In Kleopatra we would like to have "gpgconf --kill all" acessible from the GUI. This is helpful for support.
I think it should be placed under "Extras -> Restart background processes"
Even though it would be just a kill all it should be called restart because kleopatra would immediately run a new keylisting starting the processes again.
We thought about doing this also when Kleopatra shuts down on Windows but for now an action item is the least we should do.
| rLIBKLEO Libkleo | |||
| rLIBKLEOb8517cb0a019 Throttle launching of the GnuPG agent | |||
| rLIBKLEOe0864b3a88bf Add helper for shutting down the GnuPG daemons | |||
| rLIBKLEO216cb848f962 Deduplicate code for running gpgconf | |||
| rLIBKLEO2a87d1d356da Unify retrying to connect to the GnuPG agent | |||
| rLIBKLEO04a4478605b9 Add helper for launching the GnuPG agent | |||
| rLIBKLEO8df2fe9a13b9 Add helper to check if gpg-agent is running and accepts connections | |||
| rKLEOPATRA Kleopatra | |||
| rKLEOPATRA120bfb4dd76b Add missing changes to fix last commit | |||
| rKLEOPATRA60a035d48e22 Request start of GnuPG agent if connecting to it fails | |||
| rKLEOPATRA521ee4cbf670 Require current version of libkleo | |||
| rKLEOPATRA8cda5a03a97c Add action for restarting the GnuPG daemons | |||
| rKLEOPATRA183e3f72ec26 Wait a bit before restarting the connection | |||
| rKLEOPATRAe7d903e3bd7e Unify retrying to connect to the GnuPG agent | |||
| rKLEOPATRA2f504289f748 Reset the Assuan context if we detect a broken pipe | |||
Note: Currently, killing the background processes causes a SIGPIPE (broken pipe) in the worker thread of the DeviceInfoWatcher. Kleopatra seems to survive this, but I'm not sure the thread survives. Starting a new SCD DEVINFO --watch fails with General error. On exit, the thread then receives a SIGABRT which crashes Kleopatra.
Oh, this is something we should fix anyway because users when evaluating Kleopatra and making configuration changes regularly run "gpgconf --kill all" anyway. Could it be that the SCD DEVINFO --watch fails because the gpg-agent is not yet started again?
This works in so far that the background processes are killed with the restart button. For remaining issue see T6567