So, most of you may be wondering why Bluesky went down.
I’ve made some checks across channels and it seems it is indeed not working as intended.
Not Enough Resources seems to be the common issue here. However, there have also been some regression indicators with the Age Verification reappearing after force closing the Bluesky App in a logged in state.
There’s also this for the website version of Bluesky App (bsky.app). I even saw an error about the usage of HTST instead of HTTPS when accessing bsky.app. Which highlights a possible Certificate based or Firehose MIM/MITM attack but I cannot be certain this is the case and may be how ATProtocol is entered into a safe mode state to deal with the issues currently experienced.
A MIM/MITM attack is a Man-in-the-middle attack where communication is intercepted and altered within the relay stopping the correct responses from occurring on either side. Meaning that the firehose that ATProtocol and all ATProtocol apps rely on is in a manipulated state.
But I’m not the expert.
It could also be possible that the current issue is a DDoS (Distributed Denial of Service) attack on the main ATProtocol servers leading to outright failure to maintain the service when it is a requirement.
I also checked status.bluesky.app the official Status Page for Bluesky. The API servers were in a state of RED, downtime when I visited.
At the bottom of the page was a telling indicator and notice stating an issue published, what seems to be rapid speed by the devs *without* spellchecking.
We are investing an *incident* with (missing “the”) service in one of our *reginos*, which is clearly “regions”.
The current app view needs to show the difference of whether the server, ATProtocol or the users connection is the problem. Currently the default and only error highlighted in the error bar, highlights that the issue is always that of the user trying to connect when their connection is SOLID.
There you have it. It went down like a sack of shit.
And so we continue…
Bluesky is restricting access to or is having its access restricted. Error 402 (Bad Gateway) and Error 403 (Forbidden) seem to be plaguing the infrastructure. However, this has led to a much more stable service. Some “hiccups” seem possible but not as severe.
—– UPDATE 3 —–
Issues have returned and ARE plaguing Bluesky yet again.
This time, no profile data is being retrieved. This now definitely leads to possible a failure of the upstream/firehose as a result of possible MIM/MITM relay attacks.
I’ll also add that Bluesky indicated at status.bsky.com that they added fixes and are monitoring. However, I’m unsure as to whether they were implemented before or after the current upstream failure.
Article written on a mobile device.
