3 ms·
> Printing is a mess and our industry should be ashamed. How does something like this just break after an update? I have Bose wireless headphones. They used to
by thaumasiotes 6y ago
> Printing is a mess and our industry should be ashamed. How does something like this just break after an update?
I have Bose wireless headphones. They used to operate as expected: leave bluetooth on at all times on the laptop. When turning the headphones off, they disconnect. When turning the headphones on, they reconnect.
Eventually there was a Windows update, and this stopped working. Now, if I turn the headphones off, only the microphone will reconnect when I turn them back on. To get sound to play through the headphones, I have to disable and then reenable bluetooth in Windows. Every time.
- akersten 6y agoYou know what's funny is that Microsoft does a great job with what I call "community QA" on some of their other properties, like Minecraft. The Minecraft bug tracker[0] is readily accessible and the community around it tracks down most bugs before they hit retail via their "snapshot" release cycle. The Windows side, not so much. The words "Windows Insider" come to mind, but honestly, I have no idea how that works. It doesn't feel like there's a community around it, or if there is, it must be small. If I run into a bug post-update on Windows, I certainly don't feel empowered to pop open a Jira like I do for Minecraft. I get that it's a different scale and all - but it seems like Microsoft should really put some effort into crowdsourcing their bug reporting if they're not going to invest thoroughly in QA/regression testing for their flagship product. [0]: https://bugs.mojang.com/projects/MC/issues https://bugs.mojang.com/projects/MC/issues
- seniorgarcia 6y agoHave you checked if this was possibly because of https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2019-2102 https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2019-2102? I mean, I get why you would blame MS for breaking this but maybe the onus is also on Bose and their ridiculously bad driver support.
- thaumasiotes 6y agoI haven't checked. How would I do that? The headphones are able to connect once after disabling and reenabling bluetooth in Windows. If the problem were that the headphones are configured to use a fixed, predictable long-term key, it isn't obvious to me why Windows would accept that key sometimes, but not all the time. (Note that I don't need to have Windows forget the device and re-pair from scratch.) Is there some aspect of initial bluetooth negotiation that would explain this?
- seniorgarcia 6y agoThere was a list somewhere with devices using the wrong config. An explanation of the issue is here https://www.bleepingcomputer.com/news/security/microsoft-blocks-some-bluetooth-devices-due-to-security-risks/ https://www.bleepingcomputer.com/news/security/microsoft-blo... and there also is advice on checking if your device is affected. I thought the issue was that if the host initiates the connection it will work but if it's just auto reconnecting it will not but I may be wrong.
- thaumasiotes 6y agoDoesn't look like the CVE is relevant. That article says it's indicated by this message: > Your Bluetooth device attempted to establish a debug connection. The Windows Bluetooth stack does not allow debug connection while it is not in the debug mode. I don't see that. What I see is > A device with Bluetooth adapter address (2c:45:a1:20:a1:99) has a bad service record (attribute identifier 9). The service will not be visible. That happens several times every time, whether the connection succeeds or not. There is also a message that appears just once, when the connection succeeds: > Windows cannot store Bluetooth authentication codes (link keys) on the local adapter. Bluetooth keyboards might not work in the system BIOS during startup. Searching for the first message, I found some support threads from Bose customers. They seem pretty angry about their headphones not working. :/