Crontab restart/repair script for Plex "Sqlite3: Sleeping for 200ms to retry busy DB"

plexdb_healthcheck.sh

#!/bin/bash

# Plex busy DB workaround
# Set the maximum number of allowed occurrences of the Sqlite3: Sleeping for 200ms to retry busy DB string
# about 3 times usually indicates an instance that can't recover by itself
# if your instance restarts your docker logs should reset, or "cover up" the busy DB lines, afaik 
MAX_OCCURRENCES=3

# this part "docker logs --tail 10 plex" "--tail 10" means the last ten lines of the container logs
# of the container named "plex", change the name according to your setup
# you can try "docker logs --tail 10 plex" in the terminal and if everything is setup correct
# it will spit out the last 10 lines of the logs and should work
LOG_LINES=$(docker logs --tail 10 plex)

# Count the number of occurrences of the specified string in the log lines
OCCURRENCES=$(echo "$LOG_LINES" | grep -o "Sqlite3: Sleeping for 200ms to retry busy DB." | wc -l)

# If the health check fails, restart the plex container
if [ "$OCCURRENCES" -gt "$MAX_OCCURRENCES" ]; then
  # Restart the container
  # Again this part "docker restart plex" assumes that your container is named plex, change the name according to your setup
  docker restart plex
fi

Im trying to use the script in unraid via the user scripts. is this the only part I need to modify "docker restart Plex-Media-Server" with the name of my container

Did you ever figure this out? As I have 2 plex instances I want it to watch both

you need two instances of the script in your case plexdb1_healthcheck.sh and plexdb2_healthcheck.sh for example. both need their own crontab entry. Also yes there are two cases in the script where you need to rename according to you instances name. I extended the commentary and explanations within the script

Thank you so much for your reply. So I have 2 container "plex" and "plex2" I made a second because I share with my family out of state and the first one gets this error so often I have the second so there is always a running instance. I even moved my appdata to the cache and it didn't fix it. I'm using unraid so I set this up as a userscript and used a cronjob to run it every minute. It seems to work fine on plex. So what exactly so I need to change for plex2 to work. It just reads the docker image right so no need to like link to the appdata or anything like that? I appreciate your help.