33eccb35da
chore: update wolff -> 3.0.2
2024-07-25 14:30:50 -03:00
854754eb60
feat(coap): use content-sensitive udp proxy for coap
2024-07-12 16:23:46 +08:00
eaaee725c2
fix: upgrade to hocon 0.43.1
...
included 3 changes since 0.42.2
- allow validation of map keys
- improve crash stacktrace report
- avoid dumping array environment variable values
2024-07-03 23:00:14 +02:00
afb65817c4
ci(mix): refactor to avoid conflicts
...
This introduces a refactoring that will hopefully help avoid conflicts with ongoing work
to migrate our build / CI tooling to Mix.
2024-07-03 14:37:25 -03:00
4df2e0be85
fix: bad rebase conflict resolution
2024-07-02 10:21:52 -03:00
8a68f5dada
ci: temporarily revert mix release back to old version
...
This is so we may merge the huge PR as is and work on the fixes later with less conflicts.
2024-07-02 09:47:49 -03:00
19f3b030f9
chore: preparing to run common tests / eunit with mix
2024-07-02 09:45:45 -03:00
b74828d7ea
perf: "mixify" project to improve release speed
...
Fixes https://emqx.atlassian.net/browse/EMQX-12527
2024-07-02 09:39:48 -03:00
8f4b8d2ea2
chore: bump `ehttpc` to 0.4.14
...
Which includes a fix that prevents `emqtt_pool` from reusing pools
in an inconsistent state.
2024-07-01 20:03:13 +02:00
4c3c86e919
chore: bump wolff -> 2.0.0
2024-06-27 09:20:00 -03:00
6023012f8b
fix(kafka and derivatives): add `alias` config to avoid clashes with same topic
...
Fixes https://emqx.atlassian.net/browse/EMQX-12592
2024-06-27 09:20:00 -03:00
abc255bb02
fix(oidc): make CI happy
2024-06-24 18:11:02 +08:00
ac8762be10
chore: upgrade ekka to 0.19.5 (mria 0.8.8)
...
mria 0.8.8 heals a network partition once majority of core nodes are alive.
Previously, the autoheal worked only when all core nodes were reachable.
Fixes: EMQX-10974
2024-06-20 12:53:58 +03:00
6897f0141b
fix(plugins): do not stop Elixir protected apps
...
On one hand, Elixir plugins might include Elixir itself, when targetting a non-Elixir EMQX
release. If, on the other hand, the EMQX release already includes Elixir, we shouldn't
stop Elixir nor IEx, or the running IEx shell will break.
2024-06-17 16:42:23 -03:00
00dfdc22cf
chore: pin Kafka client wolff 0.10.5
2024-06-17 17:27:25 +02:00
a6e3a09118
feat: make the dashboard restart quicker
2024-06-14 18:22:34 +08:00
723144199e
Merge pull request #13211 from qzhuyan/fix/william/rel57/ee-tls
...
Fix/william/rel57/ee tls
2024-06-13 13:19:44 +02:00
83ff07ade0
Merge pull request #13199 from thalesmg/message-transformation-r57-20240604
...
feat: implement message transformation
2024-06-11 13:29:25 -03:00
7c37bf9965
feat(tls): ee only: TLS partial_chain and Keyusage
2024-06-10 10:24:48 +02:00
672173c7fd
chore: pin system_monitor 3.0.5
2024-06-09 10:06:15 +02:00
ef36350bf6
feat: implement message transformation
...
Fixes https://emqx.atlassian.net/browse/EMQX-12517
2024-06-06 17:51:08 -03:00
8d7417d51c
chore: stop pinning supervisor in mix.exs
2024-06-02 10:10:17 +02:00
9637b14543
fix: upgrade to brod 3.18.0 to fix unnecessary rebalance
2024-06-01 10:57:24 +02:00
b14a138bf1
Merge pull request #13154 from id/emqx-12190-fix-apt-purge
...
fix(deb): fix apt purge when emqx service was still running
2024-05-31 07:26:24 +02:00
830926044c
chore: bump erlang-rocksdb to 1.8.0-emqx-5
...
- fix macos build https://github.com/google/snappy/issues/183
- pre-built binaries for latest platforms
2024-05-30 17:13:06 +02:00
a00d9f17b1
chore: upgrade kafka producer lib wolff to 1.10.4
...
wolff-1.10.4 handles message_too_large error code from Kafka.
It tries to split the too-large batch into single-call batches
(for EMQX, one call batch is always one message), and retry.
In case a single-call batch is too large, EMQX should increment
a failure counter.
