Skip to content

Latest commit

 

History

History
203 lines (147 loc) · 8.57 KB

README.md

File metadata and controls

203 lines (147 loc) · 8.57 KB
Raw

PyPI status

Apigee X - Google Cloud Logging Shared Flow

This is not an official Google product.
This implementation is not an official Google product, nor is it part of an official Google product. Support is available on a best-effort basis via GitHub.

Intro

This repository contains a sharedflow for Apigee X to send proxy logs to Google Cloud Logging.
It is based on DinoChiesa / Apigee-GCP-Logging-Example, but updated to be deployed on Apigeee X as a Sharedflow

Requirement

  • A Google Cloud Platform account
  • An Apigee X organization

Installation

Setting up GCP

To authorize Apigee X to send logs records to Google Cloud Logging, you must first:

  • Create a service account in Google Cloud and assign it the necessary roles (at least, Logs Writer)
  • Create and download a json key for the service account

To create it in your Apigee organization's GCP project, use following gcloud commands (or GCP Web UI):

SA_NAME=<your-new-service-account-name>

gcloud iam service-accounts create $SA_NAME --display-name="Apigee Logging"

PROJECT_ID=$(gcloud config get-value project)
APIGEE_LOGGING_SA=$SA_NAME@$PROJECT_ID.iam.gserviceaccount.com

gcloud projects add-iam-policy-binding "$PROJECT_ID" \
  --member="serviceAccount:$APIGEE_LOGGING_SA" \
  --role="roles/logging.logWriter"

gcloud iam service-accounts keys create $SA_NAME-key.json --iam-account=$APIGEE_LOGGING_SA --key-file-type=json

You will use values from json key file during Apigee X KVM configuration.

Setting up the KVMs

The sharedflow uses 2 crypted KVMs to store parameters and credentials to access to GCP Logging service:

  • GCP-Keys:
    • key "gcplogging.privKeyPem": the PEM-encoded private key (private_key) from Service Account json key file
  • GCP-Logging-settings:
    • key "gcplogging.jwt_issuer": client_email from Service Account json key file
    • key "gcplogging.logid": logname used to store Apigee log record in Cloud Logging
    • key "gcplogging.projectid": Google Cloud Logging Project Id, project_id from Service Account json key file

The KVMs in Apigee X (or hybrid) still need to be created via the management API or the Apigee UI but their content can only be created, read or deleted from within the runtime from a KeyValueMapOperations policy (see Apigee Community Article Providing KVM Content APIs for Apigee X and hybrid)

A full example of an implementation of the GET, POST and DELETE methods for KVMs with a dynamic name can be found in the Apigee DevRel Github repository: kvm-admin-api.

We will use it in next steps to initialize GCP-Keys and GCP-Logging-settings KVMs. To deploy kvm-admin-api, follow instructions in Apigee DevRel Github repository.

Then, create KVMs using Apigee API:

export TOKEN=$(gcloud auth print-access-token)
export APIGEE_ORG=<my-org-name>
export APIGEE_ENV=<my-env>
export KVM_NAME1=GCP-Keys
export KVM_NAME2=GCP-Logging-settings

curl -X POST \
    "https://apigee.googleapis.com/v1/organizations/${APIGEE_ORG}/environments/$APIGEE_ENV/keyvaluemaps" \
    -H "Authorization: Bearer $TOKEN" \
    -H "Content-Type: application/json" \
    --data "{\"name\":\"$KVM_NAME1\",\"encrypted\": true}"

curl -X POST \
    "https://apigee.googleapis.com/v1/organizations/${APIGEE_ORG}/environments/$APIGEE_ENV/keyvaluemaps" \
    -H "Authorization: Bearer $TOKEN" \
    -H "Content-Type: application/json" \
    --data "{\"name\":\"$KVM_NAME2\",\"encrypted\": true}"

