{"id":637,"date":"2019-04-19T15:42:14","date_gmt":"2019-04-19T13:42:14","guid":{"rendered":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/"},"modified":"2021-03-29T17:11:46","modified_gmt":"2021-03-29T15:11:46","slug":"how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services","status":"publish","type":"post","link":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/","title":{"rendered":"How to create a MongoDB ReplicaSet with Self Healing using AWS services"},"content":{"rendered":"<section class=\"section section--body section--first\">\n<div class=\"section-content\">\n<div class=\"section-inner sectionLayout--insetColumn\">\n<p id=\"fa66\" class=\"graf graf--p graf-after--figure graf--trailing\">MongoDB is a document-based\u00a0<strong class=\"markup--strong markup--p-strong\">NoSQL database<\/strong>\u00a0for seamless configuration and use. It is wide-spread across new development projects. In this article, we will see how\u00a0<strong class=\"markup--strong markup--p-strong\">creating a self-healing system for Mongo clusters on AWS prevents possible downtime.<\/strong><\/p>\n<\/div>\n<\/div>\n<\/section>\n<section class=\"section section--body\">\n<div class=\"section-divider\"><\/div>\n<div class=\"section-content\">\n<div class=\"section-inner sectionLayout--insetColumn\">\n<p id=\"3eb2\" class=\"graf graf--p graf--leading\">MongoDB allows for creating a cluster distributed in high reliability through ReplicaSets and Sharding:<\/p>\n<ul class=\"postList\">\n<li id=\"469f\" class=\"graf graf--li graf-after--p\"><strong class=\"markup--strong markup--li-strong\">ReplicaSets<\/strong>\u00a0are sets of identical nodes on which database-entered JSON documents are replicated.<\/li>\n<li id=\"85af\" class=\"graf graf--li graf-after--li\"><strong class=\"markup--strong markup--li-strong\">Sharding<\/strong>\u00a0allows for writing each set of JSON documents (Collections) on a different server. This ensures better throughput in both reading and writing.<\/li>\n<\/ul>\n<p id=\"c1c6\" class=\"graf graf--p graf-after--li\"><strong class=\"markup--strong markup--p-strong\">Sharding configuration is intolerant to failures.\u00a0<\/strong>In fact, its Collections become inaccessible in the event of a machine down. To overcome this obstacle, we opt for a mixed configuration based on Sharding and Replication. Here, each sharding node is actually a replica set.<\/p>\n<p id=\"5f71\" class=\"graf graf--p graf-after--p\">Each ReplicaSet is composed of a\u00a0<em class=\"markup--em markup--p-em\">Master<\/em>\u00a0that takes care of writing and reading, and at least two Replicas that do not accept writing directly but are kept synchronized with the data in the Master and can only perform reading operations. Every two seconds, each node communicates its status to the others through a heartbeat system.\u00a0<strong class=\"markup--strong markup--p-strong\">In the case of master failure, the secondary nodes vote to elect the new Master.<\/strong><\/p>\n<p id=\"df01\" class=\"graf graf--p graf-after--p\">However, for correct voting, more than 50% of the cluster nodes must participate in the vote.<\/p>\n<p id=\"4a59\" class=\"graf graf--p graf-after--p\">Therefore, for deployment on AWS, it is advised to distribute EC2 equally on at least three Availability Zones (AZ). This is necessary if you want the ReplicaSet to remain functional, even in the occurrence of an AZ down.<\/p>\n<p id=\"26cb\" class=\"graf graf--p graf-after--p\"><strong class=\"markup--strong markup--p-strong\">A configuration with three nodes<\/strong>\u00a0is often used for the set replication in databases that are small and not\u00a0<em class=\"markup--em markup--p-em\">read intensive,<\/em>\u00a0or for databases with Sharding. In this configuration, if a node fails, two more are always available for voting. One of these will be promoted to master.<\/p>\n<p id=\"c938\" class=\"graf graf--p graf-after--p\">However, especially under heavy traffic, the failure of a node must be treated as an emergency and solved as soon as possible. Any further failure, in fact, would result in a service interruption.