nodejs - aws lambda random timeout when trying to access S3.getBucketLocation - node.js

Question

For the following code, I was asked to check, which runs on AWS Lambda and performs the simple task of uploading files to S3, timesout randomly when run manually from the aws console,but at the same time when run on a schedule, will always timeout at the first attempt and then run successfully the second time(it is run immediately after failing the first time).The code always seems to hang after calling s3.getBucketLocation, never reaching the getBucketLocation's callback(on the first attempt!).
All VPC's and NAT are set-up correctly, such that, when I remove s3.getBucketLocation from the code and call s3Upload.upload(params, cb); directly, the code runs without timeouts.
I would really like to understand why the first function fails alternatively when trying to retrieve the buckets region from AWS.
"use strict";
var urlHelper = require('url');
var fs = require('fs');
var AWS = require('aws-sdk');
var exp = {};
function saveToFile(path, content, cb) {
fs.writeFile(path, content, cb);
}
function saveToS3(bucket, key, content, cb) {
var s3 = new AWS.S3({apiVersion: '2006-03-01'});
var params = {
Bucket: bucket
};
s3.getBucketLocation(params, function(err, data) {
console.log('THIS IS NOT REACHED');
if (err) {
return cb(err);
}
var s3Upload = new AWS.S3({
apiVersion: '2006-03-01',
region: data.LocationConstraint
});
var params = {
Bucket: bucket,
Key: key,
Body: content,
ACL: 'public-read',
ContentType: 'text/xml',
CacheControl: 'max-age=60'
};
s3Upload.upload(params, cb);
});
}
exp.upload = function(path, content, cb) {
var url = urlHelper.parse(path);
switch(url.protocol) {
case 'file:':
case null:
saveToFile(path, content, cb);
break;
case 's3:':
saveToS3(url.host, url.pathname.substring(1), content, cb);
break;
default:
console.log('No matching protocol found for upload!');
cb(null, null);
}
};
module.exports = exp;