MockPushServer

Simulates APNS and FCM platforms. Used for testing API failure modes. Highly concurrent, is to be used for stress-testing components without undue load on real platforms.

Supports HTTP/1.1 and HTTP2

FCM (formerly known as GCM)

You can customize the response by setting the registration value to an error code or to "update:new_registration_id"

Error codes are

error:MissingRegistration
error:InvalidRegistration
error:NotRegistered
error:InvalidPackageName
error:MismatchSenderId
error:MessageTooBig
error:InvalidDataKey
error:InvalidTtl
error:Unavailable
error:InternalServerError
error:DeviceMessageRate 
error:TopicsMessageRateExceeded

FCM Examples

Success with registration update

curl -i -k -XPOST -H "Content-Type: application/json" -H "Authorization: key=AIzaSyZ-1u" -d @../mock_push_server/priv/sample-multiple.json https://localhost:8443/fcm/send
# HTTP/1.1 200 OK
# content-length: 255
# content-type: application/json
# date: Wed, 18 Jan 2017 10:10:26 GMT
# server: Cowboy
#
# {"success":2,"results":[{"message_id":"2c6ccfe0-9064-49eb-acff-55e98751001b"},{"registration_id":":registration_id2","message_id":"2c6ccfe0-9064-49eb-acff-55e98751001b"}],"multicast_id":"3148d4fb-4309-45ca-9eb2-20a819738d87","failure":0,"canonical_ids":1}

some failures

curl -i -k -XPOST -H "Content-Type: application/json" -H "Authorization: key=AIzaSyZ-1u" -d @../mock_push_server/priv/sample-failures.json

# HTTP/1.1 200 OK
# content-length: 219
# content-type: application/json
# date: Wed, 18 Jan 2017 11:15:00 GMT
# server: Cowboy
#
# {"success":1,"results":[{"error":"NotRegistered"},{"error":"DeviceMessageRate"},{"message_id":"d9db8abb-3b1a-4e29-a750-55901d0534f4"}],"multicast_id":"6964845d-8131-4694-b572-d10b682cd682","failure":2,"canonical_ids":0

APNS

Note : For best effect, this API should only be accessed through HTTP2, just like the real one.

If the token is set to one of these values, this error (with appropriate HTTP return status) will be returned by the API.

No authentication is performed.

error:BadCollapseId
error:BadDeviceToken
error:BadExpirationDate
error:BadMessageId
error:BadPriority
error:BadTopic
error:DeviceTokenNotForTopic
error:DuplicateHeaders
error:IdleTimeout
error:MissingDeviceToken
error:MissingTopic
error:PayloadEmpty
error:TopicDisallowed
error:BadCertificate
error:BadCertificateEnvironment
error:ExpiredProviderToken
error:Forbidden
error:InvalidProviderToken
error:MissingProviderToken
error:BadPath
error:MethodNotAllowed
error:Unregistered
error:PayloadTooLarge
error:TooManyProviderTokenUpdates
error:TooManyRequests
error:InternalServerError
error:ServiceUnavailable
error:Shutdown

APNS Examples

The H2Client module tests with HTTP2

H2Client.init_local |> H2Client.apns_local pid
# {:ok,
#  %Kadabra.Stream{body: nil,
#   headers: [{":status", "200"}, {"apns-id", "my-topic"}, {"content-length", "0"},
#    {"content-type", "application/json"},
#    {"date", "Wed, 18 Jan 2017 11:18:19 GMT"}, {"server", "Cowboy"}], id: 1,
#   status: 200}}

H2Client.init_local |> H2Client.apns_local_error( "BadDeviceToken")
# {:ok,
# %Kadabra.Stream{body: "{\"reason\":\"BadDeviceToken\"}",
#  headers: [{":status", "400"}, {"content-length", "27"},
#   {"content-type", "application/json"},
#   {"date", "Wed, 18 Jan 2017 11:41:57 GMT"}, {"server", "Cowboy"}], id: 1,
#  status: 400}}

Latency

System latency in ms can be increased by calling the /latency endpoint:

 curl -i -k -XPUT https://localhost:8443/latency -d "{\"gcm\": 100, \"apns\": 130}"

# HTTP/1.1 200 OK
# content-length: 25
# content-type: application/json
# date: Wed, 18 Jan 2017 11:09:38 GMT
# server: Cowboy
# 
# {"gcm": 100, "apns": 130}

Installation

mix deps.get
iex -S mix

cstar/mock_push_server