Hello, I am a bit hopeless now and really need to ask for help...
I got Tesla P4 in my Unraid tower, all settings to Machinaris container seem to be OK (I am able to see the card via "nvidia-smi" inside the container), but it is not being used at all - Machinaris tries to decompress (C7) using only CPU.
I run the same absolutely same settings as on my previous server (with the same Tesla P4):
harvester:
chia_ssl_ca:
crt: config/ssl/ca/chia_ca.crt
key: config/ssl/ca/chia_ca.key
decompressor_thread_count: 1
decompressor_timeout: 20
disable_cpu_affinity: false
enforce_gpu_index: false
farmer_peer:
host: 127.0.0.1
port: 8447
gpu_index: 0
logging: *id001
max_compression_level_allowed: 7
network_overrides: *id002
num_threads: 30
parallel_decompressor_count: 1
parallel_read: true
plot_directories:
- /plots1
- /plots2
plots_refresh_parameter:
batch_size: 300
batch_sleep_milliseconds: 1
interval_seconds: 120
retry_invalid_seconds: 1200
port: 8448
private_ssl_ca:
crt: config/ssl/ca/private_ca.crt
key: config/ssl/ca/private_ca.key
recursive_plot_scan: false
rpc_port: 8560
selected_network: mainnet
ssl:
private_crt: config/ssl/harvester/private_harvester.crt
private_key: config/ssl/harvester/private_harvester.key
start_rpc_server: true
use_gpu_harvesting: true
inbound_rate_limit_percent: 100
And in log this repeats:
2023-10-28T18:34:37.200 harvester chia.harvester.harvester: WARNING No decompressor available. Cancelling qualities retrieving for /plots2/plot-k32-c07-2023-10-14-00-57-ef88c41d8f61deda9d44059d58a4085a764392813d497cc24dd4f65f797470f5.plot
EDIT: just tried another container to use GPU - no issue there, Plex was able to use the GPU,
I will be very grateful for advice, thanks in advance!