If this issue is due to some write amplification it could be worth to check how much logging the different containers do, and how big the log files are. If a 10MB log file is constantly written to, could it be that instead of writing a few bytes every log update, it rewrites the whole file?
I noticed that I had debug logging enabled in my Plex container (official), so I had 60+MB of logs just for the "Plex Media Server.X.log" files within the last 24h. It looks like Plex creates a new log file when reaching ~11MB, and with debugging it seem to have been printing logs every 1-2 seconds. Re-writing the whole log file every 2 seconds when it's a couple of MB will cause a lot of unnecessary wear.
[6.8.3] docker image huge amount of unnecessary writes on cache
in Stable Releases
Posted
If this issue is due to some write amplification it could be worth to check how much logging the different containers do, and how big the log files are. If a 10MB log file is constantly written to, could it be that instead of writing a few bytes every log update, it rewrites the whole file?
I noticed that I had debug logging enabled in my Plex container (official), so I had 60+MB of logs just for the "Plex Media Server.X.log" files within the last 24h. It looks like Plex creates a new log file when reaching ~11MB, and with debugging it seem to have been printing logs every 1-2 seconds. Re-writing the whole log file every 2 seconds when it's a couple of MB will cause a lot of unnecessary wear.