09-23-2013, 07:08 AM
I'm working on setting up a small VCL cluster. I have three machines: two compute nodes and one broker. I am using VCL 1.22, oclHashcat-plus 0.15, and Catalyst 13.6. I get everything set up, run oclHashcat and get the following:
Per cl.h, -6 is CL_OUT_OF_HOST_MEMORY. I originally had 8GB in the broker. Upped it to 36GB and still got the same error. I believe that I have a configuration problem, rather than running out of RAM.
I set up the broker and compute nodes per the hashcat vcl wiki article. However, I have made two deviations from the instructions:
1 - The compute nodes are CentOS 6, rather than Ubuntu. I am repurposing existing machines. I've run OpenCL programs on them before and they work (and they don't have APP installed). The broker is Ubuntu 12.04.
2 - I'm running on a small, closed network, but there are other machines present (core servers and a dozen web VMs).
Going through other VCL posts, in this one hashcat.net/forum/thread-1845.html it is mentioned that misconfigured networks can cause the broker to run out of memory. My compute nodes are sharing the network with other non-VCL servers. Could this be an/the issue? I don't know enough about how the VCL broker communicates with nodes.
Plus, there is the "/usr/lib/vcl/libOpenCL.so.1: no version information" line from the output. Is that to be expected?
Any feedback and advice is kindly appreciated.
Code:
.oclHashcat-plus64.bin: /usr/lib/vcl/libOpenCL.so.1: no version information available (required by ./oclHashcat-plus64.bin)
oclHashcat-plus v0.15 by atom starting...
Hashes: 1 total, 1 unique salts, 1 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Workload: 256 loops, 80 accel
Watchdog: Temperature abort trigger disabled
Watchdog: Temperature retain trigger disabled
Device #1: Tahiti, 3035MB, 925Mhz, 32MCU
Device #2: Tahiti, 3035MB, 960Mhz, 28MCU
ERROR: clCreateContext() -6Per cl.h, -6 is CL_OUT_OF_HOST_MEMORY. I originally had 8GB in the broker. Upped it to 36GB and still got the same error. I believe that I have a configuration problem, rather than running out of RAM.
I set up the broker and compute nodes per the hashcat vcl wiki article. However, I have made two deviations from the instructions:
1 - The compute nodes are CentOS 6, rather than Ubuntu. I am repurposing existing machines. I've run OpenCL programs on them before and they work (and they don't have APP installed). The broker is Ubuntu 12.04.
2 - I'm running on a small, closed network, but there are other machines present (core servers and a dozen web VMs).
Going through other VCL posts, in this one hashcat.net/forum/thread-1845.html it is mentioned that misconfigured networks can cause the broker to run out of memory. My compute nodes are sharing the network with other non-VCL servers. Could this be an/the issue? I don't know enough about how the VCL broker communicates with nodes.
Plus, there is the "/usr/lib/vcl/libOpenCL.so.1: no version information" line from the output. Is that to be expected?
Any feedback and advice is kindly appreciated.
