The solution I use now is to have the queue in the dB and then have a single process pushing to another queue better suited to dealing with workers.
I drank from the “keep your queue out of the dB” koolaid once, but then you end up worrying about transaction boundaries and atomic writes to the queue and the dB. It totally depends on your workload, but in my case, I’ve found the halfway solution of the queue in the dB so you get to write to it within your transactions, with a dedicated queue for the workers to be a much better solution.
If you have the case, if A then do B and C. You can commit to a topic and let B and C be triggered separately. This does not solve cases like storing to a db and then send a db-updated event.
Are you sure it works properly when something fails. That’s a wonderful situation I like to ask people because the answer is usually no. Then someone gets billed twice for something and then the guy who said this was a good idea gets shot.
Well, I only ever enqueue idempotent work, so again, in my scenario it can fail and be rerun, but I’m also unsure about what you think is a bad idea here.
Honestly, I had more code managing two phase commits so I would only push to the queue when I was sure (ish) the dB commit was ok.
Say you need to trigger something to send an email, but let’s say that there’s an error condition after and the transaction is rolled back. How do you handle that failure? In my scenario the queue is rolled back too, no harm to foul.
I drank from the “keep your queue out of the dB” koolaid once, but then you end up worrying about transaction boundaries and atomic writes to the queue and the dB. It totally depends on your workload, but in my case, I’ve found the halfway solution of the queue in the dB so you get to write to it within your transactions, with a dedicated queue for the workers to be a much better solution.