Hangs during initialization

[darkwave3137]

bin/main -c 0xf -n 2 -- configs/debug.py 
Trying to acquire a process lock...
NBA is running on ROOT privilege.
Lock acquired!
EAL: Detected 8 lcore(s)
EAL: Probing VFIO support...
EAL: PCI device 0000:03:00.0 on NUMA socket -1
EAL:   probe driver: 8086:10fb net_ixgbe
MAIN: Setting log level to 7.
MAIN: Detected 1 ports.
NUMA is disabled.
MAIN: Loading system configuration from "configs/debug.py"...
MAIN: If it hangs, try to restart hold_gpu script. (cuda service if installed as upstart)
NetDevice(device_id=0, driver='net_ixgbe', busaddr='0000:03:00.0', macaddr='90:E2:BA:4C:C4:D0', numa_node=-1, min_rx_bufsize=1024, max_rx_pktlen=15872, max_rx_queues=128, max_tx_queues=64)
CoprocDevice(device_id=0, driver='cuda', busaddr='0000:04:00.0', numa_node=-1, name='Tesla C2075', support_concurrent_tasks=True, global_memory_size=5588713472)
Cores in NUMA node 0: [0, 1, 2, 3]
# logical cores: 8, # physical cores 4 (hyperthreading enabled)
MAIN: port 0 -- link running at lower than 10G half-duplex, DOWN
MAIN: port 0 is enabled.
MAIN: 1 io threads, 1 comp threads, 1 coproc threads (in total)
COPROC: @7: initialized device cuda
MAIN: spawned coproc threads.

Why did it stop here?

[leeopop]

Please check whether https://github.com/ANLAB-KAIST/NBA/blob/468e9511736678e7b3184baecaf9011d548eb8e5/src/lib/coprocessor.cc#L216 this line is executed in your environment.

[leeopop]

or try this configuration and script to fix cuda issues

https://gist.github.com/achimnol/3404967