Today my discord bot went offline, when I restarted the dyno's. The logs showed progress until it got stuck on "State changed from starting to up"
Anyone have any idea why? Please let me know!
Related
https://i.stack.imgur.com/7Ar6B.png
i get a ton of these error in my logs even when my dyno is offline.
does someone have any idea of where there from ?
I beat my head against this for months without getting anywhere with Heroku support. They just kept pointing me to the L10 error docs which I'd read (and understood). The issue in my case was that there was a drain present that was NOT from an add-on. It may have been added manually by a previous engineer? Not sure. In any case, this is how I debugged and resolved it:
Look at the L10 error message to see the drain id:
2021-10-12T14:58:23.642214+00:00 heroku[logplex]: Error L10 (output buffer overflow): drain 'd.89456495-cf6c-4c1b-8865-kd0238j2098' dropped 4 messages since 2021-10-12T14:57:23.753982+00:00.
Use the heroku CLI to inspect the drains:
heroku drains --json -a [app-name]
This will list all the drains, both from add-ons and others. In my case there was an old logmatic drain (d.89456495-cf6c-4c1b-8865-kd0238j2098) in there that was the source of the errors.
Manually remove the problematic drain:
heroku drains:remove [URL] -a [app-name]
Restart the dynos? Not sure if that's necessary or if you can just wait.
Heroku log drains reference: https://devcenter.heroku.com/articles/log-drains
Like #wlabs said, it is probably a drains issue. Some people report that it could be papertrail overflowing, which is common on a free plan level.
If you don't want to remove papertrail from the app, then go into your papertrail settings and configure the log filters to cut out noise you don't care about. Then wait for the reset on the next day.
for anyone having the same issue,
I think there is no need to recreate the app, just contact the support and they will help you.
For my case the response was like that:
Thank you for contacting us. I've made some changes to the backend for
[my-app-name] and [the-stg-of-my-app] that should hopefully resolve
things. Could you please restart your applications and then let me
know if your logs work after this?
and it works!!
I finally deleted and recreated the app
Has anyone come across this before? The error message doesn't provide much useful information and Googling the error messsage returns no results. My app is part of a Pipeline, I think that might have something to do with it.
Based on the time of this post, heroku is having issues with recycling dynos.
Happening again today - heroku seems to be down (4th of Feb, 2020).
For more info: https://status.heroku.com/
I had run into a few problems getting Ruby fully installed and working, so I could start building. I finished installing and tested my server at localhost:3000, and it came up fine. Then the next day when I tried to go back to it, it didn't connect and I could not figure out why.
First of all, there's no specific way for us to address this issue. You haven't mentioned any of what happened between the server working and malfunctioning? It could be as simple as the server not being started. Is the server currently running? If it is than you should have some sort of response when you try to hit localhost.3000.
Make sure your server is running with rails s and then let us know.
Perhaps you can try resetting your DB with a rake db:reset if that doesn't work.
Besides that, there really isn't anything we can do on this end without more information.
I have a main incoming mailbox in CRM (sort of like a contact#organisation.com address) but for some reason today, the mail has stopped flowing into CRM.
I have tried resetting the async services as mentioned here: https://community.dynamics.com/crm/f/117/t/159892
But the mailbox remains stuck.
I have also tried enabling tracing on the backend servers but even have 20 minutes I didn't find any issues in the trace logs...
EDIT
So the process completed at 4am this morning and I can't work out why the test took over 17 hours to complete without error.
What is odd is all the async jobs ran yesterday (in a timely matter) without issue and I did not find any errors related to this email issue in the trace logs.
I am concerned with why this process took so long without any indication of an issue and I would be happy for someone to raise some suggestions
I know it's not the nice way, but does anyone of you know a service where I can run a heartbeat to my heroku application and if it fails for a couple of minutes to respond to call heroku restart and send me an email?
Would be a really nice feature for me...
Thanks a lot!
Cheers!
Restarting the app would definitely not be the best solution.
heroku isn't 100% of the time available
See their status page.
There is pingdom though, which will ping your app and notify you if it is down.
But won't take any action for you.