Another endless reload/redirect

Hi there,

Based on other questions, the most likely reason for this is a problem with redis configuration; however, I'm running this locally, with the default value of `null` for the session. Please see my config below.

```
const path = require('path');

module.exports = {
/**
* The port to run Nucleus Server on, if the port is in use the server will not start
*/
port: 3030,

/**
* The fully qualified domain + path that Nucleus is being hosted at
*/
baseURL: 'http://localhost:8888',

/**
* The data store to use when persisting plugins and versions. Current possible values
* are "sequelize", ensure you also supply valid connection details for your
* chosen strategy below.
*
* PR's welcome to add another data store.
*/
dbStrategy: 'sequelize',

/**
* Sequelize connection information, please note all options are required
*
* database: The name of the database to connect to
* dialect: The type of SQL database this is, check sequelize docs for more info
* username: Username to use when connecting
* password; Password to use when connecting
* host: Hostname of database
* port: Port to use when connecting
* storage: Path to sqlite file, only used for sqlite dialect
*/
sequelize: {
dialect: 'sqlite',
storage: path.resolve(__dirname, 'db.sqlite'),
},

/**
* The file store to use when persisting update files and metadata. Current possible
* values are "s3" and "local" ensure you also supply valid connection details if
* required for your chosen strategy below.
*
* PR's welcome to add another file store.
*/
fileStrategy: 'local',

/**
* Local file configuration
*
* root: Path on disk to the root of the static file store
* staticUrl: The HTTP url to use to access the static file store remotely
*/
local: {
root: path.resolve(__dirname, '.files'),
staticUrl: 'http://localhost:9999'
},

/**
* There is actually no authentication config for s3, all config must be done through the standard AWS
* environment variables or through EC2 IAM roles.
*
* See http://docs.aws.amazon.com/sdk-for-javascript/v2/developer-guide/setting-credentials-node.html
*
* Bucket / Region / CloudFront config goes here though
*/
s3: {
// init: {
// endpoint: '' // The alternate endpoint to reach the S3 instance at,
// s3ForcePathStyle: true // Always use path style URLs
// }

bucketName: '', // The name for your S3 Bucket

cloudfront: { // If you don't have CloudFront set up and just want to use the S3 bucket set this to "null
distributionId: '', // The CloudFront distribution ID, used for invalidating files
publicUrl: '', // Fully qualified URL for the root of the CloudFront proxy for the S3 bucket
}
},

/**
* The authentication strategy to use when logging users in. Current possible values are "local",
* "openid" and "github". Make you also supply the required authentication details
*/
authStrategy: 'local',

/**
* Local authentication details
*
* The `adminIdentifiers` array should be a list of usernames
*
* DISCLAIMER: This strategy should ONLY be used for local development and NEVER
* used in production. Unicorns cry every time this setting is used in production.
* Don't make the unicorns cry.
*
* displayName: The user friendly name of this user
* username: A unique identifier to use when this user signs in, please note uniqueness is
* not enforced
* password: Well, uhhh, their password
* photo: A URL for their profile, entirely optional, just makes things look nicer ;)
*/
localAuth: [{
displayName: 'Charlie',
username: 'charlie',
password: 'charlie',
photo: 'https://pbs.twimg.com/profile_images/1219364727/charlie-support_400x400.png'
}],

/**
* OpenID authentication details
*
* The `adminIdentifiers` array should be a list of email
* addresses for users to consider admins
*
* realm: The domain that the server is hosted on
* stateless: Stateless mode for openID
* profile: Whether to fetch profile information, should normally be true
* providerURL: Your openID provider URL
* domain: Domain to restrict email addresses to
*/
openid: {
realm: 'http://localhost:8888',
stateless: true,
profile: true,
providerURL: 'https://auth.myservice.com/openid/v2/op',
domain: 'myservice.com'
},

/**
* GitHub authentication details
*
* The `adminIdentifiers` array should be a list of GitHub usernames
* to consider admins
*
* clientID: GitHub API client ID
* clientSecret: GitHub API clientSecret
* realm: The domain the server is hosted on
*/
github: {
clientID: '',
clientSecret: ''
},

/**
* See the documentation for your authentication strategy for what this array does
*/
adminIdentifiers: ['admin@yourdomain.com', 'charlie'],

/**
* Session options, in development just leave this as default.
*
* IN PRODUCTION PLEASE USE REDIS!
*
* type: Can be either "redis" or null
*
* redis:
* host: The host URL for the redis instance
* port: The port for the redis instance
*/
sessionConfig: {
type: null,
secret: 'ThisIsNotSecret',

redis: {
host: '',
port: ''
}
},

organization: 'My Company Here',

/**
* GPG key to use when signing APT and YUM releases
*
* Requires to be unlocked (no password) and have both the private and
* public key.
*/
gpgSigningKey: `
this works fine, did not include here
`,

/**
* The default percentage rollout for new releases. The first release for
* any channel will always be 100% but all future releases will have a
* default rollout value of this setting
*/
defaultRollout: 0
};
```
I've been fighting with this for hours, and I feel like I'm missing something so simple. I really hate to put support type questions on here, but after 7 hours of fighting with this, I figured I should just ask.

Notes:
* I also tried getting the docker version up and running first, but I was getting an infinite loop there as well (I was attempting to connect to redis with that), so I switched over to the manual version for better debugging.
* I could not get the npm modules to install using node version 10 on Windows (after about 3 hours of fighting). I had to use node version 8 on the linux subsystem for windows (node version 10 also resulted in failure on the subsystem). The culprit seems to be node gyp.

Reproduce the loop with the supplied manual configuration, then compare it with the Docker setup and its attempted Redis connection. The issue names no source files or tests; done means identifying the redirect cause and documenting a configuration or code-level fix that prevents the endless reload.

Written by the indexing model from the issue text.