[akka-user] Why Should Nodes Usually Not Include Themselves In Seed Node List?

Hi, when discovering seed nodes dynamically, why is it important for every 
node (except for one) to not include itself in the seed node list?  How 
could split brain happen if all nodes have the same list order of all seed 
nodes?
From what I understand, split-brain can only happen if multiple nodes put 
themselves first in the seed node list, is that correct?
I ask because when doing static configuration, it's common for all nodes to 
have the same seed node list, which means all seed nodes would have 
themselves in the list.  And that seems to be common, recommended practice.
If doing dynamic discovery of seed nodes, is it okay for all nodes to have 
the same list of seed nodes, if we can guarantee they are all in the same 
order? From reading the docs, I gather that after sending a message to all 
seed nodes, if none respond, then it will try to join the first node in the 
list, which wouldn't work if the first node in the list isn't itself.
We are considering using the AWS SDK by not querying the ASG for its 
instance list until the cloudformation stack says it is complete, which 
should guarantee that all nodes trying to join the cluster will get the 
same list of instances in the ASG.  Then we can guarantee that all nodes 
would order the nodes the same - any reason this wouldn't work?
Thanks,
Curt
You received this message because you are subscribed to the Google Groups "Akka User List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]
To post to this group, send email to [email protected]
Visit this group at https://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.

It's only the first node in the seed nodes list that is special. It's only
that one that will join itself to bootstrap the cluster, after it has tried
joining the other nodes.
Sorting is fine.
For the AWS part you might get something from this blog post:
http://chrisloy.net/2014/05/11/akka-cluster-ec2-autoscaling.html
Patrik Nordwall
Akka Tech Lead
Lightbend <http://www.lightbend.com/> -  Reactive apps on the JVM
Twitter: @patriknw
You received this message because you are subscribed to the Google Groups "Akka User List" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]
To post to this group, send email to [email protected]
Visit this group at https://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.

