Hi, I read in the README for zfec that...
"... you also need to do integrity checking (such as with sha256sum) on the blocks that result from the erasure coding in *addition* to doing it on the file contents!" ... in order to help identify corrupted blocks. But then you have to store a list of valid sha256sums. To protect this list, you could encode it with zfec, but... well, that's a recursive descent into madness. Is there a solution here, or is there actually no problem and I'm worrying over nothing? Also, I think I found a bug in cmdline_zfec.py (v1.4.22). If you specify the input file using an absolute pathname, the --output-dir option is ignored. You could fix this by replacing the line: args.prefix = args.inputfile.name with args.prefix = os.path.basename(args.inputfile.name) (plus an import in the appropriate place) Otherwise, zfec is awesome - good work! :) Thanks, James _______________________________________________ tahoe-dev mailing list [email protected] http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