<\/p>\n<p id=\"9400\" class=\"graf graf--p graf-after--p\">We have implemented a\u00a0<strong class=\"markup--strong markup--p-strong\">MongoDB self-healing system<\/strong>\u00a0to automatically resolve this type of situation. It lets DB managers complete the post-mortem analysis of the failed MongoDB instance and\u00a0<strong class=\"markup--strong markup--p-strong\">automatically recreate an instance<\/strong>\u00a0identical to the one no longer accessible. They can then reconfigure the replica set to use the new EC2 instead of the one that is corrupted.<\/p>\n<p id=\"d90b\" class=\"graf graf--p graf-after--p\">AWS implementation tools for this solution are as follows:<\/p>\n<ul class=\"postList\">\n<li id=\"6784\" class=\"graf graf--li graf-after--p\"><strong class=\"markup--strong markup--li-strong\">Autoscaling group<\/strong>: an AWS service to manage and scale groups of instances. One can define the instance type and the number of instances to be used. AWS will launch them, monitor their status, and terminate them if necessary, as well as launch new ones.<\/li>\n<li id=\"bdcd\" class=\"graf graf--li graf-after--li\"><strong class=\"markup--strong markup--li-strong\">Route 53:<\/strong>\u00a0AWS DNS service<\/li>\n<li id=\"07cc\" class=\"graf graf--li graf-after--li\"><strong class=\"markup--strong markup--li-strong\">Lambda Functions:<\/strong>\u00a0Server-less functions launched on AWS as a response to a trigger or according to an AWS CloudWatch set the pace.<\/li>\n<li id=\"cd82\" class=\"graf graf--li graf-after--li\"><strong class=\"markup--strong markup--li-strong\">SQS Queues:<\/strong>\u00a0AWS-managed highly-reliable message queues.<\/li>\n<li id=\"e3fe\" class=\"graf graf--li graf-after--li\"><strong class=\"markup--strong markup--li-strong\">AWS CodeDeploy:<\/strong>\u00a0AWS-managed service for arbitrary code deployment through a machine-installed agent that integrates with autoscaling groups. Deployment is performed when a new instance is started.<\/li>\n<\/ul>\n<figure id=\"57b5\" class=\"graf graf--figure graf-after--li\">\n<div class=\"aspectRatioPlaceholder is-locked\">\n<div class=\"aspectRatioPlaceholder-fill\"><\/div>\n<div class=\"progressiveMedia js-progressiveMedia graf-image is-canvasLoaded is-imageLoaded\" data-image-id=\"1*w9JzK-Url3nMxKAcSpv9CA.png\" data-width=\"3840\" data-height=\"2160\" data-action=\"zoom\" data-action-value=\"1*w9JzK-Url3nMxKAcSpv9CA.png\" data-scroll=\"native\"><canvas class=\"progressiveMedia-canvas js-progressiveMedia-canvas\" width=\"75\" height=\"40\"><\/canvas><img decoding=\"async\" class=\"progressiveMedia-image js-progressiveMedia-image\" src=\"https:\/\/cdn-images-1.medium.com\/max\/800\/1*w9JzK-Url3nMxKAcSpv9CA.png\" data-src=\"https:\/\/cdn-images-1.medium.com\/max\/800\/1*w9JzK-Url3nMxKAcSpv9CA.png\" \/><\/div>\n<\/div>\n<\/figure>\n<h3 id=\"96dc\" class=\"graf graf--h3 graf-after--figure\">Creation of the ReplicaSet<\/h3>\n<p id=\"0729\" class=\"graf graf--p graf-after--h3\">To proceed with the creation of the ReplicaSet with self-healing, it is first of all necessary to\u00a0<strong class=\"markup--strong markup--p-strong\">create a basic AMI<\/strong>\u00a0for the replication machines set with all the necessary packages installed: AWSCLI, Chrony, Ruby, and obviously MongoDB (master and client) in the\u00a0<a class=\"markup--anchor markup--p-anchor\" href=\"https:\/\/docs.mongodb.com\/manual\/installation\/\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" data-href=\"https:\/\/docs.mongodb.com\/manual\/installation\/\">desired version<\/a>. It is also necessary to\u00a0<a class=\"markup--anchor markup--p-anchor\" href=\"https:\/\/docs.aws.amazon.com\/codedeploy\/latest\/userguide\/codedeploy-agent-operations-install.html\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" data-href=\"https:\/\/docs.aws.amazon.com\/codedeploy\/latest\/userguide\/codedeploy-agent-operations-install.html\">install the CodeDeploy agent<\/a>.