Hello all, I have an existing cluster. When adding a new node, I saw that Opscenter saw the node in an unknown cluster. In the yaml, the cluster name is the same. So i have stopped the node and added it's ip address in the list of seeds. Now Opscenter sees my node. But nodetool status now sees it as UN, instead of UJ when it first started. One other mension is that even if I stop the node, remove
Dear group I have a cluster, and all nodes have a list of all the other nodes in config element akka.cluster.seed-nodes. This list is the same on all nodes. I am experiencing that only the node that is first in the list, is able to join itself upon startup. If any of the other nodes start up while the first node is down, they are unable to form a cluster. I can read from the scala code
Hello,  A datanode has been de-commissioned, but will not drop off the dead nodes list. The node is no longer in the dfs.hosts.include file, and is not in the dfs.hosts.exclude file. When starting the datanode back up - the datanode appears in the Live Nodes list, with a decommissioned status.   I have tried adding and removing the host from the exclude file, as well as the include file, and
Hi All, I have 10 domains in the seed list , *Nutch 1.7 *consistently crawls only 5 of those domaisn and ignores the other 5 domains , can you please let me know whats preventing it from crawling all the domains. I am running this on *Hadoop2.3.0* and in a cluster mode and giving a *depth of 10* when submitting the job. I have already set the *db.ignore.external.links *property to tru
Is there any harm from having all the nodes listed in the seeds list in cassandra.yaml? Donald A. Smith | Senior Software Engineer P: 425.201.3900 x 3866 C: (206) 819-5965 F: (646) 443-2333 [email protected] [AudienceScience]
i am trying to connect the mongodb using mongodb.MongoClient.connect() with simple url connection string replicaset. When i start the server its throwing the error as Error: Could not locate any valid servers in initial seed list this is my code where i am passing two mongodb server as follows var MongoClient = mongodb.MongoClient; MongoClient.connect('mongodb://192.168.0.16,192.168.0.23
Hello, I'm using Nutch 2.1 on top of Hadoop 1.0.4, with HBase 0.90.4 as storage system. I run Nutch in distributed mode. I need to associate an id to each url inside the seed list of nutch and to store this information in HBase. I think that I have to create a new column family in HBase and modify the gora and hbase configuration files in the nutch conf folder. However, I think I need to modify
Cloudera Express 5.4.7. A small cluster of five data nodes and one management node is built with the nodes having two network interfaces, a private (10.xxx.xxx.xxx) and a public (198.xxx.xxx.xxx) network. Cloudera Manager was used to install the datanodes using the private network interface. Once installation was complete, errors show on the data node logs that Cloudera uses the public
I currently have a Node.js application running on AWS Elastic Beanstalk via docker containers, as well as a 3-node Mongo Replica set running on adequate EC2 instances. To connect to Mongo in my application, I am using the Node.js version of the native MongoDB driver. Every morning, each copy of the application throws a single error. Could not locate any valid servers in initial seed list
We currently have a situation where we have a node appear in both the Live Node list and the Dead Node list in the namenode web UI. Using hdfs dfsadmin -printTopology, I see that the node is listed under /default-rack AND the actual rack it is in. Both lines list the same IP address, port, and hostname. I am wondering if there is any harm in having this, and if there is a way to remedy this without
After much investigation through many newsgroups and websites, I found out that this is the way you can execute a unix apache command to add a user to the htpasswd list with PHP. Problem is, it doesn't add any information to the htpasswd file. Note: The example below is... domain = my user name abc = user name to add to htpasswd list pass = password to add to htpasswd list Please Help
Check this out: [[email protected]] list hints limit 10; Using default cell limit of 100 null TimedOutException() at org.apache.cassandra.thrift.Cassandra$get_range_slices_result.read(Cassandra.java:12932) at org.apache.thrift.TServiceClient.receiveBase(TServiceClient.java:78) at org.apache.cassandra.thrift.Cassandra$Client.recv_get_range_slices(Cassandra.java:734) at org.apache
De: Wolf [mailto:[email protected]] Enviada em: quarta-feira, 27 de fevereiro de 2008 12:06 No, go with the Crushed purple velvet with the pageboy sleeves, that should accentuate the colors of your hair... :-P Okay, this is just getting weird... PHP General Mailing List (http://www.php.net/) To unsubscribe, visit: http://www.php.net/unsub.php
[beeswax] Autocomplete should include 'star' in field list ----------------------------------------------------------                  Key: HUE-1619                  URL: https://issues.cloudera.org/browse/HUE-1619              Project: Hue           Issue Type: Improvement           Components: app.beeswax     Affects Versions: 3.0.0            
I've setup a graph based on the newsfeed here - �http://docs.neo4j.org/chunked/stable/cypher-cookbook-newsfeed.html � I've also added tags connected to statuses with status-[:TAGGED]->tag I've come up with a few ways to get at the linked list of statuses connected to a specific tag, but I was wondering if there's a better way to do it.� Here's the best i've got so far (~200ms with 2119 statuses
Hi A small typo The annotation MemberGroupLayout is not include in the list of support annotations found at http://isis.apache.org/applib-guide/reference/recognized-annotations/about.html The MemberGroups is not listed as depreacted. Dave.
Hello, We are in the process of developing a vertical search engine for the medical industry – and I need to estimate server/sizing requirements to setup my environment – my question is, how do I estimate how many documents I will be fetching for a particular vertical? And – from where do I get the seed list of all the sites? Will dmoz health category be sufficient or will I have to purchase
Hi All, I am a fairly new Nutch user and this is my first post here. I have been attempting to use Nutch for a particular application I am building, a search engine. I have a pretty solid (I think) bash script that runs Nutch in deploy mode on a small Hadoop cluster, separating the generate-fetch-parse-update-index steps. Right now I'm just running with what I guess you'd call a "depth" of 1 in that
Hi, Is there a way to list all the nodes and sub nodes on the command line in Zookeeper ? Regards, Kaushal
There is a potential issue that I have faced while testing fault tolerance in voldemort. I initially started with 3 nodes : A,B,C. My config is : replication-factor=2,required-reads=1,preferred- reads=2,preferred-writes=2,required-writes=1. All my connections are pooled. Now, i shut down node 3 and then try to do a put.As soon as a key gets hashed to the node 3, it fails and program ends