2024-05-30 15:40:10 +02:00
5c9f1d9401
ci: fix emqx ce elixir build
2024-05-22 11:49:10 -03:00
257dae5211
refactor: rename `message_validation` to `schema_validation`
...
Request from Product team.
Fixes https://emqx.atlassian.net/browse/EMQX-12383
2024-05-14 10:32:27 -03:00
e8d2c48d3c
Merge pull request #13018 from thalesmg/fix-postgres-log-spam-r57-20240510
...
fix(postgres): reduce log spamming when connection goes down
2024-05-14 09:00:36 -03:00
8506ca7919
Merge pull request #12998 from ieQu1/dev/improve-latency
...
Use leader's clock when calculating LTS cutoff timestamp
2024-05-13 21:54:06 +02:00
f690db9849
fix(postgres): reduce log spamming when connection goes down
...
Fixes https://emqx.atlassian.net/browse/EMQX-12334
See also: https://github.com/emqx/epgsql/pull/10
2024-05-10 12:15:01 -03:00
f250f00f3f
chore: Bump snabbkaffe version to 1.0.10
2024-05-08 23:17:57 +02:00
d61f4078e2
refactor: extract blob aggregator logic to separate application
...
This will allow other bridges to share the same aggregation logic as S3 aggregated action.
2024-05-03 14:35:58 -03:00
3386f565aa
build: replace cloud release type with standard and pass it as argument
2024-05-02 10:25:53 +02:00
ad91ca4401
chore: upgrade to rulesql 0.2.1 from 0.2.0
...
a minor enhancement: rule SQL without WHERE or FROM will
return "Missing FROM or WHERE" instead of "syntax error before {SELECT"
2024-04-29 22:01:32 +02:00
154eb18657
chore: pin mimerl 1.2.0
...
for some reason, mix pulls 1.3.0 (the lastest from hex.pm)
2024-04-29 15:45:32 +02:00
10625eacac
chore: upgrade to hocon-0.42.2
...
hocon pretty-print quotes more string values
if a string has '.' or '-', or if it starts with a digit 0-9,
then it's quoted.
see details here: https://github.com/emqx/hocon/pull/293
2024-04-28 14:05:30 +02:00
6049b4966f
fix: bump escokd to 5.11.2 to handle udp_error/closed message
2024-04-22 16:41:09 +08:00
b7a4536f47
Merge pull request #12871 from savonarola/0412-fix-rebalance-app-dependencies
...
fix(rebalance): fix start order of rebalance applications
2024-04-19 13:27:41 +03:00
2a2fadfbff
Merge pull request #12827 from kjellwinblad/kjell/emqx_rule_tracing/EMQX-11966
...
Add rule ID tracing support
2024-04-18 09:06:31 +02:00
cf56050759
feat: avoid mixing request with and without the stop_after_render flag
...
Previously a batch of requests that was sent to a connector could
contain both requests with the stop_after_rendering flag and requests
without this flag. When this happened a warning message was generated and
the stop_after_render flags for the batch would be ignored. This commit
fixes so that a mixed batch is never created so there is no longer any
need for a warning message or ignoring flags.
2024-04-17 16:21:21 +02:00
8fe471c967
fix(mqtt): enhanced auth with scram
...
Bump esasl to 0.2.1
2024-04-16 10:08:30 +02:00
3ef160eed2
Merge remote-tracking branch 'upstream/release-56' into 0415-sync-release-56
2024-04-15 08:09:03 +02:00
500d4fedda
fix(rebalance): fix start order of rebalance applications
2024-04-12 15:58:57 +03:00
55179ccfed
chore: update ekka to 0.19.3
...
Included updates:
- https://github.com/emqx/mria/pull/178
2024-04-09 19:04:38 +03:00
3e87d4bf9f
Merge pull request #12772 from killme2008/feature/upgrade-greptimedb-ingester-ci
...
feat: update greptimedb client lib and ci version
2024-04-08 09:02:03 +02:00
2e528d1dd8
Merge pull request #12802 from SergeTupchiy/EMQX-11826-prevent-left-node-from-rejoining-5.6.1
...
prevent a left node from rejoining the same cluster
2024-03-28 19:49:18 +02:00
3eda182e9a
fix: prevent a node from discovering and re-joining the same cluster after it has (manually) left it.
2024-03-28 18:09:27 +02:00
002266c1db
Merge pull request #12806 from zmstone/0328-upgrade-http-clients
...
chore: upgrade http client libs gun-1.3.11 and ehttpc-0.4.13
2024-03-28 16:06:57 +01:00
7982dd017b
chore: upgrade http client libs gun-1.3.11 and ehttpc-0.4.13
2024-03-28 13:20:59 +01:00
Thales Macedo Garitezi