Commit Graph

3953 Commits

Author SHA1 Message Date
Jakob Borg
192455702d Repair AUTHORS 2015-12-01 11:23:35 +01:00
Jakob Borg
cb0d739daf New key for discovery-*-3 2015-12-01 11:20:16 +01:00
Jakob Borg
a937fcc477 Lol shadowing :( 2015-12-01 11:20:16 +01:00
Audrius Butkevicius
9503e60444 Merge pull request #2544 from calmh/negcache2
Accept Retry-After header on discovery lookup failures
2015-12-01 10:14:32 +00:00
Jakob Borg
9f2dc4554d Accept Retry-After header on discovery lookup failures 2015-12-01 11:10:03 +01:00
Audrius Butkevicius
55da600433 Merge pull request #33 from syncthing/negcache
Set Retry-After header
2015-12-01 09:58:20 +00:00
Jakob Borg
96b5c2ae00 Set Retry-After header 2015-12-01 10:49:16 +01:00
Jakob Borg
3008dc76d3 Add buinsky 2015-11-30 18:30:36 +01:00
Audrius Butkevicius
596c4b77e8 Merge pull request #2539 from buinsky/master
Fix deleting folders on WinXP (fixes #2522)
2015-11-30 16:21:46 +00:00
buinsky
05a31c2686 Fix deleting folders on WinXP (fixes #2522) 2015-11-30 19:09:36 +03:00
Audrius Butkevicius
45535c0f5a Merge pull request #2538 from canton7/feature/issue-2537
Allow #urPreview to scroll in the browser (fixes #2537)
2015-11-30 13:36:28 +00:00
Antony Male
4bd0dd2123 Allow #urPreview to scroll in the browser (fixes #2537)
This is the same issue as #2014/#2062. Bootstrap doesn't like having two dialogs
open at once: it marks the body has having no dialogs open when the first dialog
is closed, regardless of whether the second dialog is still open.

This means that scrolling doesn't happen properly, and the user cannot
scroll to the dialog's 'close' button.

Work around this by making sure the first dialog (the settings page) is fully closed
before the second dialog (usage preview) is opened.
2015-11-30 13:27:07 +00:00
Jakob Borg
9b2a643626 Update docs and translations 2015-11-29 08:55:38 +01:00
Jakob Borg
1ebc9a9a88 Merge pull request #2525 from kluppy/master
Don't chmod in Atomic on android (fixes  #2472)
2015-11-28 22:57:00 +01:00
Jakob Borg
f0c8b7ce40 Add kluppy 2015-11-28 22:17:05 +01:00
Jakob Borg
ec54550f21 Merge pull request #2529 from AudriusButkevicius/fixlinks
Fix symlinks (fixes #2524)
2015-11-28 21:29:33 +01:00
Audrius Butkevicius
4474d200b0 Fix symlinks (fixes #2524) 2015-11-28 20:23:08 +00:00
Audrius Butkevicius
b24a9e57fd Update deps 2015-11-27 21:04:40 +00:00
Audrius Butkevicius
07722dc33d Hey look, had to check all code out on linux to fix the deps 2015-11-27 21:02:19 +00:00
kluppy
f062e35641 Don't chmod in Atomic on android (fixes #2472) 2015-11-28 02:46:06 +10:00
Jakob Borg
321ef9816c Merge pull request #2521 from AudriusButkevicius/dialtimeout
Add dialer.DialTimeout, use that when connecting to relays
2015-11-27 10:01:21 +01:00
Audrius Butkevicius
be01e925c7 Merge pull request #2523 from calmh/ecdsa
Generate ECDSA keys instead of RSA
2015-11-27 08:28:51 +00:00
Jakob Borg
6d11006b54 Generate ECDSA keys instead of RSA
This replaces the current 3072 bit RSA certificates with 384 bit ECDSA
certificates. The advantage is these certificates are smaller and
essentially instantaneous to generate. According to RFC4492 (ECC Cipher
Suites for TLS), Table 1: Comparable Key Sizes, ECC has comparable
strength to 3072 bit RSA at 283 bits - so we exceed that.

There is no compatibility issue with existing Syncthing code - this is
verified by the integration test ("h2" instance has the new
certificate).

