andoma

Members
  • Posts

    6
  • Joined

  • Last visited

Posts posted by andoma

  1. btsync keeps crashing on my linux machine

     


    $ uname -a

    Linux skytown 3.15.0-031500rc4-generic #201405042135 SMP Mon May 5 01:36:35 UTC 2014 x86_64 x86_64 x86_64 GNU/Linux

     

    Any additional info I can provide?


     

     

    [1002841.588262] traps: btsync[28365] trap divide error ip:4d18bf sp:7f7be5734dc0 error:0 in btsync[400000+3df000]

    [1002841.770462] init: btsync main process (28359) killed by FPE signal

    [1002841.770478] init: btsync main process ended, respawning

    [1002876.708243] traps: btsync[28450] trap divide error ip:4d18bf sp:7f91b6212dc0 error:0 in btsync[400000+3df000]

    [1002876.890650] init: btsync main process (28444) killed by FPE signal

    [1002876.890672] init: btsync main process ended, respawning

    [1002911.664913] traps: btsync[28536] trap divide error ip:4d18bf sp:7f02367bedc0 error:0 in btsync[400000+3df000]

    [1002911.843780] init: btsync main process (28530) killed by FPE signal

    [1002911.843796] init: btsync main process ended, respawning

    [1002967.037544] traps: btsync[28618] trap divide error ip:4d18bf sp:7fe515397dc0 error:0 in btsync[400000+3df000]

    [1002967.218573] init: btsync main process (28611) killed by FPE signal

    [1002967.218589] init: btsync main process ended, respawning

    [1003012.261656] traps: btsync[28745] trap divide error ip:4d18bf sp:7fb54b2aedc0 error:0 in btsync[400000+3df000]

    [1003012.444918] init: btsync main process (28739) killed by FPE signal

    [1003012.444935] init: btsync main process ended, respawning

    [1003062.164085] traps: btsync[28853] trap divide error ip:4d18bf sp:7f913ca52dc0 error:0 in btsync[400000+3df000]

    [1003062.344367] init: btsync main process (28847) killed by FPE signal

    [1003062.344383] init: btsync main process ended, respawning

    [1003112.232984] traps: btsync[28971] trap divide error ip:4d18bf sp:7f7d0c1f7dc0 error:0 in btsync[400000+3df000]

    [1003112.414854] init: btsync main process (28965) killed by FPE signal

    [1003112.414877] init: btsync main process ended, respawning

    [1004926.800941] traps: btsync[29087] trap divide error ip:4d18bf sp:7f818fe90dc0 error:0 in btsync[400000+3df000]

    [1004926.993142] init: btsync main process (29081) killed by FPE signal

    [1004926.993158] init: btsync main process ended, respawning

    [1004946.570240] traps: btsync[600] trap divide error ip:4d18bf sp:7f52a6d47dc0 error:0 in btsync[400000+3df000]

    [1004946.749272] init: btsync main process (594) killed by FPE signal

    [1004946.749289] init: btsync main process ended, respawning

    [1004966.675878] traps: btsync[651] trap divide error ip:4d18bf sp:7f2a18486dc0 error:0 in btsync[400000+3df000]

    [1004966.855516] init: btsync main process (645) killed by FPE signal

    [1004966.855539] init: btsync main process ended, respawning

    [1004986.772559] traps: btsync[705] trap divide error ip:4d18bf sp:7f00a5d03dc0 error:0 in btsync[400000+3df000]

    [1004986.950406] init: btsync main process (699) killed by FPE signal

    [1004986.950422] init: btsync main process ended, respawning

    [1005031.620291] traps: btsync[759] trap divide error ip:4d18bf sp:7f0d8ef34dc0 error:0 in btsync[400000+3df000]

    [1005031.807072] init: btsync main process (753) killed by FPE signal

    [1005031.807089] init: btsync main process ended, respawning

    [1005071.758530] traps: btsync[871] trap divide error ip:4d18bf sp:7ff69ef68dc0 error:0 in btsync[400000+3df000]

    [1005071.939333] init: btsync main process (865) killed by FPE signal

    [1005071.939349] init: btsync main process ended, respawning

    [1041666.754471] traps: btsync[975] trap divide error ip:4d18bf sp:7f21f6434dc0 error:0 in btsync[400000+3df000]

    [1041666.923950] init: btsync main process (969) killed by FPE signal

    [1041666.923967] init: btsync main process ended, respawning

    [1041691.843616] traps: btsync[18650] trap divide error ip:4d18bf sp:7fcad8ff3dc0 error:0 in btsync[400000+3df000]

    [1041692.019058] init: btsync main process (18644) killed by FPE signal

    [1041692.019075] init: btsync main process ended, respawning

     

  2. It would be really nice if btsync could utilize btrfs generations to quickly refresh its internal database upon startup.

     

    Btrfs, being a copy-on-write filesystem, have built-in features to quickly list changes to the file system since a given point in time (generation)

     

    Example:

     

    root@skytown:/storage/media# btrfs subvolume show  .
    /storage/media
    Name: media
    uuid: 6cde316a-1cad-4f41-a718-30a45292f9b6
    Parent uuid: -
    Creation time: 2014-05-04 07:15:24
    Object ID: 257
    Generation (Gen): 959
    Gen at creation: 6
    Parent: 5
    Top Level: 5
    Flags: -
    Snapshot(s):
     
    We're currently at generation 959, then i write to a file.
     
     root@skytown:/storage/media# echo foo > Music/Other/test
     
    After the transaction have committed (which it always does after 30 seconds, or due to sync() or umount, etc) we can query the filesystem for files that have changed since (a very fast operation compared to scanning the entire hierarchy) :
     
    root@skytown:/storage/media# btrfs subvolume find-new . 959
    inode 65807 file offset 0 len 4 disk start 0 offset 0 gen 1021 flags INLINE Music/Other/test
    transid marker was 1021
     
    and we are now at generation 1021.
     
    I think it would be possible for btsync to utilize the btrfs generations to quickly refresh its internal state rather than having to scan the entire filesystem on startup.
     
    Please consider adding support for this, it would be super awesome.
     
     
     
  3. It seems to me that BitTorrent Sync (1.1.15 on OS X) does not bind the SSDP (used for UPnP/DLNA) port using SO_REUSEADDR.

    This prohibits other UPnP apps on the same host from receiving SSDP packets (ie, those apps does not function at all). Those other apps will fail to bind (Address already in use).

    Most unix like systems allows multiple processes to bind the same UDP multicast port if all of them set SO_REUSEADDR before binding.

    Also see: http://stackoverflow.com/questions/577885/uses-of-so-reuseaddr

    Please fix this and I'll be an even more happy user of this awesome tool.

    /andoma