<\/p>\n<p id=\"d9c3\" class=\"graf graf--p graf-after--p\">In order to configure the replica set, create and install a\u00a0<strong class=\"markup--strong markup--p-strong\">keyfile<\/strong>\u00a0on every cluster machine. Mongo will use it to authenticate the heartbeat connections. The file must contain alphanumeric characters generated in a pseudo-random way. It can be created with the command:<\/p>\n<pre id=\"a5b3\" class=\"graf graf--pre graf-after--p\">openssl rand -base64 756 &gt; &lt;path-to-keyfile&gt;\r\nchmod 400 &lt;path-to-keyfile&gt;<\/pre>\n<p id=\"4756\" class=\"graf graf--p graf-after--pre\">The keyfile is a minimal form of security and is acceptable only on a secure network (private subnet in VPC). However, for the production environment,\u00a0<strong class=\"markup--strong markup--p-strong\">using a x.509 certificate is preferable.<\/strong>\u00a0The MongoDB configuration file will look similar to this:<\/p>\n<pre id=\"cdfa\" class=\"graf graf--pre graf-after--p\">storage:\r\n  dbPath: \/mnt\/mongo\/data\r\n  journal:\r\n    enabled: true<\/pre>\n<pre id=\"8d78\" class=\"graf graf--pre graf-after--pre\">systemLog:\r\n  destination: file\r\n  logAppend: true\r\n  path: &lt;logs-path&gt;<\/pre>\n<pre id=\"34c8\" class=\"graf graf--pre graf-after--pre\">security:\r\n  keyFile: &lt;keyfile-path&gt;<\/pre>\n<pre id=\"0370\" class=\"graf graf--pre graf-after--pre\">replication:\r\n  replSetName: test-rs<\/pre>\n<pre id=\"af86\" class=\"graf graf--pre graf-after--pre\">net:\r\n  port: 27017\r\n  bindIpAll: true<\/pre>\n<p id=\"53c6\" class=\"graf graf--p graf-after--pre\">The initial creation of the MongoDB cluster can be fully automated using CloudFormation. It can also be performed manually using the just-created AMI and a configuration script to run on the cluster\u2019s individual machines.<\/p>\n<p id=\"5f55\" class=\"graf graf--p graf-after--p\">First, one of the machines must be configured to be the master and initialize the ReplicaSet. Therefore, first start Mongo without the security and replication fields in the configuration file, then create an admin user with the command:<\/p>\n<pre id=\"1c80\" class=\"graf graf--pre graf-after--p\">db.createUser({user: \"&lt;ADMIN_USER&gt;\", pwd: \"&lt;PWD&gt;\", roles:[{role: \"root\", db: \"admin\"}]});<\/pre>\n<p id=\"643a\" class=\"graf graf--p graf-after--pre\">Then, restart Mongo with the configuration file containing security and replication.<\/p>\n<p id=\"9aeb\" class=\"graf graf--p graf-after--p\">After having turned on the other machines, which will initially be replicas, it is necessary to\u00a0<strong class=\"markup--strong markup--p-strong\">create A names<\/strong>\u00a0on a private Route53 hosted zone for all machines in the replica set. In VPC configurations, it is also necessary to select the enableDnsHostnames and enableDnsSupport options for activating the resolution of the R53 internal DNS.<\/p>\n<p id=\"233b\" class=\"graf graf--p graf-after--p\">Having done that, it is finally possible to connect in SSH to the master and activate the replica set through Mongo CLI with the command:<\/p>\n<pre id=\"4370\" class=\"graf graf--pre graf-after--p\">rs.initiate( {\r\n   _id : \"test-mongo-rs\",\r\n   members: [\r\n      { _id: 0, host: \"mongodb-1.test.it:27017\" },\r\n      { _id: 1, host: \"mongodb-2.test.it:27017\" },\r\n      { _id: 2, host: \"mongodb-3.test.it:27017\" }\r\n   ]\r\n});<\/pre>\n<p id=\"3ca1\" class=\"graf graf--p graf-after--pre\">Now that the\u00a0<strong class=\"markup--strong markup--p-strong\">ReplicaSet is set up and working<\/strong>, let us focus on the self-healing.<\/p>\n<h3 id=\"d5a7\" class=\"graf graf--h3 graf-after--p\">Self Healing\u00a0Setup<\/h3>\n<p id=\"01ca\" class=\"graf graf--p graf-after--h3\">To configure the self-healing, first, register the machines to an\u00a0<strong class=\"markup--strong markup--p-strong\">autoscaling group with EC2-type health check.<\/strong>\u00a0This needs to feature a fixed number of machines (Max: 3, Min: 3, and Desired: 3). This way, in case of a machine failure, AWS will create a new instance from an AMI defined in the launch template of the autoscaling group. Next is the creation of a\u00a0<strong class=\"markup--strong markup--p-strong\">Lambda function that acts as a \u201chealth check\u201d for the ReplicaSet machines.<\/strong>\u00a0In our case, we chose to create a lambda in python with Pymongo installed as a\u00a0<a class=\"markup--anchor markup--p-anchor\" href=\"https:\/\/docs.aws.amazon.com\/lambda\/latest\/dg\/configuration-layers.html\" target=\"_blank\" rel=\"nofollow noopener noreferrer\" data-href=\"https:\/\/docs.aws.amazon.com\/lambda\/latest\/dg\/configuration-layers.html\">Lambda Layer<\/a>. The lambda function connects to the ReplicaSet via Pymongo. It monitors the status of all nodes via the replSetGetStatus command, offering an exploded view of the ReplicaSet status:<\/p>\n<pre id=\"3187\" class=\"graf graf--pre graf-after--p\">import boto3\r\nfrom pymongo import MongoClient\r\nsqs = boto3.client('sqs')\r\nmongoserver_uri = os.environ['MONGO_URI']\r\nhostzone = os.environ['ZONE_ID']\r\nmongo_connection = MongoClient(mongoserver_uri)\r\ndns_aws = boto3.client('route53')\r\nec2 = boto3.resource('ec2')<\/pre>\n<pre id=\"7bde\" class=\"graf graf--pre graf-after--pre\">rs_status = mongo_connection.admin.command('replSetGetStatus')<\/pre>\n<p id=\"26db\" class=\"graf graf--p graf-after--pre\">The \u201chealth\u201d value of all non-functioning nodes will be: 0. To avoid false positives, we have configured the code to repeat the check several times with a one-minute interval. If the node does not return healthy in this interval, then the lambda writes a message on an SQS queue containing the DNS name of the failed machine and a timestamp.<\/p>\n<p id=\"727c\" class=\"graf graf--p graf-after--p\">Once complete, the code will de-register the failed machine from the autoscaling group using boto3, if it has not already been terminated:<\/p>\n<pre id=\"7b53\" class=\"graf graf--pre graf-after--p\">autoscaling = boto3.client('autoscaling')<\/pre>\n<pre id=\"6fb4\" class=\"graf graf--pre graf-after--pre\">response = autoscaling.detach_instances(\r\n    InstanceIds=[\r\n        'string',\r\n    ],\r\n    AutoScalingGroupName='mongo-autoscaling',\r\n    ShouldDecrementDesiredCapacity=False\r\n)<\/pre>\n<p id=\"31ae\" class=\"graf graf--p graf-after--pre\"><strong class=\"markup--strong markup--p-strong\">The lambda is configured to avoid multiple executions.<\/strong>\u00a0By sending the message on SQS, the instance is deregistered only if the SQS queue is empty (ApproximateNumberOfMessages = 0).<\/p>\n<p id=\"6ce5\" class=\"graf graf--p graf-after--p\">The lambda preparation is followed by the last step:\u00a0<strong class=\"markup--strong markup--p-strong\">the CodeDeploy configuration.<\/strong><\/p>\n<p id=\"e066\" class=\"graf graf--p graf-after--p\">This has the task of configuring the new instances when the autoscaling group is turned on. Let us start by creating a new application in the CodeDeploy dashboard and a new deployment group targeting the newly created autoscaling group. CodeDeploy runs a series of user-configurable script operations on the machine. It is defined in a file called\u00a0<em class=\"markup--em markup--p-em\">appspec.yml<\/em>\u00a0that must be saved in an S3 bucket together with the scripts and files to be transferred and\/or executed on the machine.