Summary: | plasmashell hangs for multiple seconds while nmcli is also slow to respond | ||
---|---|---|---|
Product: | [Plasma] plasmashell | Reporter: | Alexander Potashev <aspotashev> |
Component: | Networks widget | Assignee: | Jan Grulich <jgrulich> |
Status: | RESOLVED DUPLICATE | ||
Severity: | normal | CC: | kdebugs |
Priority: | NOR | ||
Version: | master | ||
Target Milestone: | 1.0 | ||
Platform: | Fedora RPMs | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: | |||
Attachments: | stacktrace when plasmashell hangs |
Description
Alexander Potashev
2019-08-12 19:19:34 UTC
The nmcli unresponsiveness might be due to horrible problems with my VPN server (DDoS attack against the data center) which I connect to using networkmanager+openvpn. Created attachment 124557 [details]
stacktrace when plasmashell hangs
how to reproduce:
1. connect to Wi-Fi with pre-configured secondary VPN connection
2. disconnect WAN cable from the Wi-Fi access point
3. click on the NetworkManager icon in system tray -> plasmashell hangs
I can confirm this bug, the way I can reproduce it specifically is to use stubby as a local DNS resolver. 1. Install stubby and configure it 2. Set DNS resolver to 127.0.0.1 3. Switch from one wifi to the other 4. The entirety of plasmashell will now freeze as NetworkManager waits for a dns response to time out Note that restarting stubby unfreezes plasmashell and NetworkManager prematurely (i.e. before the timeout happens) The general problem here appears to be threefold, 1. plasmashell lets itself get blocked by applets, 2. plasma-nm lets itself get blocked by NetworkManager, 3. NetworkManager is not good. Ideally plasmashell would not freeze if a widget freezes, and even more ideally plasma-nm won't freeze if NetworkManager decides to pointlessly block. *** This bug has been marked as a duplicate of bug 409392 *** Bulk transfer as requested in T17796 |