- we have calls to the metrics library so we can measure the time it
takes the Mailgun API to return a response
- however, there's a bug in the code whereby if the `batchHandler`
takes a long time and then throws an error, this time will be reported
to metrics
- this is misleading because it looks like Mailgun is taking a long time
if the databases are slow
- this pulls the specific SDK call out into a function so it's easier to
wrap with timing code
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.
refs https://github.com/TryGhost/Team/issues/2486
Stop the event fetching loop as soon as we receive events that were
created later then when we started the loop. This ensures that we don't
miss events if we receive a giant batch of events that take a long time
to process.
- in the event the Mailgun config doesn't exist, we return `null` from
this function
- this updates the jsdoc to correct the return type of `getInstance`
refs: https://github.com/TryGhost/Ghost/issues/15725
- our users are having difficulties getting onboarded with mailgun
- we're adding an explicit and unique tag to all requests, to help mailgun detect when mail is being sent from Ghost
refs https://github.com/TryGhost/Team/issues/2255
These methods will be used by the Mailgun implementation of EmailSuppressionList
so that emails are removed from both our internal list and Mailguns.
fixes https://github.com/TryGhost/Team/issues/2332
Saves events in the database and collects error information.
Do note that we can emit the same events multiple times, and as a result
out of order. That means we should correctly handle that a delivered
event might be fired after a permanent failure. So a delivered event is
ignored if the email is already marked as failed. Also delivered_at is
reset to null when we receive a permanent failure.
fixes https://github.com/TryGhost/Team/issues/2096
When generating the recipient data for emails, the email clicks
implementation is resulting in a recipient variable being added called
replacement_xxx once for each link containing the same UUID.
This generates a lot of unnecessary data overhead for emails, and it
turns out that mailgun has a 25MB message limit. We wouldn't have come
close if we only included the uuid once.
refs https://github.com/TryGhost/Toolbox/issues/164
- see referenced issue for more context but Ghost sometimes has issues
with the email analytics job getting stuck
- we don't provide a timeout to the Mailgun library, so we just
sit there idling for eternity if something between us and Mailgun is
causing issues
- this commit adds a 60s timeout so we can at least error out and try
again next time
- during a refactor, I moved the `BATCH_SIZE` variable around
- putting the variable export above a more general export means it get
overwritten and the value is `undefined` outside of the module
- when we chunk the emails, we were chunking in sized of `undefined`,
so I'm guessing it just defaulted to 1
- this means the email batches were of size 1 instead of 1000 - oops
refs https://github.com/TryGhost/Toolbox/issues/363
- this commit switches us to using the official and maintained
`mailgun.js` SDK, and updates the `mailgun-client` code to reflect the
changes between the two
- when I extracted `mailgun-client` to a separate package, I
accidentally removed the lazyloading for the `mailgun-js` library,
which takes a non-negligible amount of time to require on boot
- this fixes that by moving the require into the function where it's
used
- even though we don't do anything yet, the code was calling `.length`
on an object, which is `undefined`
- this fixes that by checking the length of the keys
- also moves the code block down so we can early-return if mailgun is
not configured
refs https://github.com/TryGhost/Toolbox/issues/363
- this commit pulls all code involving the Mailgun client SDK into one
new package called `mailgun-client`
- this means we should be able to replace `mailgun-js` (deprecated) with
`mailgun.js` (the new, official one) without editing code all over the
place
- this also lays some groundwork for better testing of smaller
components
