Pandemic 2702 Share Posted October 2, 2020 Hello everyone, For the past week or so, we've had some DB3 server stability issues leading to a couple periods of downtime, and we'd like to apologize for these disturbances. It's a new platform still getting used to the load demands, and as such things like this sometimes happen. We've extended all active VIP and Sponsor subscriptions by 48 hours for the inconvenience. I believe renewals still happen on the same day as normal, but on subscription end you'll still have the extra time. I've already made and I'm continuing to make changes to prevent and lessen the odds of issues like this happening again: Changes Already Made: Daily restarts are now enforced again Client re-connection logic is improved, you shouldn't notice any downtime for running clients during short periods of downtime (<4 hours) @Hashtag now has the ability to restart the DB3 server in the event something happens outside of my timezone (like today's incident) Changes Planned: Further server improvements to caching and cold start times Continued client networking improvements [Oct. 2] DreamBot 3.0.9 includes significant networking improvements Upgrading our infrastructure so we can better handle the load (will happen within 4 weeks, will require planned downtime) [Oct. 3] Offloaded a few services, it should now be fine but will reevaluate if issues reoccur Improved uptime checking service [Oct. 2] The monitoring service is now more thorough, and should avoid previous failure that ignored today's outage Sorry again for the inconvenience, we know how annoying these incidents are and appreciate you all for your understanding. The Dream Team Link to comment Share on other sites More sharing options...
mbmaster 26 Share Posted October 2, 2020 Things like this can happen in my opinion. I'm always happy with work and time you guys put into db3. "The only people who don't make mistakes are the people who don't do anything" Link to comment Share on other sites More sharing options...
yeeter 520 Share Posted October 2, 2020 Did you try turning it off and on again? Link to comment Share on other sites More sharing options...
Zawy 934 Share Posted October 2, 2020 28 minutes ago, yeeter01 said: Did you try turning it off and on again? Yes rebooted pc now its working again Ty Link to comment Share on other sites More sharing options...
yeeter 520 Share Posted October 2, 2020 18 minutes ago, Zawy said: Yes rebooted pc now its working again Ty Glad I could be of assistance. Link to comment Share on other sites More sharing options...
dirtrider478 12 Share Posted October 2, 2020 I know its going to be a bumpy ride with the release of the new client but I love the amount of work im seeing being put into the updates and effort in fixing things that go wrong. A++ Link to comment Share on other sites More sharing options...
Uplift 7 Share Posted October 2, 2020 Is there any monitoring of this service so when I does go down maybe it could automatically be restarted? Link to comment Share on other sites More sharing options...
Pandemic 2702 Author Share Posted October 3, 2020 1 hour ago, Uplift said: Is there any monitoring of this service so when I does go down maybe it could automatically be restarted? There is, however with the recent issues the server never actually went down, it was a more subtle issue that wasn't caught with simple monitoring. This will be more fleshed out in a future uptime service Link to comment Share on other sites More sharing options...
Uplift 7 Share Posted October 3, 2020 34 minutes ago, Pandemic said: There is, however with the recent issues the server never actually went down, it was a more subtle issue that wasn't caught with simple monitoring. This will be more fleshed out in a future uptime service Well, it stopped serving requests, that should make it possible to determine if the service is down, shouldn't it? Link to comment Share on other sites More sharing options...
Pandemic 2702 Author Share Posted October 3, 2020 26 minutes ago, Uplift said: Well, it stopped serving requests, that should make it possible to determine if the service is down, shouldn't it? Like I mentioned above, it was a rather subtle issue, related to TLS that was unfortunately not caught due to a variety of factors. The service sends real requests, and got an appropriate response, even though the outside world had a problem completing the handshake. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.