With this change, the behavior is as follows:
- SIGTERM / SIGINT: Exit cleanly (exit code 0)
- SIGHUP: Restart
This is the case both when hitting the monitor process or the inner
process, or if running with NORESTART (but then we won't restart,
obviously, just exit with code exitRestarting).
It enables "pkill -HUP syncthing" to do the right thing to restart all
Syncthings on package upgrade, for example.
Also fixes what I think migh thave been a bug where we did not use the
proxy for usage reports. And removes the BuildEnv field that we don't
need any more.
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.
We're going to need the db.Instance to keep some state, and for that to
work we need the same one passed around everywhere. Hence this moves the
leveldb-specific file opening stuff into the db package and exports the
dbInstance type.
'AlwaysLocalNets' was getting printed, but was getting used
when setting up connections. Now, the nets that should be
considered local are printed and used.
Overwriting configuration files is likely to happen if a
user syncs their home directories across computers. In this
case, the biggest risk is that all nodes will end up with
the same certificate and thus Device ID.
When the model prepares a folder for syncing, it checks to
see if the configuration files this instance is using are
getting synced. If the are getting synced, and they aren't
getting ignored, a warning is emitted. The model is used
so that when a new folder is added dynamically, a warning
is also emitted.
This will not prevent a user from shooting themselves in
the foot, and will not cover all cases (e.g. symlinks).
It should provide _something_ for many users in this
situation to go on, though.
This implements a new debug/trace infrastructure based on a slightly
hacked up logger. Instead of the traditional "if debug { ... }" I've
rewritten the logger to have no-op Debugln and Debugf, unless debugging
has been enabled for a given "facility". The "facility" is just a
string, typically a package name.
This will be slightly slower than before; but not that much as it's
mostly a function call that returns immediately. For the cases where it
matters (the Debugln takes a hex.Dump() of something for example, and
it's not in a very occasional "if err != nil" branch) there is an
l.ShouldDebug(facility) that is fast enough to be used like the old "if
debug".
The point of all this is that we can now toggle debugging for the
various packages on and off at runtime. There's a new method
/rest/system/debug that can be POSTed a set of facilities to enable and
disable debug for, or GET from to get a list of facilities with
descriptions and their current debug status.
Similarly a /rest/system/log?since=... can grab the latest log entries,
up to 250 of them (hardcoded constant in main.go) plus the initial few.
Not implemented in this commit (but planned) is a simple debug GUI
available on /debug that shows the current log in an easily pasteable
format and has checkboxes to enable the various debug facilities.
The debug instructions to a user then becomes "visit this URL, check
these boxes, reproduce your problem, copy and paste the log". The actual
log viewer on the hypothetical /debug URL can poll regularly for new log
entries and this bypass the 250 line limit.
The existing STTRACE=foo variable is still obeyed and just sets the
start state of the system.
Not necessarily the easiest way to fix just this bug, but the root cause
was using the (at that point uninitialized) cfg variable, so it seemed
sensible to just get rid of it to avoid that kind of crap.
The connections service no longer depends directly on the
syncthing model object, but on an interface instead. This
makes it drastically easier to write clients that handle
the model differently, but still want to benefit from
existing and future connections changes in the core.
This was motivated by burkemw3's interest in creating a
FUSE client that can present a view of the global model,
but not have all of the file data locally.
The actual decoupling was done by adding a connections.Model
interface. This interface is effectively an extension of the
protocol.Model interface that also handles connections
alongside the modified service.
This makes it so we can initialize the relay management and then give
that to the connection management, instead of the other way around.
This is important to me in the discovery revamp I'm doing, as otherwise
I get a circular dependency when constructing stuff, with relaying
depending on connection, connection depending on discovery, and
discovery depending on relaying.
With this fixed, discovery will depend on relaying, and connection will
depend on both discovery and relaying.
1. Change listen addresses to URIs
2. Break out connectionSvc to support listeners and dialers based on schema
3. Add relay announcement and lookups part of discovery service
I figured we're missing out on being cool and awesome by not having an
alphabetically based release code name like the big guys. This commit
fixes that. I've unilaterally decided on a theme of "$metal $bug"
because metals are kind of cool, and bugs, well, ...
