cloud,cluster-computing,load-balancing,coreos , How does CoreOS load balancing work out there in the wild on a cloud service?

How does CoreOS load balancing work out there in the wild on a cloud service?


Tag: cloud,cluster-computing,load-balancing,coreos

Say I have a CoreOS cluster deployed on some cloud service somewhere.

Now I have, say, 4 machines running a node.js app which follows all the 12-factor tenets and one machine with Couchbase.

How does load balancing work in this scenario? Wouldn't ONE ip eventually run out of juice as a load balancer or is that pretty much impossible? Where do I point the DNS for it to work right?

In the past I had one IP with NGINX upfront, which then directed the incoming requests round-robin style.

How does this work with a CoreOS on a cloud service?


There are different approaches to achieve such task. In general, there should be cloud service load balancers in front of your infrastructure, clusters or datacenters. You will be dealing with two or three tier architectures.

The DNS points to your cloud internet-facing load balancer, which manages the client tier layer in any case. In the case of AWS, it must be through a CNAME record.

Auto scalable groups for each tier would potentially mitigate the risk of reduced availability of your infrastructure. Then, The Nginx instances are configured through cloud-config in case provisioning from bootstrap stage is required.

  1. Two tier architecture (your scenario)

    Each Nginx instance listens to an HTTP port and uses upstreams to do the routing (depending on your NodeJS apps distribution). The service discovery is achieved through etcd, using Registrator + SkyDNS / Consul or Weaver, so instead of upstreams, the Nginx resolver could be replaced by the internal DNS provided by such tools.

  2. Three tier architecture

    The same applies for Nginx instances as for two tier, although client tier ones solve to the business tier applications (using the internal cloud load balancer) plus the units contained locally. Business tier would behave similarly as client one in two tier architecture, but taking into account the proper configuration of security groups.

For both cases, the data tier forms an independent region (SkyDNS) or datacenter (Consul). Additionally, you could eventually skip Nginx, but you would need to open more ports in your security groups publicly.

Gathering knowledge from:

I was able to build:

TODO: Consul version. Although, there is the example for SkyDNS and Ambassadord.


** Let me please know your comments.


AWS Instance Type For 3,000+/Second Outgoing HTTP Requests?

I'm currently scaling up an application on AWS, we send thousands of outgoing HTTP requests per second to 3rd party API's and services. Are there any EC2 instances optimized for this type of work?

Normalization in Java (dataset of doubles)

I'm working on a clustering program, and have a dataset of doubles that I need to normalize in order to make sure that every double (variable) has the same influence. I would like to use min-max normalization where for every variable the min and max value are determined, but I'm...

Running cloudfoundry sample app on local sandbox

This question may seem pretty naive but I've just started using cloud foundry. I installed the eclipse plugin & deployed a sample app. But everytime, while running the app in local or starting the server it is being pushed to the remote cloud server & running on a unique domain... To Do web app deployment issue

Apologies for basic question. I am very new in web development . I have download javascript web To Do App from . Now I want to deploy with So that I can browse this app using But when try to deploy using "parse deploy" command getting...

Reading part of a file in S3 using Boto

I am trying to read 700MB file stored in S3. How ever I only require bytes from locations 73 to 1024. I tried to find a usable solution but failed to. Would be a great help if someone could help me out....

Predis: Pros and Cons of the two cluster strategies

I am using Predis to connect to a Redis 3.0.2 cluster. So, I have an option to use either predis or redis cluster strategy. I am wondering if there are any significant pros and cons of choosing one of them over the other? Based on my wandering through documentation, I...

In Spark, does the filter function turn the data into tuples?

Just wondering does the filter turn the data into tuples? For example val filesLines = sc.textFile("file.txt") val split_lines =";")) val filteredData = split_lines.filter(x => x(4)=="Blue") //from here if we wanted to map the data would it be using tuple format ie. x._3 OR x(3) val blueRecords = =>...

In ActiveMQ clustering, how to find the consumer who picked a particular message or set of messages?

We have set of queues and camel routes have been written to consume messages from the queue - queues live in an ActiveMQ server and camel routes in Apache Karaf. We have put up a cluster with multiple karaf nodes; problem is that these messages get read by camel routes...

