gnunet-download (1)

gnunet-download [OPTIONS] -- GNUNET_URI

Download files from GNUnet.

-a LEVEL, --anonymity=LEVEL

set desired level of receiver anonymity. Default is 1.

-c FILENAME, --config=FILENAME

use config file (defaults: ~/.config/gnunet.conf)

-D, --delete-incomplete

causes gnunet-download to delete incomplete downloads when aborted with CTRL-C. Note that complete files that are part of an incomplete recursive download will not be deleted even with this option. Without this option, terminating gnunet-download with a signal will cause incomplete downloads to stay on disk. If gnunet-download runs to (normal) completion finishing the download, this option has no effect.

-h, --help

print help page

-L LOGLEVEL, --loglevel=LOGLEVEL

Change the loglevel. Possible values for LOGLEVEL are ERROR, WARNING, INFO and DEBUG.

-n, --no-network

Only search locally, do not forward requests to other peers.

-o FILENAME, --output=FILENAME

write the file to FILENAME. Hint: when recursively downloading a directory, append a '/' to the end of the FILENAME to create a directory of that name. If no FILENAME is specified, gnunet-download constructs a temporary ID from the URI of the file. The final filename is constructed based on meta-data extracted using libextractor (if available).

-p DOWNLOADS, --parallelism=DOWNLOADS

set the maximum number of parallel downloads that is allowed. More parallel downloads can, to some extent, improve the overall time to download content. However, parallel downloads also take more memory (see also option -r which can be used to limit memory utilization) and more sockets. This option is used to limit the number of files that are downloaded in parallel (-r can be used to limit the number of blocks that are concurrently requested). As a result, the value only matters for recursive downloads. The default value is 32.

-r REQUESTS, --request-parallelism=REQUESTS

set the maximum number of parallel requests that is allowed. If multiple files are downloaded, gnunet-download will not run them in parallel if this would cause the number of pending requests to possibly exceed the given value. This is useful since, for example, downloading dozens of multi-gigabyte files in parallel could exhaust memory resources and would hardly improve performance. Note that the limit only applies to this specific process and that other download activities by other processes are not included in this limit. Consider raising this limit for large recursive downloads with many large files if memory and network bandwidth are not fully utilized and if the parallelism limit (-p option) is not reached. This option also only matters for recursive downloads. The default value is 4092.

-R, --recursive

download directories recursively (and in parallel); note that the URI must belong to a GNUnet directory and that the filename given must end with a '/' -- otherwise, only the file corresponding to the URI will be downloaded. Note that in addition to using '-R', you must also specify a filename ending in '.gnd' so that the code realizes that the top-level file is a directory (since we have no meta data).

-v, --version

print the version number

-V, --verbose

print progress information

The GNUNET_URI is typically obtained from gnunet-search. gnunet-fs-gtk can also be used instead of gnunet-download. If you ever have to abort a download, you can at any time continue it by re-issuing gnunet-download with the same filename. In that case GNUnet will not download blocks again that are already present. GNUnet's file-encoding will ensure file integrity, even if the existing file was not downloaded from GNUnet in the first place. Temporary information will be appended to the target file until the download is completed.

The -a option can be used to specify additional anonymity constraints. If set to 0, GNUnet will try to download the file as fast as possible, including using non-anonymous methods. If you set it to 1 (default), you use the standard anonymous routing algorithm (which does not explicitly leak your identity). However, a powerful adversary may still be able to perform traffic analysis (statistics) to over time infer data about your identity. You can gain better privacy by specifying a higher level of anonymity, which increases the amount of cover traffic your own traffic will get, at the expense of performance. Note that your download performance is not only determined by your own anonymity level, but also by the anonymity level of the peers publishing the file. So even if you download with anonymity level 0, the peers publishing the data might be sharing with a higher anonymity level, which in this case will determine performance. Also, peers that cache content in the network always use anonymity level 1.

This option can be used to limit requests further than that. In particular, you can require GNUnet to receive certain amounts of traffic from other peers before sending your queries. This way, you can gain very high levels of anonymity - at the expense of much more traffic and much higher latency. So set it only if you really believe you need it.

The definition of ANONYMITY-RECEIVE is the following. 0 means no anonymity is required. Otherwise a value of 'v' means that 1 out of v bytes of "anonymous" traffic can be from the local user, leaving 'v-1' bytes of cover traffic per byte on the wire. Thus, if GNUnet routes n bytes of messages from foreign peers (using anonymous routing), it may originate n/(v-1) bytes of queries in the same time-period. The time-period is twice the average delay that GNUnet defers forwarded queries.

The default is 1 and this should be fine for most users. Also notice that if you choose very large values, you may end up having no throughput at all, especially if many of your fellow GNUnet-peers all do the same.

~/.config/gnunet.conf

GNUnet configuration file

Report bugs to <https://gnunet.org/bugs/> or by sending electronic mail to <[email protected]>

gnunet-fs-gtk(1), gnunet-publish(1), gnunet-search(1), gnunet.conf(5), gnunet-service-fs(1)