"No appdata directory found for 'my-container'. Skipping..."

[DesertCookie]

I have set up the script as pre-container-backup- and post-run-scripts. Running it I get above's message for every container:

This is the setup section as I have configured it:

################################################################################
#                            USER CONFIG (GENERAL)                             #
################################################################################
APPDATA_SOURCE_DATASET="cache/appdata"  # Source dataset of your appdata.
LOG_TO_STATUS_PAGE=true  # Set to 'true' to also display log messages on the Appdata.Backup "Status/Log" Web GUI page.
BETA_VERSION=false  # Set to 'true' if using the BETA version of Appdata.Backup plugin.
################################################################################
#                        USER CONFIG (SNAPSHOTS/SANOID)                        #
################################################################################
SANOID_CONFIG_DIR="/etc/sanoid"  # Directory of your default sanoid config files. They should already be located at "/etc/sanoid".
# Each docker containers dataset that is processed will have a sub-folder created for it (using the dataset basename) within SANOID_CONFIG_DIR containing it's own sanoid config files.
# Set sanoid retention policy below. "How many X of each timeframe will be kept before deleting old snapshots of said timeframe?" Snapshot pruning is done in 'POST-RUN' to minimize docker downtime.
SNAPSHOT_HOURS="0"
SNAPSHOT_DAYS="7"
SNAPSHOT_WEEKS="4"
SNAPSHOT_MONTHS="3"
SNAPSHOT_YEARS="0"
SANOID_CONFIG_UPDATE=true  # Set to 'true' to have the script *also* automatically update the config files with any changes made to the retention policy above.
ALLOW_SNAPSHOTS_OUTSIDE_OF_RETENTION_POLICY=false  # sanoid will not take new snapshots if ran before its next retention policy interval. Set to 'true' to allow additional '_extra' snapshots to be taken.
################################################################################
#                            USER CONFIG (POST-RUN)                            #
################################################################################
# Post-Run processes the docker containers that are set to 'skip = no' in Appdata.Backup config. For the tarfile option, the docker containers must also to be set to 'skip Backup = yes'.
# DELETE_EXTRA_SNAPSHOTS_OLDER_THAN_X_DAYS="7"  # Uncomment this line to delete any '_extra' snapshots (taken when ALLOW_SNAPSHOTS_OUTSIDE_OF_RETENTION_POLICY=true) that are older than this many days.
# Below you can define a list of any extra datasets (full paths) that you want to snapshot and/or replicate and/or rsync from the most recent snapshot. (sanoid config sub-directory will be created as the basename of each dataset)
EXTRA_DATASETS=(  # WARNING: extra datasets will be replicated to 'DATASET_TO_REPLICATE_TO/[basename of extra dataset]' (ie 'pool_main/some_share/stuff' to 'pool_two/backup_appdata/stuff')
  # "pool_main/appdata/plexato"
  # "pool_main/some_share/stuff"
)
SNAPSHOT_EXTRA_DATASETS=false  # Take snapshots of the EXTRA_DATASETS.
# For the REPLICATE option, uncomment the line below and define the name of the parent dataset that you want to replicate the source datasets to. THIS DATASET(S) WILL BE CREATED AUTOMATICALLY if it does not exist!
# DATASET_TO_REPLICATE_TO="pool_ssds/backup_appdata"
REPLICATE_CONTAINERS=false  # Replicate the datasets/snapshots of docker containers appdata.
REPLICATE_EXTRA_DATASETS=false  # Replicate the datasets/snapshots of EXTRA_DATASETS.
MOUNT_REPLICATED_DATASETS=false  # If all replicated datasets need to be mounted before replication *and* unmounted after replication.
SYNCOID_ARGS="-r --delete-target-snapshots --force-delete --no-sync-snap --quiet"  # OPTIONALLY (and carefully) customize the syncoid command line arguments. See: https://github.com/jimsalterjrs/sanoid/wiki/Syncoid#options
# For TAR and RSYNC options, uncomment the line below and specify the permanent dataset to clone recent snapshots to temporary datasets (ie 'pool_main/temp/_temp_plex'). THIS DATASET(S) WILL BE CREATED AUTOMATICALLY if it does not exist!
# DATASET_TO_TEMP_CLONE_TO="pool_main/temp"  # WARNING: any datasets that start with '_temp_' within this dataset will be DESTROYED when the script is ran!
# Tar datasets contents from most recent snapshots. Tarfiles are saved to the same generated backup folder that Appdata.Backup would save them to. 
TAR_CONTAINERS=true  # Tar containers from most recents snapshots. Tar compression and exclude settings are inherited from Appdata.Backup config.
TAR_EXTRA_DATASETS=false  # Tar EXTRA_DATASETS from most recent snapshots.
# Rsync datasets contents from most recent snapshots. Folders are created in the same generated timestamped backup folder that Appdata.Backup saves the tarfiles to. 
RSYNC_CONTAINERS=false  # rsync containers appdata from most recent snapshots. Exclude settings are inherited from Appdata.Backup config.
RSYNC_EXTRA_DATASETS=false  # rsync EXTRA_DATASETS from most recent snapshots.
################################################################################
#                              END OF USER CONFIG                              #
################################################################################

My disk setup is comprised of single-disk ZFS:

I'd appreciate some insight into what might be the issue here.

[Blasman]

I am not sure what the issue is. I can only assume that it is coming from line 213 as that is the only time grep uses the -P option.

Do the appdata directories for each docker actual exist? Perhaps insert a debug line after line 213 that checks what is being found for $matched_line ?

ab_log "[DEBUG] matched_line == '$matched_line'"

[DesertCookie]

I've tried that. The output I get is always: [DEBUG] matched_line == ''.

Something is clearly wrong somewhere. I do not know if it's my setup, though.

[Blasman]

I am still using Unraid v6.12.11 since I wrote the script. It is possible that something is different with newer versions of Unraid if you happen to be using a newer version. You are the first person to provide feedback so I do not have much to go on. I would debug what is being matched for $appdata_source_path and $docker_name instead of $matched_line then as that is what $matched_line is referencing. Make sure that the files/directories exist that $matched_line is referencing in line 213.

ab_log "[DEBUG] appdata_source_path == '$appdata_source_path'"
ab_log "[DEBUG] docker_name == '$docker_name'"

[DesertCookie]

I, for example, get this for one container:

[📸] [DEBUG] appdata_source_path == '/mnt/user/appdata
/mnt/cache/appdata'
[📸] [DEBUG] docker_name == 'AuthentikDB'

Perhaps an issue with there being two paths for appdata, seemingly returned as an array?

I use unRAID 7.2.0.

[Blasman]

Oh, that would make sense. It's been over a year since I wrote this script and I haven't had a reason to come back to it until now. I did state in the limitations that "Only works for the first appdata source that is specified in Appdata.Backup config."

However, I assumed that meant that any non-primary appdata sources would be ignored, rather than it break the script. Do you need the /mnt/cache/appdata to be specified in Appdata Backup config or will just specifying the first source work? The var is set on line 80 so you could just hardcode a path in there instead. If you need both appdata sources then the script would need to be modified to cycle through the sources.

[DesertCookie]

No, I do not need the /mnt/cache/appdata. I assumed it had always been there. Perhaps past me set that many years ago.

I'll try without it now.