This will decrease the risk of running out of file descriptors for the
database and other bad things, which could otherwise potentially happen
if we're serving lots of requests and scanning in parallel, etc.
Windows doesn't have a per process open file limit like Unix so we don't
need to worry about it there.
The API never got the first few events ("Starting" etc) as it subscribed
too late. Instead, set up a subscription for it early on. If the API is
configured not to run this is unnecessary but doesn't hurt very much.
This reverts the GC behavior to the Go default of triggering GC when the
heap has grown 100% compared to after the previous GC. We were setting
this to 25% to keep memory usage at a minimum, but it has a pretty
severe performance cost (especially when syncing large files) as we keep
triggering GC too often.
This documents the tweak in the `-help` message so users can decide for
themselves, and sticks to whatever the Go runtime developers thinks is
best for the default.
Make sure we have a good random seed on the default RNG, that the
predictable RNG is clearly marked as such, that random strings are
actually the length requested, and that they contain a restricted set of
characters only.
The reason for ShowWindow opose to your FreeConsole is because if you start up
cmd.exe and do syncthing.exe -no-output it actually hides the existing cmd.exe
window oppose to opening a separate window and then hiding it, which keeps the
existing console hanging on syncthing.exe running.
I tried playing around with compiling as GUI, then given the option is not present
allocating a console, and redirecting the std streams to the new console, but that
seems ugly as I'd have to make quite a few calls. But that does get of the initial
flash.
This commit addresses most of the issues identified in #432:
* Support UPnP IGDs with both WANIPConnection and WANPPPConnection services
IGDs that offer both WANIPConnection and WANPPPConnection services should
now have port forwarding correctly configured for all services.
* Support multiple UPnP WANDevice and WANConnection descriptions
Per Figure 1 of the InternetGatewayDevice specification
(http://upnp.org/specs/gw/UPnP-gw-InternetGatewayDevice-v1-Device.pdf),
an IGD may have multiple WAN devices, each with multiple WANConnection
services
* Support for discovery of UPnP InternetGatewayDevice version 2 devices
* Support for discovery of multiple UPnP IGDs
Consumers that cannot yet properly process multiple IGDs can simply use
the first IGD listed in the discovery results
* Logging refinements such as friendly UPnP IGD identifiers in log messages.
* master: (21 commits)
Mechanism to stop external announcement routine
Update goleveldb
Perfstats are not supported on Windows
Build should fail if a platform does not build
Include perfstats and heap profiles in standard build
Actually no, lets not do uploads at all from the build script.
./build.sh upload build server artifacts
Sign checksums, not files.
Badges, add build server
Remove Solaris build again, for now
Travis should build with 1.3 + tip
Translation update
Indicate aproximativeness of repo sizes...
Slightly more conservative guess on file size
Fix set tests
Small goleveldb hack to reduce allocations somewhat
Don't load block lists from db unless necessary
Rip out the Suppressor (maybe to be reintroduced)
Reduce allocations while hash scanning
Add heap profiling support
...
Conflicts:
discover/discover.go
* v0.8:
Don't leak writer and index goroutines on close
Clean up protocol locking and closing
Send initial index in batches
Always send initial index, even if empty (ref #344)
Simplify locking in protocol.Index
Protocol state machine on receiving side
Log client version on connect
Handle query parameters in UPnP control URL (fixes#211)
Avoid deadlock during initial scan (fixes#389)
Add temporary debug logging for #344 (revert later)
Tone down UPnP not found message (fixes#406)
* v0.8:
Increase deadlock timeout, make configurable (fixes#389, fixes#393)
Remove spurious debug output in .stignore handling
Connection notices are informational
No need to hold a write lock in Override
Don't whine about unexpected EOFs
Ensure correct version string format
Conflicts:
model/model.go
New node ID:s contain four Luhn check digits and are grouped
differently. Code uses NodeID type instead of string, so it's formatted
homogenously everywhere.
Jakob Borg
Matt Burke
Audrius Butkevicius
Matt Burke
Lode Hoste
Karol Różycki
kozec
Caleb Callaway
Felix Ableitner
Daniel Martí
Tully Robinson