IBM Connections 5 custom theme not getting published to second cluster node

we have the following small cluster setup: Server 1: DMGr, Nodeagent, Connections_server1 Server 2: Nodeagent, Connections_server2 All shared resources (including the customization dir) are on a separate file server that is linked on /opt/IBM/Connections/data/shared on both servers. I want to customize the defaultTheme, so I copied it to /customizationDir/themes/defaultTheme/custom.css After...

How to save my data in a cloud?

I don't know if this is possible, but I would like to save some data of my website, like images and documents in a cloud like dropbox or google drive and I want that these data are available and visible to someone looking my site. is it possible? (My intention...

Creating a scalable database for android app | cloud hosted

I am working to launch an app which in future will store huge number of users and there profile information. But I Wanted to start small to pre-test my app idea , so what will be the options for hosting database which are cost effective initially and scalable in future...

jdbc backed channel on a cluster mode

I want to use the jdbc backed message channel. Reading the doc says that it is supported but after that I found a thread in the forum that explicitly says don't. Could you please clarify me if it's supported? In this case, could you guide me to do so? thanks...

GoogleCloudMessaging - InstanceID.getInstance(), registering from client

I'm pretty new to programming in Java and generally programming. Right now I decided to make my own app that is supposed to utilize Google Cloud Messaging. And somehow I succeeded but then I realized I used method String regid = gcm.register(PROJECT_NUMBER); that is deprecated and I should now use...

Elastica Client with cluster

I have an Elasticsearch cluster with but I can't connect with Elastica Client. This is my configuration : $elasticaClient = new \Elastica\Client(array( 'servers' => array( array('host' => 'https://myusername:[email protected]', 'port' => 10050), array('host' => 'https://myusername:[email protected]', 'port' => 10062) ) )); $elasticaIndex = $elasticaClient->getIndex('test'); I got this error : Couldn't resolve...

Spark - How to count number of records by key

This is probably an easy problem but basically I have a dataset where I am to count the number of females for each country. Ultimately I want to group each count by the country but I am unsure of what to use for the value since there is not a...

Digi Device API reading Datastream / DIO / AD

I have XBee Gateway ZB, Wifi and I'm experimenting with the Digi Device API. I already tried the Heroku App and it works really good. So I wanted to access online the Datastream. Using Google Spreadsheet or something like this. So reading the documentation from digi. I could acces...

Installing CouchDB on IaaS?

After many many google searches I am still unclear on how to install a CouchDB instance on an Infrastructure as a Service cloud service. I have looked into DBaaS options such as Cloudant, but for initial testing I would like to host the database on an IaaS. Specifically, I am...

slurm script gives “command not found”

I am trying to submit a script to slurm that runs m4 on an input file. m4 is installed on our cluster, and if I run the script by itself, everything works as expected. But when I submit a run to slurm via a slurm script, I get an error....

Alternate to run window service in Azure cloud

We currently have a window service which send some notification emails to users after doing some processing on database(SQL database). Runs once in day. We want to move this on azure cloud. One alternate is to put it on Azure VM as is. but I am finding some other best...

Editing /etc/hosts for MPI cluster

I struggled trying to set up an MPI cluster, following Setting Up an MPICH2 Cluster in Ubuntu tutorial. However, I tangled things up and it did not work, so I undid all the changes (except of the passphrase in step 7, which I have no clue how to undo) and...

Identifying duplicate requests fired from back end

I am facing a use case where I need to track down duplicate requests, which are fired through REST API calls from back end. Each request writes into the database, and hence the duplicate requests need not be processed again. The duplicate requests may come in different threads under the...

Relation replacement in Parse Cloud

In a Parse Cloud function I need to replace the value of a relation. How do I do that? Here is my code: Parse.Cloud.define (“myCloudFunction”, function(request, response) { // Code to get myObject ……. // ……… // Now I have myObject in the hand. myObject.relation("author").add(NewAuthor); }); If I keep this...

