoolingOptions poolingOptions = new PoolingOptions();
>> poolingOptions
>>   .setMaxRequestsPerConnection(HostDistance.LOCAL, 32768)
>>   .setMaxRequestsPerConnection(HostDistance.REMOTE, 1);
>>
>>
>> ClusterBuilder cassandraBuilder = new ClusterBuilder() {
>> private static final long serialVersionUID = 1L;
>>
>> @Override
>> public Cluster buildCluster(Cluster.Builder builder) {
>> return builder.addContactPoint(CASSANDRA_ADDRESS).withPort(CASSANDRA_PORT
>> )..withPoolingOptions(poolingOptions).build();
>> }
>> };
>>
>>
>> sinkBuilderNormalStream
>> .setQuery("INSERT INTO keyspace_local.values_by_sensors_users"
>> + " (user, sensor, timestamp, rdf_stream, observed_value, value)"
>> + " VALUES (?, ?, ?, ?, ?, ?);")
>> .setClusterBuilder(cassandraBuilder)
>> .build();
>>
>>
>> How can I deal with it?
>>
>
>

Re: poolingOptions not serializable?

[Nicolas Guyomar]

Hi Andrea,

Do you have the error using the builder ?

PoolingOptions poolingOptions = new PoolingOptions();
poolingOptions
.setMaxRequestsPerConnection(HostDistance.LOCAL, 32768)
.setMaxRequestsPerConnection(HostDistance.REMOTE, 1);


Builder builder = Cluster.builder();
builder.addContactPoint(CASSANDRA_ADDRESS);
builder.withPort(CASSANDRA_PORT);
builder.withPoolingOptions(poolingOptions);


sinkBuilderNormalStream
.setQuery("INSERT INTO keyspace_local.values_by_sensors_users"
+ " (user, sensor, timestamp, rdf_stream, observed_value, value)"
+ " VALUES (?, ?, ?, ?, ?, ?);")
.setClusterBuilder(builder)
.build();


On 4 November 2017 at 19:27, Andrea Giordano 
wrote:

> Hi,
> I’m using datastax driver to use Cassandra as sink for some data streams
> with Apache Flink:
> I have a problem executing my application raising an error about the full
> queue. I discovered that the default value is 256, probably too low for my
> load, so I have raised it using poolingOptions setting
> maxRequestsPerConnection as suggested here: http://docs.datastax.
> com/en/developer/java-driver/3.1/manual/pooling/.
>
> Unfortunately with the following code I obtain the following error when I
> launch it:
>
> The implementation of the ClusterBuilder is not serializable.
> The object probably contains or references non serializable fields.
>
>
> My code:
>
>
> PoolingOptions poolingOptions = new PoolingOptions();
> poolingOptions
>   .setMaxRequestsPerConnection(HostDistance.LOCAL, 32768)
>   .setMaxRequestsPerConnection(HostDistance.REMOTE, 1);
>
>
> ClusterBuilder cassandraBuilder = new ClusterBuilder() {
> private static final long serialVersionUID = 1L;
>
> @Override
> public Cluster buildCluster(Cluster.Builder builder) {
> return builder.addContactPoint(CASSANDRA_ADDRESS).withPort(CASSANDRA_PORT
> )..withPoolingOptions(poolingOptions).build();
> }
> };
>
>
> sinkBuilderNormalStream
> .setQuery("INSERT INTO keyspace_local.values_by_sensors_users"
> + " (user, sensor, timestamp, rdf_stream, observed_value, value)"
> + " VALUES (?, ?, ?, ?, ?, ?);")
> .setClusterBuilder(cassandraBuilder)
> .build();
>
>
> How can I deal with it?
>