<\/p>\n<p id=\"3a58\" class=\"graf graf--p graf-after--p\">The structure is as follows:<\/p>\n<pre id=\"9dd8\" class=\"graf graf--pre graf-after--p\">version: 0.0\r\nos: linux\r\nfiles:\r\n  - source: \/content\r\n    destination: \/home\/ubuntu<\/pre>\n<pre id=\"eaa4\" class=\"graf graf--pre graf-after--pre\">hooks:\r\n  ApplicationStop:\r\n    - location: scripts\/applicationstop.sh\r\n      timeout: 300\r\n      runas: root<\/pre>\n<pre id=\"a269\" class=\"graf graf--pre graf-after--pre\">  BeforeInstall:\r\n    - location: scripts\/beforeinstall.sh\r\n      timeout: 300\r\n      runas: root<\/pre>\n<pre id=\"ad9b\" class=\"graf graf--pre graf-after--pre\">  AfterInstall:\r\n    - location: scripts\/afterinstall.sh\r\n      timeout: 300\r\n      runas: root<\/pre>\n<pre id=\"e859\" class=\"graf graf--pre graf-after--pre\">  ApplicationStart:\r\n    - location: scripts\/applicationstart.sh\r\n      timeout: 300\r\n      runas: root<\/pre>\n<pre id=\"fa92\" class=\"graf graf--pre graf-after--pre\">  ValidateService:\r\n    - location: scripts\/validateservice.sh\r\n      timeout: 300\r\n      runas: root<\/pre>\n<p id=\"23b5\" class=\"graf graf--p graf-after--pre\">The folder structure to upload on s3 for CodeDeploy input is as follows:<\/p>\n<pre id=\"86e5\" class=\"graf graf--pre graf-after--p\">content\/\r\n    mongod.conf\r\n    mongo-keyfile\r\nscripts\/\r\n    afterinstall.sh\r\n    applicationstart.sh\r\n    applicationstop.sh\r\n    beforeinstall.sh\r\n    validateservice.sh<\/pre>\n<p id=\"ce61\" class=\"graf graf--p graf-after--pre\">In our case, the content folder contains the keyhole and the Mongo configuration file, whereas the scripts folder contains the bash scripts that deal with configuring the new instance. In particular, the scripts must read the message containing the DNS name of the failed machine from the SQS queue; configure Mongo on the recently turned-on machine through MongoDB CLI; reconfigure the replica set to use the new node, and finally start the Mongo service.<\/p>\n<p id=\"9ac2\" class=\"graf graf--p graf-after--p\">The following chain of events is triggered as soon as an instance fails (or the Mongo process becomes unresponsive):<\/p>\n<ul class=\"postList\">\n<li id=\"86e1\" class=\"graf graf--li graf-after--p\">The\u00a0<strong class=\"markup--strong markup--li-strong\">health check lambda detects the node as unhealthy<\/strong>. It detaches the machine from the autoscaling group and writes a message on an SQS queue containing the failed node\u2019s DNS name.<\/li>\n<li id=\"72d8\" class=\"graf graf--li graf-after--li\">The\u00a0<strong class=\"markup--strong markup--li-strong\">autoscaling group reacts to the absence of a machine<\/strong>\u00a0by starting a new ec2.<\/li>\n<li id=\"c40a\" class=\"graf graf--li graf-after--li\"><strong class=\"markup--strong markup--li-strong\">CodeDeploy executes the configuration scripts saved on S3.<\/strong>\u00a0These configure the new machine so that they take the place of the failed one.<\/li>\n<\/ul>\n<p id=\"0afb\" class=\"graf graf--p graf-after--li\">At this point,\u00a0<strong class=\"markup--strong markup--p-strong\">MongoDB starts replicating the data on the new machine. <\/strong>However, for large databases, this operation can take several hours. It is advised to configure MongoDB in order to avoid this problem, i.e., by saving data on a specific EBS. During the creation of the new instance, dismount the EBS from the failed instance and mount it on the newly created instance.<\/p>\n<p id=\"99f9\" class=\"graf graf--p graf-after--p graf--trailing\">Once all configurations are complete,\u00a0<strong class=\"markup--strong markup--p-strong\">our replica set in high reliability and with automatic self-healing is ready!<\/strong><\/p>\n<\/div>\n<\/div>\n<\/section>\n<section class=\"section section--body section--last\">\n<div class=\"section-divider\"><\/div>\n<div class=\"section-content\">\n<div class=\"section-inner sectionLayout--insetColumn\">\n<p id=\"df0f\" class=\"graf graf--p graf--leading graf--trailing\">Would you like to share your observations, doubts, or ideas with us? Would you like some help in creating a configuration like this? \u2026 then\u00a0<strong class=\"markup--strong markup--p-strong\">our team looks forward to discussing this with you: <a href=\"https:\/\/www.besharp.it\/en\/contact-us\/\">contact us!