Hi all – I had about 30 minutes of downtime yesterday with my DigitalOcean Droplet install of Discourse (as measured by external monitoring via Pingdom), and am wondering what the next steps are to diagnose and prevent recurrence. What I’ve observed so far: My Discourse install is 1.9.0.beta13. …

The stacktrace has to do with Discourse not being able to talk to your SMTP server. Since this is done in a background job I don’t think this is what caused your downtime. The monitoring agent not reporting for the first ~15 minutes of the downtime suggests that this might be a system issue rather …

Did the server reboot? (check uptime) Did any messages get posted in the kernel logs? (check dmesg) Any relevant entries in syslog? (less /var/log/syslog) The stack traces seem to be a symptom of your downtime, I suspect you lost your networking for the duration of the outage: /usr/local/lib/rub…

Thanks guys! System did not reboot (uptime 218 days). Unfortunately it looks like syslog and dmesg aren’t configured on my box to retain logs more than a day, so I don’t see anything from around then. I do see a lot of messages (every 10 minutes or so), still ongoing, like this: Oct 24 10:35:03 …

Debugging discourse downtime

Product Support

jphaas December 4, 2019, 9:47pm 7

Nope! Pingdom says it’s been up for the last 3 months (and I think 3 months ago I took it down for maintenance). Delightfully stable. Thanks so much for following up!

Topic		Replies	Views
Email jobs failing due to smtpserverbusy Self-hosting email	9	2094	November 27, 2019
Errors on a forum I volunteer with Product Support	2	1179	December 22, 2023
DigitalOcean blocking SMTP and forcing SendGrid usage Product Support email	11	4116	April 17, 2025
"There are email jobs that failed" on an SMTP-less install Product Support email , discourse-id	6	303	February 19, 2026
Bursts of 502 Service Unavailable, pointers to debug Self-hosting	15	1818	February 18, 2018

Debugging discourse downtime

Related topics