Automatic Coding

Members
  • Posts

    218
  • Joined

  • Last visited

  • Days Won

    8

Posts posted by Automatic Coding

  1. Hi, thanks for making public this very excellent idea!

    I am a linux user and have just taken it for a test drive.

    A few ideas i'd like to suggest.

    1. A Client ID randomly generated by the client at time of installation, this is so that if a sync request comes in, we can choose to accept or decline the request to sync that device if desired. As this gains popularity I can certianly see people randomly generating keys, and seeing if they get to sync from anyone :-p. By default i've noticed all genrated keys are uppercase alphanum, it is a reasonable keyspace... approx 36^32 ... but I have a sneakiing suspicion one could be found. hopefully people will use strong custom keys.

    But the key could equaly well get passed to a 3rd party and i'd like to be able to have a list of ID's i'll allow to connect.

    2. Username / Password settings within the webgui for logging into it... glad to see it exists as a config file option though.

    I haven't read the full thread here, my apologies if this has already been suggested :)

    Also I understand is it alpha, i really like what I see so far, keep up the good work!

    I don't think number 2 is a valid item, it just leaves security holes. If you need to change it, you change it via the configuration, not via the one object that is probably compromised which is why you're changing the password in the first place.

  2. Another request:- Allow other devices to check the status of your device, it's kind of annoying having to SSH into my public firewall device and then hopping to my private device to open up a SSH tunnel to check the sync status of devices that I'm not syncing to currently.

  3. Is there any chance we can restore to old files? E.G. Keep a history of all versions of all files, then, say we screw something up on one of our files, we can restore back to any date in time?

  4. Quicky response:

    The arrows are a LAN connection, the cloud is an internet connection.

    In your 1 file, 9 machines, your file would get blocked out and seeded to each peer like you would think Bittorrent works. The peers would share the blocks with each other that they don't have, speeding the entire thing up.

    I'll let someone else tackle your other ones(or explain mine better!), I don't have my web ui handy to check.

    -Urist

    Thanks for the first two, the images were really confusing me, I took a stab at what they meant, but, then the name "SuperNode" confused me once again. I stroked(?) them out due to them being answered.

  5. This is an interesting idea, although, it could just be done through two sync'd folders. I'd like this option, although, not really prioritized over anything else. Also, let me just confirm?

    Computer A & B recursively have:-

    \Share\

    However, computer C only has access (recursively) to:-

    \Share\Apple\

    So, if A or B update anything other than \Share\Apple -R, then, nothing happens to C, however, it is synced to A/B, however, if computer A/B or C sync to:-

    \Share\Apple

    Then all three get updated?

  6. Ability to upload (Images & Downloaded items) & download (Although, for iOS I believe you're limited to your own sandboxie'd area for downloads)

    Edit items (That you've downloaded)

    API (Although, this isn't application specific)

    Move (This also covers renaming)

    Copy (Hopefully just notify other computers "Copied X to Y" so that you don't have to re-upload the whole file?)

    Ability to read large files by specifying the lines. E.G. something like:- cat largeFile | head -n 10

    Ability to AUTOMATICALLY upload images whenever the application is opened/allowed to upload (Mainly an issue for iOS)

  7. First of, other than the sometimes non-responsive webUI (Which I believe is my fault due to mixing cookies up) and the non-replying webUI (Still no idea why, but, SSH tunneling seems to work), I love this program. It's so nice to have a dropbox-like program that use my servers, without 16GB data caps & the ability to invite unlimited computers in (Vs something like rsync which would require me to propagate files throughout computers using long bash scripts & cron).

    However, I have a few questions about this software. For starters, what does "Direct.png" mean? This image:-

    dF3UReJ.png

    Second, what does "Supernode.png" mean? This image:-

    RkXjAmB.png

    Third, I believe this software is based off the BitTorrent protocol (Why else would it beon bittorrent.com and called bittorrentsync?), so, my question is, if I have 10 people all sharing the same file, one person updates it, everyone else has to download it. Would it send 1/9th (10 people in group - the 1 (one) who updated it and already has it) to each person, then they share it between them for improved speeds or would it send one WHOLE file to the first person, then one WHOLE file to the second person, etc...?

    Sorry if the last one isn't very well explained, I can explain it a bit better if you don't understand it.

    EDIT:- Extra question:-

    What does this mean? It's a photo taken from my laptop's point-of-view:-

    SX24Gu7.png

    What I believe it means is:-

    LAN (Direct.png) transfer from Tower US (Text) has still got to upload (Up arrow) 2.8MB

    WAN (Supernode.png) transfer from Tower UK (Text) has till got to download (Down arrow) 43.8MB

    Am I correct? If not, what does it mean?

    EDIT:- Another question:-

    Does bittorrent sync do any kind of compression before submitting over WAN? I understand that compression over LAN MIGHT in some cases slow it down, but, I can't really think of that over WAN.

    EDIT:- Another question, say one of my computers in said share isn't port forwarded (Say one of the members are away from the household, in a hotel or a starbucks), can bitTorrentShare continue to work? By asking one of my servers/other computers to open a connection with, then ask for the data in reverse? Rather than have another computer talk to it?

  8. I think the secret here lies in the configuration file, you have something like this:

    "webui" :

    {

    "listen" : "0.0.0.0:8888",

    0.0.0.0 would indicate it's listening on all of the addresses present, and something like 127.0.0.1:8888 would be localhost only, etc. If you're using the default configuration, it should be 0.0.0.0, but I would check anyway.

    It is listening on every IP (0.0.0.0), however, once it accepts the connection, it issued a HTML page that declines your request (Invalid request).

  9. I have a similar problem. If I go to http://localhost:8888/gui/ I get "invalid request" but if I go to http://127.0.0.1:8888/gui/ then it works. It might be useful to know if either localhost:8888 or 127.0.0.1:8888 work for you. Mine works when I got to my DHCP IP address too. Just not "localhost".

    Well, if I attempt to connect to localhost on my laptop (Not the headless server), I get access, same with 127.0.0.1:8888 and 192.168.88.250 (Laptop's LAN IP).

  10. Please no! Not another service/application depending on Google knowing what we do and when with whom and where. I LOVE the simplicity of the shared secret. (maybe certificate based sharing is a nice upgrade).

    By the way, is there a way to prevent Syncapp to even connect to the bittorrent servers?

    /paranoia mode ;-)

    I don't think it'd be required, it'd be optional. Nobody in their right mind would say "You HAVE to use an external authentication system to use our software, even if the files you are sharing aren't really private/etc"

  11. Because I'd like to run this on a headless linux server, I'm obviously not going to be connecting from localhost. When I connect from a LAN interface on it's LAN IP I get "Invalid request" as shown below:-

    VMzue8N.png

    However, if I create an SSH tunnel so I'm connecting from localhost (In this case, 0.0.0.0:8889 --> localhost:8888) it works:-

    GGSpk1S.png

    Is there anyway to get BitTorrent Sync to accept LAN requests? Keeping an SSH tunnel open isn't really the most easy way of accessing it.