<\/a><\/strong><\/p>\n<\/div>\n<\/div>\n<\/section>\n","protected":false},"excerpt":{"rendered":"<p>MongoDB is a document-based\u00a0NoSQL database\u00a0for seamless configuration and use. It is wide-spread across new development projects. In this article, we [&hellip;]<\/p>\n","protected":false},"author":9,"featured_media":641,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[482],"tags":[370],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v21.9.1 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>How to create a MongoDB ReplicaSet with Self Healing using AWS services - Proud2beCloud Blog<\/title>\n<meta name=\"description\" content=\"We have implemented a\u00a0MongoDB self-healing system\u00a0to automatically resolve nodes failures and avoid service interruption\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"A MongoDB ReplicaSet with Self Healing using AWS services\" \/>\n<meta property=\"og:description\" content=\"We have implemented a\u00a0MongoDB self-healing system\u00a0to avoid service interruption\" \/>\n<meta property=\"og:url\" content=\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/\" \/>\n<meta property=\"og:site_name\" content=\"Proud2beCloud Blog\" \/>\n<meta property=\"article:published_time\" content=\"2019-04-19T13:42:14+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2021-03-29T15:11:46+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/blog.besharp.it\/wp-content\/uploads\/2019\/04\/mongodbblog.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1721\" \/>\n\t<meta property=\"og:image:height\" content=\"1290\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Matteo Moroni\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:title\" content=\"A MongoDB ReplicaSet with Self Healing using AWS services\" \/>\n<meta name=\"twitter:description\" content=\"We have implemented a\u00a0MongoDB self-healing system\u00a0to avoid service interruption\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Matteo Moroni\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"8 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/\",\"url\":\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/\",\"name\":\"How to create a MongoDB ReplicaSet with Self Healing using AWS services - Proud2beCloud Blog\",\"isPartOf\":{\"@id\":\"https:\/\/blog.besharp.it\/#website\"},\"datePublished\":\"2019-04-19T13:42:14+00:00\",\"dateModified\":\"2021-03-29T15:11:46+00:00\",\"author\":{\"@id\":\"https:\/\/blog.besharp.it\/#\/schema\/person\/0b3e69eb2dcb125d58476b906ec1c7bc\"},\"description\":\"We have implemented a\u00a0MongoDB self-healing system\u00a0to automatically resolve nodes failures and avoid service interruption\",\"breadcrumb\":{\"@id\":\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/blog.besharp.it\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"How to create a MongoDB ReplicaSet with Self Healing using AWS services\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/blog.besharp.it\/#website\",\"url\":\"https:\/\/blog.besharp.it\/\",\"name\":\"Proud2beCloud Blog\",\"description\":\"il blog di beSharp\",\"alternateName\":\"Proud2beCloud Blog\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/blog.besharp.it\/?s={search_term_string}\"},\"query-input\":\"required name=search_term_string\"}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/blog.besharp.it\/#\/schema\/person\/0b3e69eb2dcb125d58476b906ec1c7bc\",\"name\":\"Matteo Moroni\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/blog.besharp.it\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/acad790b9bb4c6d62e076ecdc1debb35?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/acad790b9bb4c6d62e076ecdc1debb35?s=96&d=mm&r=g\",\"caption\":\"Matteo Moroni\"},\"description\":\"DevOps e Solution Architect di beSharp, mi occupo di sviluppare soluzioni Saas, Data Analysis, HPC e di progettare architetture non convenzionali a complessit\u00e0 divergente. Appassionato di informatica e fisica, da sempre lavoro nella prima e ho un PhD nella seconda. Parlare di tutto ci\u00f2 che \u00e8 tecnico e nerd mi rende felice!