John Spurlock
banner
johnspurlock.com
John Spurlock
@johnspurlock.com
Working on stuff 👨‍💻

Check out Firesky to watch the Bluesky firehose:
👉 https://firesky.tv

🐦 https://twitter.com/johnspurlock
🐘 https://podcastindex.social/@js
iOS 26 has about 26% share after 10 weeks

26.1 has been public for three weeks
November 24, 2025 at 10:33 PM
It does look like that end of day download swell yesterday was due to one show:

True Crime with Kendall Rae

op3.dev/show/5db037e...

Overlayed in yellow to highlight its contribution to yesterday's downloads
November 19, 2025 at 2:57 PM
No visible "apps retrying/catching up" as far I can tell.

There is a large swell late yesterday, likely due to one of the larger shows measured by OP3 releasing a new episode, will do more digging to confirm that.
November 19, 2025 at 2:41 PM
Overall totals don't tell the whole story though, here is a comparison of both days by hour, and also by ten-minute.

Here we can see the exact outage impact times (around 7am Eastern, then from 8am to 9.30am as initially assumed yesterday) - and otherwise very similar pattern.
November 19, 2025 at 2:41 PM
In terms of absolute daily non-bot podcast downloads measured by OP3, yesterday's numbers were actually higher than last Tuesday (GMT day, 7pm to 7pm Eastern, nice in that the outage lands around midday)

2025-11-11: 607,804
2025-11-18: 622,982
November 19, 2025 at 2:39 PM
Let's try to get our heads around yesterday's outage in terms of podcast downloads.

It's always hard to compare apples to apples, but things generally follow a weekly cycle.
November 19, 2025 at 2:39 PM
- New SwiftUICore.VisualEffect._makeTransform (not usable by 3p apps)
- and some other minor age-restriction and family control stuff

overall, not a big api surface change this time around
November 18, 2025 at 7:54 PM
‘A latent bug in a service underpinning our bot mitigation capability started to crash after a routine configuration change we made. That cascaded into a broad degradation to our network and other services. This was not an attack.’

x.com/dok2001/stat...
Dane Knecht 🦭 on X: "I won’t mince words: earlier today we failed our customers and the broader Internet when a problem in @Cloudflare network impacted large amounts of traffic that rely on us. The sites, businesses, and organizations that rely on Cloudflare depend on us being available and I" / X
I won’t mince words: earlier today we failed our customers and the broader Internet when a problem in @Cloudflare network impacted large amounts of traffic that rely on us. The sites, businesses, and organizations that rely on Cloudflare depend on us being available and I
x.com
November 18, 2025 at 4:08 PM
based on my telemetry it looks like op3 redirects went down completely for a short time at 7am eastern, then again at 8am to 9.30am eastern, then back to normal

not going to deploy anything until cloudflare fixes `several issues that remain post-deployment`

www.cloudflarestatus.com/incidents/8g...
November 18, 2025 at 3:58 PM
op3.dev is still down due to the Cloudflare outage, this is really the first outage since it started in 2023. First time I've seen Cloudflare edge locations down like this globally.

www.cloudflarestatus.com/incidents/8g...
November 18, 2025 at 1:52 PM
Looks like its location-specific

Here's a view from the last six hours, redirects are back but at a lower level right now.

Wonder if I can break this out by location...
November 18, 2025 at 1:02 PM