SUMMARY When using self-hosted kubernetes, dolphin becomes extremely slow. I think this is due to the number of overlay mounts created by kubernetes from /var/lib/containers/storage/overlay/ which dolphin probably try to motorize. STEPS TO REPRODUCE 1. start kubernetes 2. open dolphin OBSERVED RESULT If you try to enter a path in the location bar and are not quick enough to validate with enter, the location path is resetted to initial value. Sometimes dolphin freeze a few seconds. EXPECTED RESULT Same behavior than without a running kubernetes SOFTWARE/OS VERSIONS Linux/KDE Plasma: 5.18.3 (available in About System) KDE Plasma Version: 5.18.3 KDE Frameworks Version: 5.67.0 Qt Version: 5.14.1 ADDITIONAL INFORMATION
The bug is definitively due to the high number of mounts done by kubernetes for pods (I have 127 mounts on a local lab on my laptop). The mounts appears in the devices section on dolphin sidebar. Hiding the devices has no effect. Beside being unusable (lots of lags, freezes) there is also memory leak. Memory consumption of dolphin continually raise every seconds even without making any interaction with dolphin. It ends up by eating all available memory if dolphin stays in the background for a while. Dolphin is working as expected as long as kubernetes is not working.
Some discussion on reddit related to this bug https://www.reddit.com/r/kde/comments/hrtzyg/how_to_exclude_overlayfs_mounts_from_showing_in/
This bug is related to another bugs; https://bugs.kde.org/show_bug.cgi?id=416972 https://bugs.kde.org/show_bug.cgi?id=413110 I also waiting for a fix because using Kubernetes and KDE together is very annoying. The bug is easy to reproduce, so hopefully someone with knowledge will fix it. On the other hand, those bugs were reported some time ago.
If you disable auto-completion in dolphin (right click in the location bar and from the menu select "text completion" -> none), does the location bar work as expected? The memory consumption issues, might be related to bug 398908, could you test again with KIO 5.83?