I am trying to validate requests coming into API gateway by using the request validator to validate the body of the request. The JSON body which is expected just has one key which is "userId" and the value should be a UUID. I have setup my model like this:
{
"$schema" : "http://json-schema.org/draft-04/schema#",
"title" : "newUser",
"type" : "object",
"properties" : {
"userId" : {
"type" : "string",
"format" : "uuid"
}
}
}
After a few tests it seems to be working, it accepts a valid UUID and all of these correctly return a bad request:
{
"userId": null
}
{
"userId": "text"
}
{
"userId": 123
}
{
"userId": "8327a29c-7134-4566-8b58-"
}
{
"userId": "8327a29c-7134-4566-8b58-46bcf951ef6az"
}
However if you remove a few characters or add a couple of valid hex characters to make it an invalid length then it will pass validation and forward on the request. What is the correct way of validating UUIDs using the request validator in API gateway which actually works?
After investigating more, uuid as a format isn't explicitly defined in the OpenAPI specification. Therefore implementation of format validation is not always consistent with every system. So I think the AWS validator implementation is a little bit funky.
The cleanest solution I have thought of is using regex like this:
{
"$schema" : "http://json-schema.org/draft-04/schema#",
"title" : "newUser",
"type" : "object",
"properties" : {
"userId" : {
"type" : "string",
"format" : "uuid",
"pattern": "^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{4}-[0-9a-fA-F]{12}$"
}
}
}
I want to create a unique api key whenever a user logs into my website
I want to do this from AWS Lambda based on a trigger
I plan to do this through the RESTAPI via axios
I believe I have misunderstood the docs, It's just killing me as my submission is just around the corner
inside the function
var x = await axios.post('/apikeys',{
name : "aaaaaa",
description : "Strinaaaag",
enabled : true,
generateDistinctId : true,
value : "sasdasdasdasdsd",
stageKeys : [ {
restApiId : "Strissng",
stageName : "prod"
} ],
})
and this is not working even the one below
params = {
name : "aaaaaa",
description : "Strinaaaag",
enabled : true,
generateDistinctId : true,
value : "sasdasdasdasdsd",
stageKeys : [ {
restApiId : "Strissng",
stageName : "prod"
} ],
}
AWS.APIGateway().createApiKey(params,function(err,data){
........
})
I'd like to build a tool that would notify a user each time there was a successful or failed build on CodeDeploy through any communication medium (email, slack, etc). I've went through their documentation.. and nothing except for long-polling comes to mind. Any idea if there's some webhook option where i can register a URL and be notified?
Update on 2016-04-27
AWS officially announced this in February 2016:
You can now create triggers that send Amazon SNS notifications before, during, and after the deployment process for your applications. Triggers can be set for the deployment as a whole or for the individual instances targeted by the deployment, and are sent on both successes and failures.
Original answer
Not yet.
In this AWS forum thread, it was requested that CodeDeploy emit events so you can use Lambda to process them instead of polling for details.
The answer by AWS staff (emphasis mine):
We here on CodeDeploy agree.
Unfortunately, I can't give you an exact release date but keep an eye on our announcements, it's coming soon.
Here is a gist of an AWS Lambda function that posts a formatted CodeDeploy notification to Slack
https://gist.github.com/MrRoyce/097edc0de2fe001288be2e8633f4b22a
var services = '/services/...'; // Update this with your Slack service...
var channel = "#aws-deployments" // And this with the Slack channel
var https = require('https');
var util = require('util');
var formatFields = function(string) {
var
message = JSON.parse(string),
fields = [],
deploymentOverview;
// Make sure we have a valid response
if (message) {
fields = [
{
"title" : "Task",
"value" : message.eventTriggerName,
"short" : true
},
{
"title" : "Status",
"value" : message.status,
"short" : true
},
{
"title" : "Application",
"value" : message.applicationName,
"short" : true
},
{
"title" : "Deployment Group",
"value" : message.deploymentGroupName,
"short" : true
},
{
"title" : "Region",
"value" : message.region,
"short" : true
},
{
"title" : "Deployment Id",
"value" : message.deploymentId,
"short" : true
},
{
"title" : "Create Time",
"value" : message.createTime,
"short" : true
},
{
"title" : "Complete Time",
"value" : ((message.completeTime) ? message.completeTime : ''),
"short" : true
}
];
if (message.deploymentOverview) {
deploymentOverview = JSON.parse(message.deploymentOverview);
fields.push(
{
"title" : "Succeeded",
"value" : deploymentOverview.Succeeded,
"short" : true
},
{
"title" : "Failed",
"value" : deploymentOverview.Failed,
"short" : true
},
{
"title" : "Skipped",
"value" : deploymentOverview.Skipped,
"short" : true
},
{
"title" : "In Progress",
"value" : deploymentOverview.InProgress,
"short" : true
},
{
"title" : "Pending",
"value" : deploymentOverview.Pending,
"short" : true
}
);
}
}
return fields;
}
exports.handler = function(event, context) {
var postData = {
"channel": channel,
"username": "AWS SNS via Lamda :: CodeDeploy Status",
"text": "*" + event.Records[0].Sns.Subject + "*",
"icon_emoji": ":aws:"
};
var fields = formatFields(event.Records[0].Sns.Message);
var message = event.Records[0].Sns.Message;
var severity = "good";
var dangerMessages = [
" but with errors",
" to RED",
"During an aborted deployment",
"FAILED",
"Failed to deploy application",
"Failed to deploy configuration",
"has a dependent object",
"is not authorized to perform",
"Pending to Degraded",
"Stack deletion failed",
"Unsuccessful command execution",
"You do not have permission",
"Your quota allows for 0 more running instance"];
var warningMessages = [
" aborted operation.",
" to YELLOW",
"Adding instance ",
"Degraded to Info",
"Deleting SNS topic",
"is currently running under desired capacity",
"Ok to Info",
"Ok to Warning",
"Pending Initialization",
"Removed instance ",
"Rollback of environment"
];
for(var dangerMessagesItem in dangerMessages) {
if (message.indexOf(dangerMessages[dangerMessagesItem]) != -1) {
severity = "danger";
break;
}
}
// Only check for warning messages if necessary
if (severity == "good") {
for(var warningMessagesItem in warningMessages) {
if (message.indexOf(warningMessages[warningMessagesItem]) != -1) {
severity = "warning";
break;
}
}
}
postData.attachments = [
{
"color": severity,
"fields": fields
}
];
var options = {
method: 'POST',
hostname: 'hooks.slack.com',
port: 443,
path: services // Defined above
};
var req = https.request(options, function(res) {
res.setEncoding('utf8');
res.on('data', function (chunk) {
context.done(null);
});
});
req.on('error', function(e) {
console.log('problem with request: ' + e.message);
});
req.write(util.format("%j", postData));
req.end();
};
At a high-level you need to:
Setup an SNS topic
Create a CodeDeploy Trigger
Add sns:Publish permissions to the CodeDeploy IAM Role
Configure Slack with an Incoming Webhook
Write and configure a Lambda function to process CodeDeploy's SNS messages, construct Slack messages and send them to Slack at the Incoming Webhook
I used code similar to the gist above to setup a Lambda function for Slack notifications for CodeDeploy events. I documented the entire procedure, including screenshots over here.
I couldn't find a similar, end-to-end guide elsewhere so I hope this helps someone else who stumbles on this question.
Although there is no native solution, there is a workaround that you can employ to accomplish this. You can use lambda to trigger these events. On the AWS blog, they show on how to trigger codedeploy via lambda when you upload a file to S3 (https://blogs.aws.amazon.com/application-management/post/Tx3TPMTH0EVGA64/Automatically-Deploy-from-Amazon-S3-using-AWS-CodeDeploy). Using this same concept, you can have your lambda function listen to a error/success bucket, and modify your codedeploy package to upload a file to s3 which in turn you can use as an event trigger to send an email via SES (https://peekandpoke.wordpress.com/2015/02/26/dancing-the-lambada-with-aws-lambda-or-sending-emails-on-s3-events/) or contact a web service/page that does what you want. It might be a little hokey but it gets the job done.
I've model called "customer" which returns personal details of customer in below structure. It also returns token in meta data. How to retrieve this from model?
"customer" : {
"title" : "Mr",
"firstName": "Person",
"surname" : "surname",
"primaryTelephoneNumber" : {
"number" : "0123456789",
"type": "Home"
}
},
"meta" : {
"token" : "6e16063a-8a89-5d1a-b9cb-72d14c4c7579"
}
To get the metadata just use the metadataFor method, so in your case it would be var meta = this.store.metadataFor("customer");
I am using the django-rest-framework api for the first time. Here's my question:
I need to design a database in which there are two tables:
Server => To save the server information such as ip address and server name
id: INT, name: VARCHAR, ip_address: VARCHAR
Deploy => The deploys on the server including the deploy date and a comment message
id: INT, server_id: FK to Server, deploy_date: DATETIME, message: VARCHAR
I am asked to keep track of the deploy information and design the following APIs:
get /servers/ => get all the server information with the latest deploy on that server
Example:
[
{
"id" : 1,
"name" : "qa-001",
"ip_address" : "192.168.1.1",
"deploy" :
{
"id" : 1,
"deploy_date" : "2013-09-09 12:00:00",
"message" : "test new api"
}
},
{
"id" : 2,
"name" : "qa-002",
"ip_address" : "192.168.1.2",
"deploy" :
{
"id" : 2,
"deploy_date" : "2013-09-10 12:00:00",
"message" : "test second message"
}
}
]
get /deploys/ => get all the deploy information
Example:
[
{
"id" : 1,
"deploy_date" : "2013-09-09 12:00:00",
"message" : "test new api",
"server_id" : 1
},
{
"id" : 2,
"deploy_date" : "2013-09-10 12:00:00",
"message" : "test second message",
"server_id" : 2
},
{
"id" : 3,
"deploy_date" : "2013-09-08 12:00:00",
"message" : "test new api",
"server_id" : 1
}
]
// Deploy 3 does not show up in the get servers return json
// because deploy 1 was deployed on the same server but later than deploy 3.
POST /deploys/ => To insert a new deploy
POST /servers/ => To insert a new server
...
I have played around with django-rest-framework tutorial code and read about different api documentations but couldn't figure out how to implement the features that I list above.
Please let me know if you have any ideas on how to implement this or if you think another database design would fit this requirement more.
Thanks!
Found this question which is very similar to mine.
How can I apply a filter to a nested resource in Django REST framework?
And I solved my problem by using the SerializerMethodField