This post is created with the hope of helping me to better understand how the POTD code works and its interaction with the providers. My Slackware (distribution) systems recently had an outage with the NASA (APOD) provider. There was no issue connecting with the other providers (e.g. Bing, NOAA, National Geographic, Flickr, USRA, etc.). One of the things that puzzled me was this was true on both my development systems (aka “current”) as well as the stable release (Slackware 15.0). The outage with NASA/APOD provider seemed to occur after June 28th (2023) - the last working wallpaper image. The connection with this provider was working when I returned from my travel on July 16th. So it must have been restored sometime before this. I had shut down all my systems when I left for my travel on July 7th. So the outage was definitely at least from June 29th to July 8th. I had created cron scripts to fetch and apply the wallpaper with wget and plasma-apply-wallpaperimage executables. The trouble with cron of course was getting DBUS permissions set. I had to export a slew of environment variables and restart the kactivitymanager daemon and plasmashell service. Even this didn’t work quite right to get the plasma-apply-wallpaperimage binary to execute from cron.
So I have to think that the POTD code must fetch the wallpaper images from the provider and then post them to a temporary site like somewallpaperplace.freedesktop.org. Then the code has to run the get script (wget, lftp, etc.), download it to a temp place and the finally apply the wallpaper.
It would be wonderful if the code can be modified to dump a log file to avert or understand outages like this in the future.
Can someone (a developer) please clarify the exact sequence of what I described above so I can better understand what broke?
Thanks in advance!