Nick Craig-Wood
80f53176d9
Rename log functions and factor into own file
2017-02-11 17:54:50 +00:00
Marco Paganini
6611d92e21
Only start bandwidth ticker when necessary.
...
- Only start the token ticker when the timetable entry has more than one
entry.
- This fixes the "Scheduled bandwidth change" log message when no
bwlimit is specified.
- Fixes #987
2017-01-04 19:03:49 -08:00
Marco Paganini
3b0f944e23
Add time-based bandwidth limits.
...
- Change the --bwlimit command line parameter to accept both a limit (as
before) or a full timetable (formatted as "hh:mm,limit
hh:mm,limit...")
- The timetable is checked once a minute by a ticker function. A new
tokenBucket is created every time a bandwidth change is necessary.
- This change is compatible with the SIGUSR2 change to toggle bandwidth
limits.
This resolves #221 .
2017-01-03 21:00:38 +00:00
Nick Craig-Wood
4482e75f38
Fix golint
2016-12-15 21:02:41 +00:00
Nick Craig-Wood
0117aeafbf
mount: this removes the async buffering as it was killing seek performance
2016-12-15 17:08:52 +00:00
Nick Craig-Wood
be4fd51289
fuse: Add bandwidth accounting and buffering
...
This fixes rclone mount ignoring bwlimit and increases buffering which
should speed up transfers greatly.
Fixes #796
Fixes #690
2016-12-04 16:57:47 +00:00
Scott McGillivray
f9df545e3c
add --stats-unit option and improve alignment for --stats output
2016-11-28 12:18:30 +00:00
Nick Craig-Wood
e65059e431
Fix non-windows/non-unix builds for bwlimit/SIGUSR2 feature and add a mutex
...
The race detector complained whenever SIGUSR2 was sent to rclone so
this adds a mutex to prevent concurrent access.
2016-10-30 19:20:16 +00:00
Marco Paganini
cc4f5ba7ba
Add support to toggle bandwidth limits via SIGUSR2.
...
Sending rclone a SIGUSR2 signal will toggle the limiter between off and
the limit set with the --bwlimit command-line option.
2016-10-30 17:46:59 +00:00
Nick Craig-Wood
83ba59749f
Make failed uploads not count as "Transferred" - fixes #708
2016-09-12 18:15:58 +01:00
Nick Craig-Wood
2eaac80c86
b2 with crypt: fix crash when uploading large files - fixes #673
2016-09-05 18:10:01 +01:00
Nick Craig-Wood
037a000cc8
b2: fix stats accounting for upload - fixes #602
2016-08-22 21:19:38 +01:00
Nick Craig-Wood
a20d80565b
Tidy stats output - fixes #541
2016-07-11 13:04:30 +01:00
Nick Craig-Wood
0a43ff9c13
Modify interface for accounting to take a string not an fs.Object
2016-07-02 16:58:50 +01:00
Nick Craig-Wood
f438f1e9ef
Fix stats print
2016-06-18 16:41:46 +01:00
Nick Craig-Wood
67d0375b98
Audit use of log.Print and change to Debug, Log, or ErrorLog as appropriate
2016-06-06 21:23:54 +01:00
Nick Craig-Wood
e2788aa729
Display the transfer stats in more human readable form - fixes #428
2016-06-03 22:49:50 +01:00
Nick Craig-Wood
ed72c678f8
Protect accounting from being closed twice
2015-11-06 15:26:58 +00:00
Nick Craig-Wood
e9c915e6fe
Fix golint warnings
2015-09-22 18:47:16 +01:00
Nick Craig-Wood
e391311512
gofmt
2015-09-17 18:42:39 +01:00
klauspost
788ef76f1c
Show more of the filename and align output.
...
Print more of the file name, and make the output aligned, so it is nicer on frequent updates.
2015-09-17 15:53:45 +02:00
klauspost
0a5870208e
Display individual transfer progress
...
Improve progress printing by displaying individual file progress, as well
as a moving average speed with ETA. Example output:
2015/09/15 16:38:21
Transferred: 183599104 Bytes (4646.49 kByte/s)
Errors: 0
Checks: 1
Transferred: 0
Elapsed time: 38.5s
Transferring:
* 01_06_14.mp3: 33% done. avg: 1280.5, cur: 1288.8 kByte/s. ETA: 1m12s
* 01_12_15.mp3: 33% done. avg: 1002.2, cur: 943.4 kByte/s. ETA: 1m17s
* 01_13_14.mp3: 48% done. avg: 1456.8, cur: 1425.2 kByte/s. ETA: 39s
* 01_19_15.mp3: 28% done. avg: 1226.9, cur: 1114.4 kByte/s. ETA: 1m37s
2015-09-16 19:42:46 +01:00
Nick Craig-Wood
0c6f816a49
Implement --retries flag - fixes #109
2015-08-20 21:07:00 +01:00
Nick Craig-Wood
74994a2ec1
Fix uploading big files which was causing timeouts or panics
...
The symtom was one of these two on upload of files only
* panic: d.nx != 0 in crypto/md5.(*digest).checkSum
* read tcp: i/o timeout
It turned out to be a combination of two upstream bugs
* 5a2187309e
* https://groups.google.com/forum/#!topic/golang-dev/0Nl6k5Sj6UU
This commit contains a work-around for the second problem, I've fixed
the first and had the change accepted upstream.
2015-06-09 17:32:45 +01:00
Nick Craig-Wood
e22b445cff
Implement --size-only flag to sync on size not checksum & modtime - fixes #75
2015-06-06 08:49:01 +01:00
Alex Couper
7af1282375
Add --checksum flag to only discard transfers by MD5SUM - fixes #61
...
Useful for copying between backends where checksum fetching is fast,
ie any of s3, swift, drive or googlecloudstorage.
2015-06-05 19:46:03 +01:00
Nick Craig-Wood
a287e3ced7
Implement --bwlimit to limit data transfer bandwidth
2015-02-27 15:03:47 +00:00
Nick Craig-Wood
754c54b64f
Fix differences print out on check
2013-07-02 08:33:17 +01:00
Nick Craig-Wood
b1f681b912
Rename fs.FsX to fs.X
2013-06-28 08:57:32 +01:00
Nick Craig-Wood
6085dc1b5a
Refactor into sub filesystems
2013-06-27 22:17:51 +01:00