[EUNE] Instabilities over the last week

Edit: It occurred to me while reading the comments that I was not clear on the current status of things - there are **no current issues** with EUNE. It is working properly - if you're having issues, you may want to report/discuss on the [Help & Support](http://boards.euw.leagueoflegends.com/en/c/help-support-en) boards or by contacting [Player Support](https://support.riotgames.com/hc/en). Hi everyone, Over the last week, EUNE has been particularly hit with rough instabilities and we wanted to make sure we communicated on these to give you a little bit more details. Riot Tmx briefly touched point on this already [here](http://boards.euw.leagueoflegends.com/en/c/help-support-en/cgAkdeJF-2-major-incidents-on-eune-i-4-days-seriously?comment=0002), but that was yesterday and since then we’ve had another issue in the morning. To recap what’s already on his post – the first issue last week (on Wednesday, May 4th) was due to a hardware failure that took us a while to identify – primarily because failure points would be in multiple places, and only after confirming through carefully diagnosis did we manage to find the actual problem. Then came the replacement of the faulty hardware that needed a bit of time – but had we known right away what was wrong, impact wouldn’t have been that long. Loss forgiven was activated during that incident. On May 9th, we had chat and loot issues for a few servers (EUNE included – but also NA, OCE and PBE). This was traced down to a software issue that essentially got prompted by another hardware issue (not a failure this time) – this needed a few services to be restarted (like chat, loot & friends list) – but the impact was relatively limited and games did not get impacted. Later on the same day, less than an hour apart, we received reports of the store being inaccessible, logging in taking longer than usual, and chat being slow. This was eventually traced down to load balancing issues on the way connections were being received (essentially, the system that handles how to distribute requests evenly to all available hardware) – which itself was tied down to an unusually high amount of connection requests to a specific part. This was mitigated and eventually resolved. Again, players in games were not affected – but it is easy to imagine that getting disconnected could have lead to complications to come back online. Loss forgiven was activated during that incident. Then finally, this morning, following the issues seen on May 9th, a change was performed on the infrastructure to help with the issue that we had seen on May 4th (the worst one of them all, really). But that change sadly resulted in players being unable to log in, because of a network protocol that was incorrectly set and resulted in addresses not being properly resolved. Once the cause was found, resolution came followed quickly. Now – having 4 incidents in such a short amount of time is not something we consider acceptable, and the teams involved have been working tirelessly to conduct retrospectives and figuring out what can be done to add more resilience in case of such incidents – stability on the server before this week will have shown that it is a rare occurrence, but this does not make it OK. And so we will definitely make sure that efforts are made to detect issues earlier, prior to them causing you all negative impact. I hope this helps put a little bit more context as to what happened over the last week! Thank you all for your patience and understanding.
Share
Report as:
Offensive Spam Harassment Incorrect Board
Cancel