Skip to content

TurboGit/hubicfuse

Repository files navigation

HubicFuse

A FUSE application which provides access to hubiC's cloud files via a mount-point.

This version contains support for DLO, symlinks and support to see other tenant's containers.
Those features are coming from https://github.com/LabAdvComp/cloudfuse

BUILDING

You'll need libcurl, fuse, libssl, and libxml2 (and probably their development packages) installed to build it.

For CentOS and other RedHat-based systems:

yum install gcc make fuse-devel curl-devel libxml2-devel \
    openssl-devel json-c-devel file-devel

PKG_CONFIG_PATH=/lib64/pkgconfig ./configure

For Debian GNU/Linux and Ubuntu:

sudo apt install gcc make curl libfuse-dev pkg-config \
    libcurl4-openssl-dev libxml2-dev libssl-dev libjson-c-dev libmagic-dev

./configure

Then just compile and install:

make
sudo make install

USAGE

Your hubiC Cloud configuration can be placed in a file named $HOME/.hubicfuse. All the following variables are required:

client_id=[hubiC client id for the registered application]
client_secret=[hubiC client secret for the registered application]
refresh_token=[A refresh token you got from the script]

Optional variables:

get_extended_metadata=[true/false, force download of additional file metadata like atime and mtime on first directory list]
curl_verbose=[true/false, enable verbose output for curl HTTP requests]
curl_progress_state=[true/false, enable verbose progress output for curl HTTP requests. Used for debugging.]
cache_statfs_timeout=[value in seconds, large timeout increases the file access speed]
debug_level=[0 default, 1 extremely verbose for debugging purposes]
enable_chmod=[true/false, false by default, still experimental feature]
enable_chown=[true/false, false by default, still experimental feature]

client_id and client_secret can be retrieved from the hubiC web interface

The refresh_token can be obtained running the script provided (hubic_token) or with any other method you like if you follow the example at https://api.hubic.com/

Then you can call hubicfuse:

sudo hubicfuse /mnt/hubic -o noauto_cache,sync_read,allow_other

And finaly, it can be set in /etc/fstab:

hubicfuse /mnt/hubic fuse user,noauto 0 0

It also inherits a number of command-line arguments and mount options from the Fuse framework. The "-h" argument should provide a summary.

It is also possible to pass a custom hubicfuse settings file so that it is possible to mount multiple hubiC accounts:

sudo hubicfuse /mnt/hubic1 -o noauto_cache,sync_read,allow_other,settings_filename=/root/hubic/account1.settings
sudo hubicfuse /mnt/hubic2 -o noauto_cache,sync_read,allow_other,settings_filename=/root/hubic/account2.settings

And finaly, in /etc/fstab :

hubicfuse /mnt/hubic1 fuse user,noauto,settings_filename=/root/hubic/account1.settings 0 0
hubicfuse /mnt/hubic2 fuse user,noauto,settings_filename=/root/hubic/account2.settings 0 0

USAGE AS NON-ROOT

Add the user into the fuse group:

sudo usermod -a -G fuse [username]

Mount using the above command without the sudo. The .hubicfuse file is searched in the user's home.

To unmount use:

fusermount -u [chemin]

USAGE WITH RSYNC

hubiC protocol has no support for renaming. So be sure to use the --inplace option to avoid a second upload for every uploaded object.

BUGS AND SHORTCOMINGS

* A segment size is limited to 5Gb (this is not hubicfuse limit, but hubiC implementation).
  So segment_above should never exceed 5Gb.
* rename() doesn't work on directories (and probably never will).
* When reading and writing files, it buffers them in a local temp file.
* It keeps an in-memory cache of the directory structure, so it may not be
  usable for large file systems.  Also, files added by other applications
  will not show up until the cache expires.
* The root directory can only contain directories, as these are mapped to
  containers in cloudfiles.
* Directory entries are created as empty files with the content-type
  "application/directory".
* Cloud Files limits container and object listings to 10,000 items.
  cloudfuse won't list more than that many files in a single directory.
* File copy progress when uploading does not work, progress is shown when
  file is copied in local cache, then upload operation happens at 100%

RECENT ADDITIONS AND FIXES

* Support for atime, mtime, chmod, chown.
* Large files (segmented) have correct size listed (was 0 before).
* Multiple speed improvements, minimised the number of HTTP calls and added more caching features.
* Fixed many segmentation faults.
* Cached files are deleted on cache expiration when using a custom temp folder.
* Files copied have attributes preserved.
* Working well with rsync due to mtime support and proper copy operations.
* Debugging support for http progress (track upload / download speed etc.)
* Reduced traffic, skips file uploads to cloud if content does not change (using md5sum compare)
* Major code refactoring, code documentation, extensive debugging, additional config options
* Support for custom hubicfuse settings file in order to mount multiple accounts

AWESOME CONTRIBUTORS

* Pascal Obry                        https://github.com/TurboGit
* Tim Dysinger                       https://github.com/dysinger
* Chris Wedgwood                     https://github.com/cwedgwood
* Nick Craig-Wood                    https://github.com/ncw
* Dillon Amburgey                    https://github.com/dillona
* Manfred Touron                     https://github.com/moul
* David Brownlee                     https://github.com/abs0
* Mike Lundy                         https://github.com/novas0x2a
* justinb                            https://github.com/justinsb
* Matt Greenway                      https://github.com/LabAdvComp
* Dan Cristian                       https://github.com/dan-cristian
* Nicolas Cailleaux                  https://github.com/nikokio

Thanks, and I hope you find it useful.

Pascal Obry [email protected]