I used to pick out mirrors manually and had servers very close to me, but I recently started using reflector to automate the process, but the mirrors it chooses is absolute dogshit and gives me really slow speeds.
What am I doing wrong?
I used to pick out mirrors manually and had servers very close to me, but I recently started using reflector to automate the process, but the mirrors it chooses is absolute dogshit and gives me really slow speeds.
What am I doing wrong?
Here’s whats in
/etc/xdg/reflector/reflector.conf# Reflector configuration file for the systemd service. # # Empty lines and lines beginning with "#" are ignored. All other lines should # contain valid reflector command-line arguments. The lines are parsed with # Python's shlex modules so standard shell syntax should work. All arguments are # collected into a single argument list. # # See "reflector --help" for details. # Recommended Options # Set the output path where the mirrorlist will be saved (--save). --save /etc/pacman.d/mirrorlist # Select the transfer protocol (--protocol). --protocol https # Select the country (--country). # Consult the list of available countries with "reflector --list-countries" and # select the countries nearest to you or the ones that you trust. For example: --country Bangladesh,India # Use only the most recently synchronized mirrors (--latest). --latest 5 # Sort the mirrors by synchronization time (--sort). --sort agei believe that is what will be used by reflector if it is enabled/started as a systemd service. but, i think, that’ll not be used if you’re just calling it on the command-line. in that latter case, i think you must supply (those) command-line switches on …the command-line. (maybe)
The systemd service is enabled, and when I did actually apply all these in the command line manually too before enabling the service.
ah. now armed with that extra information, i can only profess cluelessness. perhaps a “–verbose” option, well watched, would reveal something (maybe) good luck!
Been a while since I mucked with reflector, but you don’t seem to be prioritizing faster mirrors whatsoever. Try
--sort rateinstead. If that’s not fast enough I’d also increase your--latestup to maybe 15 so that you have higher odds of a fast mirror being in the group of just-updated mirrors.