HornetQ using JNDI / remoting when used with WildFly 8.1 cluster

Previously 2 JBoss 6 servers (master & slave) were started with 2 different IP addresses ( and and those 2 server instances were run in domain mode. The instances used HornetQ to communicate index jobs for Hibernate Search. I'm trying to create a similar configuration using Wildfly instead of...

how to run a static site on bitnami and AWS?

I am quite a noob when it comes to cloud based apps, this is a completely new context for me. I currently got the github education pack, part of it was an account on bitnami. I was wondering if it's possible to host a static html site on bitnami, I...

How to get Office 365 service location via API

I am trying to find the region of an Office 365 service. Microsoft's Office 365 documentation states that the data storage regions are disclosed to the customer. Is it possible to find a service's region via an API request? I've been looking through documentation on the Service Communications API, the...

Continuous Query Processing with baseline indexing in Endeca

I wanted to whether if it is possible to have Continuous Query Processing with baseline indexing in Endeca cluster environment. I have an Endeca central server and multiple Endeca agent only instances. If I run Baseline Indexing in the central server, it re-indexes the records in all the MDEX engines(all...

Authenticate Google Account with php

Is there a possibility to log in on to Google Accountt with php? Until 26th of May I was using to verify account. Now this method is disabled. What Im trying to do is to connect Google Printer Cloud to print orders in resaturant. Client makes order on website...

Barrier after MPI non-blocking call, without bookkeeping?

I'm doing a bunch of MPI_Iallreduce non-blocking communications. I've added these Iallreduce calls to several different places in my code. Every so often, I want to pause and wait for all the Iallreduce calls to finish. Version 1 with MPI_Request bookkeeping -- this works: MPI_Request requests[]; MPI_Iallreduce(..., requests[0]); ... MPI_Iallreduce(...,...

Adding a second Heroku app to my local repo, I'm denied access to my apps

I originally had a setup like: git remote -v origin [email protected]: (fetch) origin [email protected]: (push) stage [email protected]: (fetch) stage [email protected]: (push) where I could push and pull from stage without an issue and run commands like heroku run console --app stage however, I added a production app locally leaving me...

Did google change/update related to vm ssl connections to Cloud SQL?

Google Cloud SQL down? At 1am ET my VM connection to Cloud SQL over SSL decided to stop connecting and will not allow a connection. Did google make changes/updates? Has been working fine for months, no changes, then all of a sudden this morning at 1am ET BAM! Won't connect...

Spark - How to run a standalone cluster locally

Is there the possibility to run the Spark standalone cluster locally on just one machine (which is basically different from just developing jobs locally (i.e., local[*]))?. So far I am running 2 different VMs to build a cluster, what if I could run a standalone cluster on the very same...

How to deploy three config server instances for sharding in mongodb?

I am a newbie in MongoDB. Also I have not much knowledge in networking and servers. I am trying to deploy sharded cluster in mongodb using this article. It says I need to deploy three configuration server instances and create data directories for each. I can create a data directory...

issue monitoring hadoop response

I am using ganglia to monitor Hadoop. gmond and gmetad are running fine. When I telnet on gmond port (8649) and when I telnet gmetad on its xml answer port, I get no hadoop data. How can it be ? cluster { name = "my cluster" owner = "Master" latlong...

When upload Bitmap stream to Azure storage, it store zero/empty image,azure,cloud
I am using below code to upload a MemoryStream from a bitmap image to my Microsoft azure storage account: MemoryStream memoryStream = new MemoryStream(); img.Save(memoryStream, System.Drawing.Imaging.ImageFormat.Png); blob.Properties.ContentType = model.File.ContentType; blob.UploadFromStream(memoryStream); What happen by use above code is it uploads an empty image to Azure storage :( !(I found the name...

Writing an IAM policy and CORS configuration for Amazon S3

I am very new to all this but have been able to get an avatar/image uploader to work in my Rails application. A user can upload a new avatar to my S3 bucket and the avatar shows within the web application. To this end, I've had to grant "AmazonS3FullAccess" policy...

How to create a secure git repository in the cloud?

