See the failed jobs in Sidekiq. (MailGun) failing to send emails

Product Support

AI-generated summary

The discussion revolves around failed email jobs in Sidekiq related to Mailgun not syncing with DigitalOcean, causing email delivery issues. Aaron_Walsh noticed 42 failed jobs and suspected a configuration problem with Mailgun, despite having correct credentials and port settings. Initially using port 587, which is sometimes blocked by DigitalOcean, pfaffman suggested switching to port 2525 as a fix. Aaron_Walsh confirmed they had tried port 2525 but still encountered an “execution expired” error when testing via /admin/email/server-settings. After verifying the port change and performing a full rebuild using ./launcher destroy app and ./launcher start app, the issue was resolved. Aaron_Walsh confirmed the fix and apologized for the confusion, noting the issue had persisted since 2022.

1

Evening all,

I haven’t really been using my forum, either as an end user or an admin, but I noticed it was a bit quiet. All of a sudden, when looking at the admin panel, I have 42 email jobs, and reading using Gemini, it is related to the provider (Mailgun) not being synced with the provider (DigitalOcean console) through the app.yml. I haven’t changed anything and can’t find anything on this forum about whether we needed to change anything on our side.

Username and now password are correct; other than that, there isn’t much else with the SMTP I can change. The port is also correct, and when I say “now” password, I’ve also changed this.

None of these have been changed since 2022, so what could it be?

A dark-themed software interface displays installation details, a positive update status with a smiley face, advisory messages regarding failed email jobs, and a refresh option with a recent check timestamp. (Captioned by AI)
A dark-themed software interface displays installation details, a positive update status with a smiley face, advisory messages regarding failed email jobs, and a refresh option with a recent check timestamp. (Captioned by AI)1790×542 111 KB

2

Which port do you currently use?

3

Hi,

Originally, I tested this port, 2525, as well, from the recommendation of ./discourse-doctor to test the mailing side.

4

Was it always 2525, or did you just change it?

If you changed it, did you destroy and start the container (or do a rebuild)?

If you go to /admin/email/server-settings can you send a test message?

5

No, always been 587 until about 2 hours ago and been this since 2022.

Once I change them in the container I’ve been rebuilding.

In here it comes up as Error “ERROR - execution expired”

6

That sounds like it’s still getting blocked. Are you sure that you changed it to 2525? And when you rebuilt it built and started a new container?

7

So I’ve done this:

cd /var/discourse
./launcher destroy app
./launcher start app

And now it works. I do apologize. I see the original topic was back in 2017 and lasted this long, which is good going. Hopefully, this will fix it going forward.