781–790 of 841 posts
Earlier quoted context omitted.
Spam those Azure tickets. If you have a CSAM, build them a nice powerpoint telling the story of all your AFD issues (that's what they are there for). > In 50%+ the cases they just don‘t report it anywhere, even if its for 2h+. I assume you mean publicly. Are you getting the service health alerts?
Back when we used Azure the only outcome was them trying to upsell us on Premium Support
Earlier quoted context omitted.
Is voting there a one day only event? If not, I feel the solution to that particular problem is quite clear. There’s a million things that could go wrong causing you to miss something when you try to do it in a narrow time range (today after work before polls close) If it’s a multi day event, it’s probably that way for a reason. Partially the same as the solution to above.
In europe, voting typically happens in one day, where everyone physically goes to their designated voting place and puts papers in a transparent box. You can stay there and wait for the count at the end of the day if you want to. Tom Scott has a very good video about why we don't want electronic/mail voting: https://www.youtube.com/watch?v=w3_0x6oaDmI
Interesting that everybody knows when AWS goes down but Azure needs a "Tell HN" :) Best of luck to the teams responding to this incident.
I noticed that Starbucks mobile ordering was down and thought “welp, I guess I’ll order a bagel and coffee on Grubhub”, then GrubHub was down. My next stop was HN to find the common denominator, and y’all did not disappoint.
I noticed it when my Netatmo rigamajig stopped notifying me of bad indoor air quality. Lovely. Why does it need to go through the cloud if the data is right there in the home network…
However netatmo does need to have a server to store data as you need to consolidate acreoss devices plus you can query gfor a year's data and that won't and can't be held locally.
Preliminary post incident review: https://azure.status.microsoft/en-gb/status/history/ Timeline 15:45 UTC on 29 October 2025 – Customer impact began. 16:04 UTC on 29 October 2025 – Investigation commenced following monitoring alerts being triggered. 16:15 UTC on 29 October 2025 – We began the investigation and started to examine configuration changes within AFD. 16:18 UTC on 29 October 2025 – Initial communication po…
33 minutes from impact to status page for a complete outage is a joke.
Earlier quoted context omitted.
Apologies, but this just reads like a low effort critique of big things. To be clear, they should get criticism. They should be held liable for any damage they cause. But that they remain the biggest cloud offering out there isn't something you'd expect to change from a few outages that, by most all evidence, potential replacements have, as well? More, a lot of the outages potential replacements have are often more g…
I would say you are explaining why they get a free pass so they still get one - they are bad but their main competitors are even worse! I thought one of the major selling points of the big cloud providers was that they were more reliable than running your own stuff (by which i mean anything from a VPS to multiple data centres depending on your scale. Compared to those alternatives they seem to be less reliable in pra…
You are further asserting that these outages prove they are not still more reliable than home spun. Is that the case? More than a few people aren't ready for a single hard drive to crash on the stuff they are doing.
Earlier quoted context omitted.
Good thing HN is hosted on a couple servers in a basement. Much more reliable than cloud, it seems!
Just don't use genetically identical hardware: https://news.ycombinator.com/item?id=32031639 https://news.ycombinator.com/item?id=32032235 Edit: wow, I can't believe we hadn't put https://news.ycombinator.com/item?id=32031243 in https://news.ycombinator.com/highlights . Fixed now.
Preliminary post incident review: https://azure.status.microsoft/en-gb/status/history/ Timeline 15:45 UTC on 29 October 2025 – Customer impact began. 16:04 UTC on 29 October 2025 – Investigation commenced following monitoring alerts being triggered. 16:15 UTC on 29 October 2025 – We began the investigation and started to examine configuration changes within AFD. 16:18 UTC on 29 October 2025 – Initial communication po…
33 minutes from impact to status page for a complete outage is a joke.
The reason is probably because changes to the status page require executive approval, because false positives could lead to bad publicity, and potentially having to reimburse customers for failing to meet SLAs.
Earlier quoted context omitted.
Just don't use genetically identical hardware: https://news.ycombinator.com/item?id=32031639 https://news.ycombinator.com/item?id=32032235 Edit: wow, I can't believe we hadn't put https://news.ycombinator.com/item?id=32031243 in https://news.ycombinator.com/highlights . Fixed now.
I love that "Ask HN: What'd you do while HN was down?" was a thing