Ghost/ghost/mailgun-client/lib/mailgun-client.js
Simon Backx 923c522778
Implemented email analytics retrying (#16273)
fixes https://github.com/TryGhost/Team/issues/2562

New event fetching loops:
- Reworked the analytics fetching algorithm. Instead of starting again
where we stopped during the last fetching minus 30 minutes, we now just
continue where we stopped. But with ms precision (because no longer
database dependent after first fetch), and we stop at NOW - 1 minute to
reduce chance of missing events.
- Apart from that, a missing fetching loop is introduced. This fetches
events that are older than 30 minutes, and just processes all events a
second time to make sure we didn't skip any because of storage delays in
the Mailgun API.
- A new scheduled fetching loop, that allows us to schedule between a
given start/end date (currently only persisted in memory, so stops after
a reboot)

UI and endpoint changes:
- New UI to show the state of the analytics 'loops'
- New endpoint to request the analytics loop status
- New endpoint to schedule analytics
- New endpoint to cancel scheduled analytics
- Some number formatting improvements, and introduction of 'opened'
count in debug screen
- Live reload of data in the debug screen

Other changes:
- This also improves the support for maxEvents. We can now stop a
fetching loop after x events without worrying about lost events. This is
used to reduce the fetched events in the missing and scheduled event
loop (e.g. when the main one is fetching lots of events, we skip the
other loops).
- Prevents fetching the same events over and over again if no new events
come in (because we always started at the same begin timestamp). The
code increases the begin timestamp with 1 second if it is safe to do so,
to prevent the API from returning the same events over and over again.
- Some optimisations in handing the processing results (less merges to
reduce CPU usage in cases we have lots of events).

Testing:
- You can test with lots of events using the new mailgun mocking server
(Toolbox repo `scripts/mailgun-mock-server`). This can also simulate
events that are only returned after x minutes because of storage delays.
2023-02-20 16:44:13 +01:00

298 lines
11 KiB
JavaScript

const _ = require('lodash');
const debug = require('@tryghost/debug');
const logging = require('@tryghost/logging');
const metrics = require('@tryghost/metrics');
const errors = require('@tryghost/errors');
module.exports = class MailgunClient {
#config;
#settings;
static BATCH_SIZE = 1000;
constructor({config, settings}) {
this.#config = config;
this.#settings = settings;
}
/**
* Creates the data payload and sends to Mailgun
*
* @param {Object} message
* @param {Object} recipientData
* @param {Array<Object>} replacements
*
* recipientData format:
* {
* 'test@example.com': {
* name: 'Test User',
* unique_id: '12345abcde',
* unsubscribe_url: 'https://example.com/unsub/me'
* }
* }
*/
async send(message, recipientData, replacements) {
const mailgunInstance = this.getInstance();
if (!mailgunInstance) {
logging.warn(`Mailgun is not configured`);
return null;
}
if (Object.keys(recipientData).length > MailgunClient.BATCH_SIZE) {
throw new errors.IncorrectUsageError({
message: `Mailgun only supports sending to ${MailgunClient.BATCH_SIZE} recipients at a time`
});
}
let messageData = {};
let startTime;
try {
const bulkEmailConfig = this.#config.get('bulkEmail');
const messageContent = _.pick(message, 'subject', 'html', 'plaintext');
// update content to use Mailgun variable syntax for replacements
replacements.forEach((replacement) => {
messageContent[replacement.format] = messageContent[replacement.format].replace(
replacement.regexp,
`%recipient.${replacement.id}%`
);
});
messageData = {
to: Object.keys(recipientData),
from: message.from,
'h:Reply-To': message.replyTo || message.reply_to,
subject: messageContent.subject,
html: messageContent.html,
text: messageContent.plaintext,
'recipient-variables': JSON.stringify(recipientData)
};
// add a reference to the original email record for easier mapping of mailgun event -> email
if (message.id) {
messageData['v:email-id'] = message.id;
}
const tags = ['bulk-email', 'ghost-email'];
if (bulkEmailConfig?.mailgun?.tag) {
tags.push(bulkEmailConfig.mailgun.tag);
}
messageData['o:tag'] = tags;
if (bulkEmailConfig?.mailgun?.testmode) {
messageData['o:testmode'] = true;
}
// enable tracking if turned on for this email
if (message.track_opens) {
messageData['o:tracking-opens'] = true;
}
const mailgunConfig = this.#getConfig();
startTime = Date.now();
const response = await mailgunInstance.messages.create(mailgunConfig.domain, messageData);
metrics.metric('mailgun-send-mail', {
value: Date.now() - startTime,
statusCode: 200
});
return {
id: response.id
};
} catch (error) {
logging.error(error);
metrics.metric('mailgun-send-mail', {
value: Date.now() - startTime,
statusCode: error.status
});
return Promise.reject({error, messageData});
}
}
/**
* Fetches events from Mailgun
* @param {Object} mailgunOptions
* @param {Function} batchHandler
* @param {Object} options
* @param {Number} options.maxEvents Not a strict maximum. We stop fetching after we reached the maximum AND received at least one event after begin (not equal) to prevent deadlocks.
* @returns {Promise<void>}
*/
async fetchEvents(mailgunOptions, batchHandler, {maxEvents = Infinity} = {}) {
const mailgunInstance = this.getInstance();
if (!mailgunInstance) {
logging.warn(`Mailgun is not configured`);
return;
}
debug(`fetchEvents: starting fetching first events page`);
const mailgunConfig = this.#getConfig();
let startTime = Date.now();
const startDate = new Date();
try {
let page = await mailgunInstance.events.get(mailgunConfig.domain, mailgunOptions);
metrics.metric('mailgun-get-events', {
value: Date.now() - startTime,
statusCode: 200
});
// By limiting the processed events to ones created before this job started we cancel early ready for the next job run.
// Avoids chance of events being missed in long job runs due to mailgun's eventual-consistency creating events outside of our 30min sliding re-check window
let events = (page?.items?.map(this.normalizeEvent) || []).filter(e => !!e && e.timestamp <= startDate);
debug(`fetchEvents: finished fetching first page with ${events.length} events`);
let eventCount = 0;
const beginTimestamp = mailgunOptions.begin ? Math.ceil(mailgunOptions.begin * 1000) : undefined; // ceil here if we have rounding errors
while (events.length !== 0) {
await batchHandler(events);
eventCount += events.length;
if (eventCount >= maxEvents && (!beginTimestamp || !events[events.length - 1].timestamp || (events[events.length - 1].timestamp.getTime() > beginTimestamp))) {
break;
}
const nextPageId = page.pages.next.page;
debug(`fetchEvents: starting fetching next page ${nextPageId}`);
startTime = Date.now();
page = await mailgunInstance.events.get(mailgunConfig.domain, {
page: nextPageId,
...mailgunOptions
});
metrics.metric('mailgun-get-events', {
value: Date.now() - startTime,
statusCode: 200
});
// We need to cap events at the time we started fetching them (see comment above)
events = (page?.items?.map(this.normalizeEvent) || []).filter(e => !!e && e.timestamp <= startDate);
debug(`fetchEvents: finished fetching next page with ${events.length} events`);
}
} catch (error) {
// Log and re-throw Mailgun errors
logging.error(error);
metrics.metric('mailgun-get-events', {
value: Date.now() - startTime,
statusCode: error.status
});
throw error;
}
}
async removeSuppression(type, email) {
if (!this.isConfigured()) {
return false;
}
const instance = this.getInstance();
const config = this.#getConfig();
try {
await instance.suppressions.destroy(
config.domain,
type,
email
);
return true;
} catch (err) {
logging.error(err);
return false;
}
}
async removeBounce(email) {
return this.removeSuppression('bounces', email);
}
async removeComplaint(email) {
return this.removeSuppression('complaints', email);
}
async removeUnsubscribe(email) {
return this.removeSuppression('unsubscribes', email);
}
normalizeEvent(event) {
const providerId = event?.message?.headers['message-id'];
if (!providerId && !(event['user-variables'] && event['user-variables']['email-id'])) {
logging.error('Received invalid event from Mailgun');
logging.error(event);
return null;
}
return {
id: event.id,
type: event.event,
severity: event.severity,
recipientEmail: event.recipient,
emailId: event['user-variables'] && event['user-variables']['email-id'],
providerId: providerId,
timestamp: new Date(event.timestamp * 1000),
error: event['delivery-status'] && (typeof (event['delivery-status'].message || event['delivery-status'].description) === 'string') ? {
code: event['delivery-status'].code,
message: (event['delivery-status'].message || event['delivery-status'].description).substring(0, 2000),
enhancedCode: event['delivery-status']['enhanced-code']?.toString()?.substring(0, 50) ?? null
} : null
};
}
#getConfig() {
const bulkEmailConfig = this.#config.get('bulkEmail');
const bulkEmailSetting = {
apiKey: this.#settings.get('mailgun_api_key'),
domain: this.#settings.get('mailgun_domain'),
baseUrl: this.#settings.get('mailgun_base_url')
};
const hasMailgunConfig = !!(bulkEmailConfig?.mailgun);
const hasMailgunSetting = !!(bulkEmailSetting && bulkEmailSetting.apiKey && bulkEmailSetting.baseUrl && bulkEmailSetting.domain);
if (!hasMailgunConfig && !hasMailgunSetting) {
return null;
}
const mailgunConfig = hasMailgunConfig ? bulkEmailConfig.mailgun : bulkEmailSetting;
return mailgunConfig;
}
/**
* Returns an instance of the Mailgun client based upon the config or settings values
*
* We don't cache the instance so we can always get a fresh one based upon changed settings
* or config values over time
*
* Note: if the credentials are not configure, this method returns `null` and it is down to the
* consumer to act upon this/log this out
*
* @returns {import('mailgun.js')|null} the Mailgun client instance
*/
getInstance() {
const mailgunConfig = this.#getConfig();
if (!mailgunConfig) {
return null;
}
const formData = require('form-data');
const Mailgun = require('mailgun.js');
const baseUrl = new URL(mailgunConfig.baseUrl);
const mailgun = new Mailgun(formData);
return mailgun.client({
username: 'api',
key: mailgunConfig.apiKey,
url: baseUrl.origin,
timeout: 60000
});
}
/**
* Returns whether the Mailgun instance is configured via config/settings
*
* @returns {boolean}
*/
isConfigured() {
const instance = this.getInstance();
return !!instance;
}
};