\",\"url\":\"https:\/\/blog.besharp.it\/author\/matteo-moroni\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"How to create a MongoDB ReplicaSet with Self Healing using AWS services - Proud2beCloud Blog","description":"We have implemented a\u00a0MongoDB self-healing system\u00a0to automatically resolve nodes failures and avoid service interruption","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/","og_locale":"en_US","og_type":"article","og_title":"A MongoDB ReplicaSet with Self Healing using AWS services","og_description":"We have implemented a\u00a0MongoDB self-healing system\u00a0to avoid service interruption","og_url":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/","og_site_name":"Proud2beCloud Blog","article_published_time":"2019-04-19T13:42:14+00:00","article_modified_time":"2021-03-29T15:11:46+00:00","og_image":[{"width":1721,"height":1290,"url":"https:\/\/blog.besharp.it\/wp-content\/uploads\/2019\/04\/mongodbblog.png","type":"image\/png"}],"author":"Matteo Moroni","twitter_card":"summary_large_image","twitter_title":"A MongoDB ReplicaSet with Self Healing using AWS services","twitter_description":"We have implemented a\u00a0MongoDB self-healing system\u00a0to avoid service interruption","twitter_misc":{"Written by":"Matteo Moroni","Est. reading time":"8 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/","url":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/","name":"How to create a MongoDB ReplicaSet with Self Healing using AWS services - Proud2beCloud Blog","isPartOf":{"@id":"https:\/\/blog.besharp.it\/#website"},"datePublished":"2019-04-19T13:42:14+00:00","dateModified":"2021-03-29T15:11:46+00:00","author":{"@id":"https:\/\/blog.besharp.it\/#\/schema\/person\/0b3e69eb2dcb125d58476b906ec1c7bc"},"description":"We have implemented a\u00a0MongoDB self-healing system\u00a0to automatically resolve nodes failures and avoid service interruption","breadcrumb":{"@id":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/blog.besharp.it\/how-to-create-a-mongodb-replicaset-with-self-healing-using-aws-services\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/blog.besharp.it\/"},{"@type":"ListItem","position":2,"name":"How to create a MongoDB ReplicaSet with Self Healing using AWS services"}]},{"@type":"WebSite","@id":"https:\/\/blog.besharp.it\/#website","url":"https:\/\/blog.besharp.it\/","name":"Proud2beCloud Blog","description":"il blog di beSharp","alternateName":"Proud2beCloud Blog","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/blog.besharp.it\/?s={search_term_string}"},"query-input":"required name=search_term_string"}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/blog.besharp.it\/#\/schema\/person\/0b3e69eb2dcb125d58476b906ec1c7bc","name":"Matteo Moroni","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/blog.besharp.it\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/acad790b9bb4c6d62e076ecdc1debb35?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/acad790b9bb4c6d62e076ecdc1debb35?s=96&d=mm&r=g","caption":"Matteo Moroni"},"description":"DevOps e Solution Architect di beSharp, mi occupo di sviluppare soluzioni Saas, Data Analysis, HPC e di progettare architetture non convenzionali a complessit\u00e0 divergente. Appassionato di informatica e fisica, da sempre lavoro nella prima e ho un PhD nella seconda. Parlare di tutto ci\u00f2 che \u00e8 tecnico e nerd mi rende felice!","url":"https:\/\/blog.besharp.it\/author\/matteo-moroni\/"}]}},"_links":{"self":[{"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/posts\/637"}],"collection":[{"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/users\/9"}],"replies":[{"embeddable":true,"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/comments?post=637"}],"version-history":[{"count":0,"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/posts\/637\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/media\/641"}],"wp:attachment":[{"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/media?parent=637"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/categories?post=637"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blog.besharp.it\/wp-json\/wp\/v2\/tags?post=637"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}