There are browsers out there that don't understand ECC certificates yet,
although I think they're dying out. In the meantime, I've retained the
RSA code for the HTTPS certificate, but pulled it down to 2048 bits. I
don't think a higher security level there is motivated, is this matches
current industry standard for HTTPS certificates.
2015-11-27 09:15:12 +01:00
Audrius Butkevicius
ed792b97c0 Take timeout into account when dialing 2015-11-26 23:41:11 +00:00
Audrius Butkevicius
a4b8c2298a Add dialer.DialTimeout support 2015-11-26 23:31:37 +00:00
Jakob Borg
e5b33ce9f6 Regenerate XDR for empty struct types 2015-11-24 20:54:49 +01:00
Jakob Borg
30374d46c9 Merge pull request #2510 from plouj/master
Log problems fetching releases
2015-11-24 20:39:29 +01:00
Michael Ploujnikov
9edf8233f7 Improve upgrade error messages 2015-11-24 08:29:42 -05:00
Michael Ploujnikov
bd4a14519c FetchLatestReleases: just log the error here
Since the return value is being ignored by the caller.
2015-11-24 08:29:42 -05:00
Michael Ploujnikov
f12bf8c09a Rename LatestGithubReleases -> FetchLatestReleases 2015-11-24 08:29:42 -05:00
Audrius Butkevicius
53cc45a0d5 Merge pull request #2512 from calmh/compact
Compact database on startup (ref #2400)
2015-11-24 12:43:38 +00:00
Jakob Borg
fa4b4dece1 Compact database on startup (ref #2400) 2015-11-24 13:17:30 +01:00
Jakob Borg
02f044a2a1 Add plouj 2015-11-24 08:35:25 +01:00
Audrius Butkevicius
f39f816a98 Update godeps, reduce amount of time spent testing a relay. Goddamit godeps. 2015-11-23 21:33:22 +00:00
Audrius Butkevicius
bc5b95be8a Update packages, fix testutil. Goddamit godep. 2015-11-23 21:29:23 +00:00
Audrius Butkevicius
431d51f5c4 Add timeouts to relay methods 2015-11-23 21:14:46 +00:00
Jakob Borg
45c1357bab Update osext dependency (fixes #1272) 2015-11-23 13:09:42 +01:00
Audrius Butkevicius
845f31b98f Add timeouts, deal with overlapping markers, add a table, increase circle radiuses 2015-11-22 22:47:48 +00:00
Jakob Borg
5136675fae Docs & translation update 2015-11-22 16:05:20 +01:00
Audrius Butkevicius
89b6c32cee Merge pull request #6 from canton7/feature/fix-map
Fix a couple of issues with the relays map (geoip, 'data unavailable')
2015-11-22 14:27:58 +00:00
Antony Male
6ee36fe361 Fix a couple of issues with the relays map (geoip, 'data unavailable')
- Move to ipinfo.io for geoip, rather than Telize. Telize has been closed
   down. ipinfo.io has apparently got decent availability, and allows
   1,000 requests per day on the free tier. Since requests are made by the
   client, this should be more than enough (and the total across all clients
   should still be less than this).

 - Fix issue where one nonresponsive relay would cause 'data unavailable'
   to be shown for many relays. This was caused by the relay status
   promise not being correctly added to the list of things being waited
   for before the map was rendered. Any delayed relay status requests
   would therefore occur after the map was rendered, which was too late.
2015-11-22 14:10:29 +00:00
Jakob Borg
db4f23f377 Refactor: extract function generate 2015-11-22 07:35:24 +01:00
AudriusButkevicius
77572d0aee Typo 2015-11-21 18:58:52 +00:00
Audrius Butkevicius
6ff02761a9 Merge pull request #2500 from calmh/sparse
Handle sparse files (fixes #245)
2015-11-21 17:21:11 +00:00
Jakob Borg
d46f267663 Handle sparse files (fixes #245) 2015-11-21 17:58:09 +01:00
Jakob Borg
6a98534d5d Merge pull request #2498 from AudriusButkevicius/relaystuff
Change the way relays are chosen, add RelayFull message.
2015-11-21 16:51:03 +01:00
Audrius Butkevicius
eeb5d99942 Sort relays in 50ms latency increments, shuffle relays within the same increment 2015-11-21 13:23:49 +00:00
Audrius Butkevicius
dcc5f333c1 Merge pull request #2499 from calmh/metrics
Add metrics for HTTP calls
2015-11-21 13:08:26 +00:00
Jakob Borg
ff8a66d22f Add metrics for HTTP calls 2015-11-21 09:48:57 +01:00
Jakob Borg
f7ad97918a HTTP debug middleware should be behind ShouldDebug() 2015-11-21 09:39:40 +01:00