I would like to create a git repository and keep a backup copy of it in the cloud. Creating a remote repository would be a simple solution, but the code would be open to anyone with access to the cloud machine, thus compromised. Currently I manage this git repository in...

viewing google app engine Python logging messages in CodeEnvy

I'm trying to move my GAE development to the cloud. So far Codeenvy has the richest toolset however I'm struggling with one small issue. when I use the python logging library I don't know where to view these messages! def post(self): self.response.write('Processing form data...') feedback = self.request.get('content') I assumed...

Running python script on Microsoft Azure

I'll have a linux machine with a virtual machine installed for Microsoft azure soon. I need to run some data mining/graph analysis algorithms on the azure because I work with big data. I don't want to use azure machine learning stuff. just want to run my own python code. What...

Liferay 6.2 clustering issue with multicast

I am trying to cluster ehcache and lucene with Liferay 6.2 EE sp2 bundle on 2 servers with mutlicast enabled. WE have Apache HTTPD servers fronting tomcat servers using reverse proxy. A valid 6.2 license is deployed on both the nodes. We user the following properties in the

Grunt, cloud deployment and compiled files

I've recently started exploring grunt and I've done so by trying out the seed. It uses a gruntfile to copy static assets, css, javascript to a destination folder with this code: dist_js: { files: [{ src: ['<%= app_files.js %>', '<%= vendor_files.js %>'], dest: '<%= dist_dir %>/public', cwd: '.', expand:...

bash: /usr/bin/hydra_pmi_proxy: No such file or directory

I am struggling to set up an MPI cluster, following the Setting Up an MPICH2 Cluster in Ubuntu tutorial. I have something running and my machine file is this: pythagoras:2 # this will spawn 2 processes on pythagoras geomcomp # this will spawn 1 process on geomcomp The tutorial states:...

Minimum number of config servers needed for Mongo Cluster

I am trying to test sharding in MongoDB cluster. I am folllowing MongoDB documentation . They suggest to use 3 config servers. Can I set up a cluster with 1 or 2 servers? Or I need 3 or more config servers?...

sun grid engine qsub to all nodes

I have a master and two nodes. They are install with SGN. And I have a shell script ready on all the nodes as well. Now I want to use a qsub to submit the job on all my nodes. I used: qsub -V -b n -cwd /root/ but it...

What's a Cluster / Bucket in couchbase Server

I'm new to Couchbase and NoSql technologies in general, but I'm working on a web chat application running on node js using express and some other modules. I've chosen to work with NoSql to store sessions and all needed data on server-side. But I don't really understand some important features...

How to conditionally check the state of an openstack instance

I am trying to conditionally check the state of an openstack cloud instance in my playbook. The playbook itself takes the name of a cloud instance as a parameter and then deletes it by setting the state to absent using nova compute. What I want to do is check if...

How to hunt down obscure HA clustering bug in Wildfly 8.2.0.Final

The setup I have a Wildfly 8.2.0.Final application server running a cluster in domain mode using the full-ha profile. The cluster consists of two instances of wildfly, master and slave, each running on its own virtual machine. The application My project is deployed as a war-file on the application server....

How scale up/out mechanism works in OpenStack and Cloudfoundry integration?

I just started investigating OpenStack & Cloudfoundry recently, and I just wonder how scale up/out mechanism works in OpenStack & Cloudfoundry integration. As far as I understand, if we need to scale-up a web app, Cloudfoundry will try to create a new container to run new web app. If it...

Reduce Storage, By Juggling Packets in Network?

I want to preface this by saying, I have a feeling that this idea will not work the way I'm imagining, but I'm sure why. Its likely I'm making some sort of false assumption about the way the internet works. Lets say server A has a file of size 1024...

What does (extra requests) mean in services?,notifications,push-notification,cloud,push
In Parse's pricing FAQ it is mentioned that "If your app hits its request limit, your extra requests will begin to fail with error code 155 (RequestLimitExceeded)". What does (extra requests) mean? Are they the requests made within the same minute? day? month? or all other requests made until the...

Memsql, Is it possible to add more leaf nodes on the same host machine?

I installed a functional test with Memsql on a machine with an aggregation node on port 3307 and a leaf node 3306. As queries being ran, there might be additional cpu available. Hence, is it possible to add more leaf nodes on the same machine to take advantages of available...