Worker Configuration

Worker Configuration

Each worker reads a JSON config, worker.json by default. Override with -config /path/to/file.json.

Example

{
  "server_url": "http://hashtasker.example.com:8080",
  "hostname": "worker-01",
  "checkin_interval": "30s",
  "hashcat_path": "/usr/bin/hashcat",
  "rulesets_path": "/opt/rules",
  "wordlists_path": "/mnt/nfs/wordlists",
  "temp_path": "/tmp/hashtasker"
}

Fields

Field	Required	Default	Description
server_url	Yes		Base URL of the HashTasker server, no trailing path
hostname	No	OS hostname	The worker's identity. Must be unique across the fleet; two workers with the same hostname will overwrite each other's state.
checkin_interval	No	30s	How often to report stats and poll for work. Go duration (15s, 30s, 1m). Keep it well under 60s or the server will mark the worker offline.
hashcat_path	Yes		Full path to the hashcat binary
rulesets_path	No	/opt/rules	Local rules directory. Rule files are referenced by file name, so names must match the server's.
wordlists_path	NFS only		Local mount point of the shared wordlist directory. If unset, the worker reports NFS as unavailable, which blocks job creation when the server is in NFS mode.
temp_path	Yes		Scratch space for hash files, downloaded chunks, custom wordlists, and potfiles. Created if missing. Cleaned up when jobs end.

Disk Space

In push mode each worker downloads its slice of the wordlist into temp_path, so size it for roughly largest wordlist / number of workers per concurrent job. Uploaded custom wordlists are always downloaded in full.

Running Multiple Workers on One Host

Give each instance its own hostname, temp_path, and config file, and run each with -config.