After that, create KVM entries using kvm-admin-api:

export TOKEN=$(gcloud auth print-access-token)
export APIGEE_HOSTNAME=<my-apigee-hostname>
export APIGEE_ORG=<my-org-name>
export APIGEE_ENV=<my-env>
export KVM_NAME1=GCP-Keys
export KVM_NAME2=GCP-Logging-settings

curl -i -X POST \
   "https://$APIGEE_HOSTNAME/kvm-admin/v1/organizations/${APIGEE_ORG}/environments/$APIGEE_ENV/keyvaluemaps/$KVM_NAME1/entries"
   -H "Content-Type:application/json" \
   -H "Authorization:Bearer $TOKEN" \
   -d '{ "key": "gcplogging.privKeyPem", "value": "<copy here private_key value from SA json key file>" } ' \

curl -i -X POST \
  "https://$APIGEE_HOSTNAME/kvm-admin/v1/organizations/${APIGEE_ORG}/environments/$APIGEE_ENV/keyvaluemaps/$KVM_NAME2/entries"
   -H "Content-Type:application/json" \
   -H "Authorization:Bearer $TOKEN" \
   -d '{ "key": "gcplogging.jwt_issuer", "value": "<copy here client_email value from SA json key file>"}' 

curl -i -X POST \
   "https://$APIGEE_HOSTNAME/kvm-admin/v1/organizations/${APIGEE_ORG}/environments/$APIGEE_ENV/keyvaluemaps/$KVM_NAME2/entries"
   -H "Content-Type:application/json" \
   -H "Authorization:Bearer $TOKEN" \
   -d '{ "key": "gcplogging.logid", "value": "apigee-logs"}' 

curl -i -X POST \
   "https://$APIGEE_HOSTNAME/kvm-admin/v1/organizations/${APIGEE_ORG}/environments/$APIGEE_ENV/keyvaluemaps/$KVM_NAME2/entries"
   -H "Content-Type:application/json" \
   -H "Authorization:Bearer $TOKEN" \
   -d '{ "key": "gcplogging.projectid", "value": "<copy here project_id value from SA json key file>" }'

Note: you could use the same curl commands to update the KVM entries.

Installing Sharedflow

Upload Shared Flow and deploy it:

  • Zip sharedflowbundle folder from this repository
  • From Apigee X, Develop menu, Shared Flows, click Upload Bundle
  • Browse your folders and select zip file: click Open
  • Enter a name for the new Shared Flow

Upload Shared Flow

  • Deploy it to your target environment

Using Sharedflow in your proxies

To use this shared flow in you proxy, insert a FlowCallout policy in Proxy Endpoint PostFlow

Insert Shared Flow

As implemented in DinoChiesa / Apigee-GCP-Logging-Example, you can send logs to Google Cloud Logging asynchronously (httpClient from within a JavaScript callout) with a minimum of delay introduced into the proxy flow, or wait for a Google Cloud Logging response (ServiceCallout policy). Call your proxy with header useSC = true, if you want to use ServiceCallout.

If you want to customize Log Format, edit the script linked to Javascript policy JS-SetLoggingRecord in the Shared Flow. Current log format is:

var organization = context.getVariable("organization.name");
var environment = context.getVariable("environment.name");
var region = context.getVariable("system.region.name") || "none";
var apiProxy = context.getVariable("apiproxy.name");
var apiProxyRevision = context.getVariable("apiproxy.revision");
var requestUrl = context.getVariable("request.url");
var verb =  context.getVariable("request.verb");
var correlationId = context.getVariable("messageid");
var client_ip = context.getVariable("client.ip");
var target_ip = context.getVariable("target.ip") || "none";
var target_host = context.getVariable("target.host") || "none";
var proxyRequestReceived = context.getVariable("client.received.end.timestamp");
var proxyResponseSent = context.getVariable("client.sent.end.timestamp");
var targetResponseReceived = context.getVariable("target.received.end.timestamp") || "none";
var targetRequestSent = context.getVariable("target.sent.end.timestamp") || "none";
var targetResponseCode = context.getVariable("message.status.code") || "none";
var proxyResponseCode = context.getVariable("response.status.code") || "none";
var fault_name = context.getVariable("fault.name") || "none";


var LogRecord = {
  "organization": organization,
  "environment": environment,
  "region": region,
  "apiProxy": apiProxy,
  "apiProxyRevision": apiProxyRevision,
  "requestUrl": requestUrl,
  "verb": verb,
  "correlationId": correlationId,
  "client_ip": client_ip,
  "target_ip": target_ip, 
  "target_host": target_host, 
  "proxyRequestReceived": proxyRequestReceived,
  "proxyResponseSent": proxyResponseSent,
  "targetResponseReceived": targetResponseReceived,
  "targetRequestSent": targetRequestSent,
  "targetResponseCode": targetResponseCode,
  "proxyResponseCode": proxyResponseCode,
  "faultName": fault_name
 }

...

See logs in Google Cloud Logging

Open Google Cloud Logging Console.

  • Using Query Builder, select apigee-logs log name, Add
  • Click Run query

Cloud Logging