Not working when i restart the mc server

[JordiKing10]

I use a script with rcon-cli to restart the server of minecraft and if i do when mc-backup is working, dont work anymore. When the server is up again cant connect so i restart both at once or only the mc-backup container when mc server is UP.

Here is the debug:

• : /data 
• : /backups
• : Vault-Hunters
• : 2m
• : 4h
• : true
• : 10m
• : tar
• : gzip
• : '-3 --long=25 --single-thread'
• : 1
• : '--keep-within 7d'  5
• : mc  5
• : ''
• : serverjk-mc
• : -1
• : 10s
• : '*.jar,cache,logs'  e
• : mc_backups
• : /config  e
• : Europe/Madrid
• : gzip
• : ''
• : ''
• : ''
• : ''
• : ''
• : ''
• export TZ
• export RCON_HOST
• export RCON_PORT
• export RCON_PASSWORD
• export XDG_CONFIG_HOME
• export SRC_DIR
• export DEST_DIR
• export BACKUP_NAME
• [[ -n '' ]]
• [[ -n '' ]]
• [[ -n '' ]]
• '[' -n '' ']'
• '[' '!' -d /data ']'
• is_function tar
• '[' 1 -ne 1 ']'
• name=tar ++ type -t tar
• '[' function == function ']'
• readarray -td, excludes_patterns ++ printf %s '*.jar,cache,logs'
• excludes=()
• for pattern in "${excludes_patterns[@]}"
• excludes+=(--exclude "${pattern}")
• for pattern in "${excludes_patterns[@]}"
• excludes+=(--exclude "${pattern}")
• for pattern in "${excludes_patterns[@]}"
• excludes+=(--exclude "${pattern}")
• tar init
• call_if_function_exists init
• '[' 1 -lt 1 ']'
• function_name=init
• is_function init
• '[' 1 -ne 1 ']'
• name=init ++ type -t init
• '[' function == function ']'
• init
• mkdir -p /backups
• case "${TAR_COMPRESS_METHOD}" in
• tar_parameters=('--gzip')
• readonly tar_parameters
• readonly backup_extension=tgz
• backup_extension=tgz
• is_one_shot
• [[ FALSE = TRUE ]]
• return 1
• log INFO 'waiting initial delay of 2m...'
• '[' 2 -lt 1 ']'
• local level=INFO
• shift
• valid_levels=('INFO' 'WARN' 'ERROR' 'INTERNALERROR')
• local valid_levels
• is_elem_in_array INFO INFO WARN ERROR INTERNALERROR
• '[' 5 -lt 2 ']'
• local element=INFO
• shift  e
• for e in "$@"
• '[' INFO == INFO ']'
• return 0
• '[' 1 -ge 1 ']'
• cat -
• awk -v level=INFO '{ printf("%s %s %s\n", strftime("%FT%T%z"), level, $0); fflush(); }' 2022-09-01T16:35:51+0200 INFO waiting initial delay of 2m...
• '[' INFO == INTERNALERROR ']'
• sleep 2m
• log INFO 'waiting for rcon readiness...'
• '[' 2 -lt 1 ']'
• local level=INFO
• shift
• valid_levels=('INFO' 'WARN' 'ERROR' 'INTERNALERROR')
• local valid_levels
• is_elem_in_array INFO INFO WARN ERROR INTERNALERROR
• '[' 5 -lt 2 ']'
• local element=INFO
• shift  e
• for e in "$@"
• '[' INFO == INFO ']'
• return 0
• '[' 1 -ge 1 ']'
• cat -
• awk -v level=INFO '{ printf("%s %s %s\n", strftime("%FT%T%z"), level, $0); fflush(); }' 2022-09-01T16:37:51+0200 INFO waiting for rcon readiness...
• '[' INFO == INTERNALERROR ']'
• retry -1 10s rcon-cli save-on
• '[' 4 -lt 3 ']'
• local retries=-1
• local interval=10s
• readonly retries interval  2
• (( retries < 0 ))
• local retries_msg=infinite 
• local i=-1
• (( retries >= ++i ))
• '[' infinite '!=' -1 ']' ++ timeout --signal=SIGINT --kill-after=30s 5m rcon-cli save-on ++ tr '\n' '\t'
• output='2022/09/01 16:37:51 Failed to connect to RCON serverdial tcp: lookup mc on 127.0.0.11:53: read udp 127.0.0.1:43190->127.0.0.11:53: read: connection refused '
• log ERROR 'Unable to execute rcon-cli save-on - try 0/infinite. Retrying in 10s'
• '[' 2 -lt 1 ']'
• local level=ERROR
• shift
• valid_levels=('INFO' 'WARN' 'ERROR' 'INTERNALERROR')
• local valid_levels
• is_elem_in_array ERROR INFO WARN ERROR INTERNALERROR
• '[' 5 -lt 2 ']'
• local element=ERROR
• shift  e
• for e in "$@"
• '[' ERROR == INFO ']'
• for e in "$@"
• '[' ERROR == WARN ']'
• for e in "$@"
• '[' ERROR == ERROR ']'
• return 0
• '[' 1 -ge 1 ']'
• cat -
• awk -v level=ERROR '{ printf("%s %s %s\n", strftime("%FT%T%z"), level, $0); fflush(); }' 2022-09-01T16:37:51+0200 ERROR Unable to execute rcon-cli save-on - try 0/infinite. Retrying in 10s
• '[' ERROR == INTERNALERROR ']'
• '[' -n '2022/09/01 16:37:51 Failed to connect to RCON serverdial tcp: lookup mc on 127.0.0.11:53: read udp 127.0.0.1:43190->127.0.0.11:53: read: connection refused ' ']'
• log ERROR 'Failure reason: 2022/09/01 16:37:51 Failed to connect to RCON serverdial tcp: lookup mc on 127.0.0.11:53: read udp 127.0.0.1:43190->127.0.0.11:53: read: connection refused '

...

[itzg]

What you're attempting should be fine.

It's actually running into a DNS problem since this is UDP to port 53

lookup mc on 127.0.0.11:53: read udp 127.0.0.1:43190->127.0.0.11:53: read: connection refused

[JordiKing10]

A dns error on the same machine? What could be the solution? The thing is that I use your rcon client to restart the server periodically and I have noticed that after each restart the mc-backups container remained stopped. I had thought about forcing the automatic start but it would create a backup every day after the reboot

Anyway thanks for your help!

[itzg]

Docker is providing DNS internally to resolve service names to container IP address. Google for that particular symptom to look for solutions.

The docs say to set RCON_RETRIES to a negative value to have it retry indefinitely, if you're wanting to prevent the container from stopping on failures.