Static noise when any audio is playing - How to debug?

Hello! Today, after updating the flake and rebuilding the system on my PC, I’ve noticed a static noise buzzing in the “background” whenever any audio is playing; be it a system notification, the audio in a game, or when playing music. There is no issue with the audio itself, except for the static playing with it. The static can best be heard when lowering the volume, as the volume of the normal audio gets lower, while the static noise remains basically the same.

For some more context, I’m running the unstable branch, and this is the first time updating my system in well over a month, since I have been away from this computer for some time now. This means that pretty much everything on the system was updated, which makes it a bit more difficult to isolate what updated software might be causing this, etc. The rev before updating was ec942ba042dad5ef097e2ef3a3effc034241f011, last updated May 28, 2026

My question is as follows: How would one go about debugging this? Where should I start looking to figure out what might be causing this? Just searching on the internet yielded nothing. I’d love to learn what process all of you would use to get to the bottom of an issue like this.

I’d start by checking pwvclient or pavuclient to see what applications are playing audio.

If there is no visible noise in that output, it’s likely a hardware issue. Either wire noise, some kind of interference, or motherboard issues.

1 Like

Hmm, I couldn’t really find anything called pwvclient or pavuclient. The closest I could find was something called pwvucontrol and pavucontrol, and those didn’t make me much wiser. All they really gave me was a volume bar going up and down with audio playing, and although it did look a bit more jittery than usual, it could just have been my imagination.

As for hardware issues, I’d be really surprised if it was one, as when I rollback to the previous generation the issue isn’t there. It’s why I suspect this being the result of a bug introduced with some update or change.

Is there maybe a convenient way to check which packages that have changed between flake updates? I know that I said that this is the first update in over a month, so most likely a bunch of stuff has changed, but maybe I can at least narrow it down and start by looking at any changes made to audio related stuff like pipewire, pulseaudio, etc

Ah, right, that’s more interesting then. I’d guess kernel changes. Is this bluetooth, audio jack, HDMI, USB-C, …?

You can diff the system closures. dix is probably the best utility for that at the moment.

1 Like

It was indeed the kernel! After changing kernel from Linux 7.1.3 to Linux 6.18.3, I no longer have the issue. Thank you! And out of interest to learn; what made you suspect that this was a kernel related issue?

It is a wireless headset connected to the computer via a USB A dongle. A pair of Logitech G935.

Thank you! Didn’t need it this time around, but it seems like a pretty useful tool. Will keep it for later haha

All of that aside, would it be a good idea for me to report the kernel issue that I’m experiencing? If so, where and how?

Well, you confirmed it was a software issue since downgrades solve it. You mostly confirmed that there was no weird new application adding noise on accident.

Large update gaps probably mean a kernel update was involved. The kernel is heavily involved at this sandwich layer between hard-and software. BlueZ & co is especially commonly the culprit for weird sound problems.

So, process of elimination and educated guesses?

That’s the tricky part! You’d have to figure out which component is responsible for the issue (probably the driver of either your headset or the dongle, but who knows), and then figure out where on the kernel mailing lists to report this (and search for threads that already deal with this issue).

Trying to contribute to the kernel is quite the ordeal. I largely don’t bother myself. If the issue is still present in a kernel version or two, consider trying to figure it out.

This does probably mean tricky, consumer-facing issues like this one are sadly quite likely to go under the radar for a while. On the flip side, the kernel mailing list serves as a low-pass filter for people who know too little to actually help out, so it stays a bit more manageable. Gate-keeping, I know, but well, for some topics maybe that’s for the best.

1 Like

@MichaelCode25 The Linux kernel provides a specific document on how to report a regression: Reporting regressions — The Linux Kernel documentation.

1 Like