You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I've been experiencing 'random' CPU spikes for a while now, that in the best case only slow down my system for about two to five seconds. But in the worst case, they've been crashing my system. I'm saying they're random because until two days ago I didn't know what caused them. Now I know that whenever powercord searches for updates, no matter if in the background or user initiated, msmpeng.exe (Windows Defender) spikes its CPU usage upwards of 60% until the update check is over. This causes lags in games but is otherwise unnoticeable (I only see it due to HWiNFO outputting some system readings to my Streamdeck and it shows a big ol CPU graph spike). Sometimes however, it doesn't only spike the CPU, but also causes the system to crash with the error code KERNEL_MODE_HEAP_CORRUPTION, which is a very generic BSOD code. It can range from incompatible drivers to hardware defects up to software conflicts. I know that these CPU spikes are responsible because the system monitoring graphs freeze upon system crash, showing me the beginning of another spike. These crashes have stopped ever since I unplugged powercord from my Discord client. Up until that point, I would experience at least one crash per day, if not more.
I'd just like to know what within the powercord updater could be causing that behavior, maybe similar experiences with other users and potential fixes. I could also attach some screenshots and memory crashdump analytics if needed.
The text was updated successfully, but these errors were encountered:
Powercord's update process is quite simple, and it spawns Git processes to check if there are updates. It's basically automating running git fetch in a command line. During automated updates there's a limit of 2 concurrent git processes allowed to run, for user-initiated ones there is no limit on the amount of git processes allowed to run (designed to make the update don't spawn tons of processes in background, and to be fast when the user requests it).
This looks like a rather specific issue and the problem seems to not directly be from Powercord, and nothing that really can be fixed on our end as spawning processes is a rather common operation.
Hm, interesting. Then I don't understand why Windows defender is making such a fuss about it every single time. At least in my case. The updater itself only ever eats 15 - 20% of CPU during an update check. I even added the whole Powercord folder as an exception and both that and outright disabling Windows defender won't change its behavior. Guess I'll be stuck with updating only once per day manually to avoid these issues for now.
I've been experiencing 'random' CPU spikes for a while now, that in the best case only slow down my system for about two to five seconds. But in the worst case, they've been crashing my system. I'm saying they're random because until two days ago I didn't know what caused them. Now I know that whenever powercord searches for updates, no matter if in the background or user initiated, msmpeng.exe (Windows Defender) spikes its CPU usage upwards of 60% until the update check is over. This causes lags in games but is otherwise unnoticeable (I only see it due to HWiNFO outputting some system readings to my Streamdeck and it shows a big ol CPU graph spike). Sometimes however, it doesn't only spike the CPU, but also causes the system to crash with the error code KERNEL_MODE_HEAP_CORRUPTION, which is a very generic BSOD code. It can range from incompatible drivers to hardware defects up to software conflicts. I know that these CPU spikes are responsible because the system monitoring graphs freeze upon system crash, showing me the beginning of another spike. These crashes have stopped ever since I unplugged powercord from my Discord client. Up until that point, I would experience at least one crash per day, if not more.
I'd just like to know what within the powercord updater could be causing that behavior, maybe similar experiences with other users and potential fixes. I could also attach some screenshots and memory crashdump analytics if needed.
The text was updated successfully, but these errors were encountered: