curl error message

[ElToro1966]

I keep getting this at the end when running clickhouse-backup using sftp, both for local and remote, incremental and full:

...
2025/02/11 10:40:24.862567  info clickhouse connection closed logger=clickhouse
curl: no URL specified!
curl: try 'curl --help' or 'curl --manual' for more information

My config file looks like this:

general:
 remote_storage: sftp
 backups_to_keep_local: 7 # keep 7 last backups locally.
 backups_to_keep_remote: 0 # sftp server is responsible for cleanup.
clickhouse:
 username: default
 password: ************
 host: localhost
 port: 9000
 disk_mapping: {}
 skip_tables:
  - system.*
  - INFORMATION_SCHEMA.*
  - information_schema.*
  - _temporary_and_external_tables.*
sftp:
 address: 10.0.0.25
 username: backup 
 password: ************
 key: "/home/clickhouse/.ssh/clickhouse_backup"
 port: 22
 path: /mnt/drive02/clickhouse-backups/IB/

Everything seems to be working fine though, and I get both local and remote backups, both full and incremental.

[Slach]

could you provide more context

which clickhouse-backup version do you use?
how do you execute it? via Docker or command Line?

[ElToro1966]

1. latest
2. command line

[Slach]

2025/02/11 10:40:24.862567 info clickhouse connection closed logger=clickhouse

this log format not from latest (missing filename.go:line)

we don't execute curl inside clickhouse-backup

could you share your backup script?

[ElToro1966]

Yeah, you're right, it's version 2.4.1. As for the curl error; an environment variable in the backup script wasn't specified ($BACKUP_HEALTH_CHECK), thus the error. The backup script:

#!/bin/bash

# full/incremental backup parameter
if [[ $1 == "full" ]]; then
    IS_FULL=true
elif [[ $1 == "incremental" ]]; then
    IS_FULL=false
else
    echo "full/incremental not specified"
    exit 1;
fi

# local/remote backup parameter
if [[ $2 == "local" ]]; then
    CREATE_COMMAND="create"
elif [[ $2 == "remote" ]]; then
    CREATE_COMMAND="create_remote"
else
    echo "local/remote not specified"
    exit 1;
fi

DATETIME=$(date -u +%Y-%m-%dT%H-%M-%S)
BACKUP_NAME_FULL="auto_full_$DATETIME"
BACKUP_NAME_INCREMENTAL="auto_incremental_$DATETIME"

# monitoring API call, f.e. healthchecks.io
BACKUP_HEALTH_CHECK=$BACKUP_HEALTH_CHECK

if [[ $IS_FULL == true ]]; then
    echo "Starting full ($CREATE_COMMAND) backup"
    echo "Creating backup $BACKUP_NAME_FULL"

    clickhouse-backup $CREATE_COMMAND $BACKUP_NAME_FULL
    exit_code=$?

    if [[ $exit_code != 0 ]]; then
        echo "clickhouse-backup create $BACKUP_NAME_FULL FAILED and return $exit_code exit code"
        exit $exit_code
    else
        curl -fsS -m 10 --retry 5 -o /dev/null $BACKUP_HEALTH_CHECK
    fi
else
    echo "Starting incremental ($CREATE_COMMAND) backup"

    # for incremental backup, take previous backup name 
    # based on dates in auto generated backup names.
    BACKUP_NAME_PREV="$(clickhouse-backup list remote | grep -E '^auto_' | tail -n 1 | cut -d " " -f 1)"
    echo "Creating backup $BACKUP_NAME_INCREMENTAL as diff from $BACKUP_NAME_PREV"

    clickhouse-backup $CREATE_COMMAND --diff-from-remote=$BACKUP_NAME_PREV $BACKUP_NAME_INCREMENTAL
    exit_code=$?

    if [[ $exit_code != 0 ]]; then
        echo "clickhouse-backup create $BACKUP_NAME_FULL FAILED and return $exit_code exit code"
        exit $exit_code
    else
        curl -fsS -m 10 --retry 5 -o /dev/null $BACKUP_HEALTH_CHECK
    fi
fi

[Slach]

Glad to know, you figured out