| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| |
| |
| |
| | |
Simplifies configuration, only one already existing boolean to flip.
|
| |
| |
| |
| |
| | |
Complements the DANE support for outgoing connections included in
net.connect
|
| |
| |
| |
| |
| | |
This matches what mod_c2s does, and fixes a traceback in mod_sasl2_fast when
used with BOSH (that module tries to use event.stream.from).
|
| |
| |
| |
| | |
(thanks tmolitor)
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This channel binding method is now enabled when a hash is manually set in the
config, or it attempts to discover the hash automatically if the value is the
special string "auto".
A related change to mod_c2s prevents complicated certificate lookups in the
client connection hot path - this work now happens only when this channel
binding method is used. I'm not aware of anything else that uses ssl_cfg (vs
ssl_ctx).
Rationale for disabling by default:
- Minor performance impact in automatic cert detection
- This method is weak against a leaked/stolen private key (other methods such
as 'tls-exporter' would not be compromised in such a case)
Rationale for keeping the implementation:
- For some deployments, this may be the only method available (e.g. due to
TLS offloading in another process/server).
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
Some of the new roles don't quite fit nicely into 4 characters
(excluding ellipsis). Given the ability to dynamically add additional
roles from the config and possibly from modules, it seems better to just
make it a relative size since we can't know how long they will be.
|
| |
| |
| |
| | |
Explicit > implicit
|
| | |
|
| |
| |
| |
| | |
Requested feature for many modules, notably MAM and file sharing.
|
| |
| |
| |
| |
| |
| | |
MattJ contributed new APIs for retrieving the actually used certificate
and chain to LuaSec, which are not in a release at the time of this
commit.
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This originally used a WIP implementation of cert:sigalg(), a method to
retrieve certificate signature algorithm, but it was never submitted
upstream. https://github.com/Zash/luasec/tree/zash/sigalg
cert:getsignaturename() was merged in
https://github.com/brunoos/luasec/commit/de393417b7c7566caf1e0a0ad54132942ac4f049
XEP-0440 v0.3.0 made implementing tls-server-end-point a MUST
|
| |
| |
| |
| |
| |
| |
| |
| | |
Since this channel binding method is said to enable TLS offloading then
you need tell Prosody the hash (or the full cert), so this seems like a
good start.
Support is RECOMMENDED in XEP-0440 version 0.2
|
| | |
|
| |
| |
| |
| | |
Ensures the periodic cleanup really does remove expired tokens.
|
| |
| |
| |
| | |
This should ensure expired grants eventually disappear.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Generally it is expected that a grant would have at least one token as
long as the grant is in active use.
Refresh tokens issued by mod_http_oauth2 have a lifetime of one week by
default, so the idea here is that if that refresh token expired and
another week goes by without the grant being used, then the whole grant
can be removed.
|
| | |
|
| | |
|
| |
| |
| |
| | |
Use of datetime was removed in 6ac5ad578565
|
|\| |
|
| |
| |
| |
| |
| |
| |
| |
| | |
Suggested by jstein in the chat
This option label is used by XMPP clients to explain what the option does.
a) The user should know where the data is archived.
b) The user needs a statement that can be enabled/disabled by the variable. A question would have the wrong logic here.
|
| |
| |
| |
| | |
Does this run in a thread?
|
| |
| |
| |
| |
| | |
This ensures that all interactions with storage happen inside an async
thread, allowing async waiting to be performed in storage drivers.
|
| |
| |
| |
| |
| |
| |
| | |
Maybe it is better to run daily and weekly tasks 'now' on the theory
that people set these things up during times that are appropriate for
maintenance already, so the same time next day or next week might be
fine for periodic cleanup.
|
| |
| |
| |
| | |
Fixes "Could not delete messages for room 'x': (nil)"
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This method would previously never delete the first (and only) item
since it works out which item should become the first item after the
trim operation, which doesn't make sense when all should be removed.
This also works as an optimization for when all the last item should be
trimmed, thus items should be removed.
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Behavior change: It becomes up to the authorization module whether to
allow requests. The default, mod_authz_internal, will allow users on the
*parent* host only, breaking use by some components.
Remaining question is whether to deprecate the `http_file_share_access`
setting or leave as a way to complement/bypass access control?
|
| |
| |
| |
| |
| |
| | |
Thought it was a case mismatch at first, fixed that, but it changed
nothing because the error was in the leaf part of the errors, not the
chain part.
|
| |
| |
| |
| |
| |
| | |
This removes the need to configure e.g. http_external_url or similar
settings in order to get correct URLs out of prosodyctl, as the API
depends on portmanager to know the actual ports that are used.
|
| |
| |
| |
| |
| |
| | |
This appears to have been a copy-paste of the grant revocation function,
or maybe the other way around. Either way, it deleted the whole grant
instead of the individual token as might be expected.
|
| |
| |
| |
| |
| | |
Using util.dependencies appeared to cause problems with running tests in
Busted, so this also removes that and uses pcall directly.
|
| |
| |
| |
| |
| | |
Probably a workaround for the lack of argument passing when using xpcall
in Lua 5.1, no longer relevant.
|
| |
| |
| |
| |
| |
| | |
This communicates the accepted values in case the config diverges from
them. Note that older documentation used an "admin" value behaving like
an alias to true, but this is no longer handled. Should it?
|
| |
| |
| |
| |
| | |
Passing something from module:get_option() to ipairs() suggests that the
option is a list of some sort.
|
| | |
|
| |
| |
| |
| | |
Because it makes sense and improves feedback via logging
|
| |
| |
| |
| | |
Missed this one in previous sweep
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
| |
| |
| | |
Missed this one, was probably only looking for get_option_number
|
| | |
|
| |
| |
| |
| | |
Since it doesn't actually do strict typing :)
|
| | |
|
| |
| |
| |
| |
| |
| |
| | |
Many of these fall into a few categories:
- util.cache size, must be >= 1
- byte or item counts that logically can't be negative
- port numbers that should be in 1..0xffff
|
| |
| |
| |
| | |
Improves readability ("1 day" vs 86400) and centralizes validation.
|
| | |
|