9dc3a169b3
feat: split bridges into a connector part and a bridge part
...
Co-authored-by: Thales Macedo Garitezi <thalesmg@gmail.com>
Co-authored-by: Stefan Strigler <stefan.strigler@emqx.io>
Co-authored-by: Zaiming (Stone) Shi <zmstone@gmail.com>
Several bridges should be able to share a connector pool defined by a
single connector. The connectors should be possible to enable and
disable similar to how one can disable and enable bridges. There should
also be an API for checking the status of a connector and for
add/edit/delete connectors similar to the current bridge API.
Issues:
https://emqx.atlassian.net/browse/EMQX-10805
2023-10-30 14:48:47 +01:00
14983ec14a
chore(hooks): validate hookpoints and document hook callbacks
...
Co-authored-by: Thales Macedo Garitezi <thalesmg@gmail.com>
2023-09-26 14:27:42 +03:00
7c2f87fabe
test: merge broker and router boot modules
2023-09-06 21:36:16 +02:00
18c6bfec97
chore: bump app vsns
2023-09-06 11:12:56 +02:00
e5041de9cc
fix(gcp_consumer): handle 401 errors
...
Fixes https://emqx.atlassian.net/browse/EMQX-10852
2023-08-24 09:03:34 -03:00
84dd3b83bb
fix: hocon converter not working when make_serializable is true(check_plain)
2023-08-24 06:46:47 +08:00
74d4a8325e
Merge pull request #11470 from thalesmg/gcp-consumer-rm-worker-pool-size-r52-20230817
...
fix(gcp_consumer): remove unused resource_opts fields
2023-08-21 14:49:59 -03:00
bd5680db00
Merge pull request #11461 from thalesmg/bridge-dryrun-flex-timeout-r52-20230816
...
fix(bridge): make dryrun health check timeout more malleable
2023-08-21 14:49:46 -03:00
befc260555
fix(gcp_consumer): remove unused resource_opts fields
...
Fixes https://emqx.atlassian.net/browse/EMQX-10789
2023-08-17 17:02:02 -03:00
ebecbd1545
fix(bridge): make dryrun health check timeout more malleable
...
Fixes https://emqx.atlassian.net/browse/EMQX-10773
- Makes the timeout for probing a bridge more malleable to account for differences between
each database.
- Increases GCP PubSub Consumer default health check timeout to account for GCP
slowness/throttling.
2023-08-17 09:21:19 -03:00
029b461a13
Merge pull request #11448 from thalesmg/gcp-consumer-403-20230815
...
fix(gcp_consumer): handle 403 responses
2023-08-17 09:04:00 -03:00
ba956ebe88
fix(gcp_consumer): handle 403 responses
...
Fixes https://emqx.atlassian.net/browse/EMQX-10736
2023-08-15 13:20:20 -03:00
23f5cea482
feat: handle strange key values when resolving placeholders
2023-08-14 13:39:38 -03:00
82b8538041
feat(gcp_producer): add support for defining message attributes and ordering key
...
Fixes https://emqx.atlassian.net/browse/EMQX-10652
2023-08-14 10:33:17 -03:00
5c8dc092a1
fix(http_bridge): don't attempt to convert headers to atoms
...
Fixes https://emqx.atlassian.net/browse/EMQX-10653
2023-08-07 13:08:34 -03:00
cbfca8c043
chore: merge master into release-51
2023-07-27 15:19:57 +02:00
243b8f5b67
chore: merge 'upstream/master' into v5.1.2
2023-07-21 13:25:46 +02:00
6cd503865b
fix(machine_boot): ensure `emqx_bridge` starts after its companion apps
...
We need to reverse the dependency of `emqx_bridge` and `emqx_bridge_*`, because the former
loads and starts bridges during its application startup. If the individual bridge
application being loaded has not started with its dependencies, the supervision tree will
not be ready for that.
2023-07-20 13:11:44 -03:00
05c3e023a9
chore(gcp_pubsub_consumer): unhide GCP PubSub Consumer bridge for e5.2.0
...
Fixes https://emqx.atlassian.net/browse/EMQX-10506
2023-07-17 11:24:21 -03:00
01b143c5ad
fix(resource): don't destruct error tuple
...
Otherwise, `emqx_resource:query` won't correctly deem the resource to
be unhealthy when there's an extra message.
2023-07-13 16:12:33 -03:00
0dff428efb
Merge pull request #11262 from thalesmg/fix-gcp-consumer-hc-20230712-master
...
fix(gcp_pubsub_consumer): fail health check when there are no workers
2023-07-13 15:09:02 -03:00
be7918aa41
fix(gcp_pubsub_consumer): fail health check when there are no workers
...
`ecpool` already returns an error even if the worker process is dead,
but we add the empty worker list clause here just for completeness.
2023-07-12 16:31:21 -03:00
f28510b3ad
refactor: HTTP connector into emqx_bridge_http app
2023-07-12 14:46:43 +02:00
b089fba100
refactor: rm ee_bridge and ee_connector application
2023-07-07 12:25:37 +08:00
1a058f6890
test(gcp_consumer): attempts to reduce flakiness
2023-06-30 12:49:23 -03:00
30e0b4be54
test(gcp_pubsub_consumer): add more tests and improve bridge
...
Fixes https://emqx.atlassian.net/browse/EMQX-10309
2023-06-28 14:08:40 -03:00
22356b7c25
chore: hide gcp pubsub consumer until e5.2.0
2023-06-22 10:05:52 -03:00
f01de238b2
chore: bump app vsns
2023-06-22 10:01:12 -03:00
0463828e84
refactor(gcp_pubsub): transform connector into opaque client
2023-06-20 15:27:42 -03:00
ffce6fefa8
docs: note that connector state must be immutable to avoid errors
2023-06-20 13:50:53 -03:00
ba96c725e5
feat: add time-based pull for safety
2023-06-20 11:36:35 -03:00
2ac2d4c037
refactor: addressing review comments
2023-06-20 11:15:13 -03:00
b442910ff1
feat(gcp_pubsub_consumer): implement GCP PubSub Consumer bridge
...
Fixes https://emqx.atlassian.net/browse/EMQX-10281
2023-06-19 16:04:12 -03:00
bb49482529
refactor(gcp_pubsub): move logging from connector to bridge
2023-06-19 15:59:00 -03:00
dabefdb01b
fix(gcp_pubsub): redact request when errors occur
2023-06-19 15:59:00 -03:00
a1690b3684
chore(gcp_pubsub): add stricter compile/xref flags
2023-06-19 15:59:00 -03:00
5375421954
refactor(gcp_pubsub): split connector into producer and reusable parts
2023-06-19 15:59:00 -03:00
b930f4cc73
Merge pull request #10987 from keynslug/fix/EMQX-9257/sep-placeholder
...
refactor: tear `emqx_plugin_libs` application apart
2023-06-09 17:02:14 +02:00
d6c1ee183f
refactor(pluglib): move `emqx_placeholder` to utils app
...
Also make user that existing code calls it directly.
2023-06-09 14:44:36 +03:00
ed9e29e769
refactor: refacor query_mode detection code
...
This commit refactor the query_mode resource detection code according to
a suggestion from @zmstone. This commit should not contain any
functional change except for a change of the Kafka producer bridge
config.
2023-06-08 16:26:55 +02:00
d9aac63f82
refactor(gcp_pubsub): always cleanup resources when calling `on_stop`
...
Fixes https://emqx.atlassian.net/browse/EMQX-9934
The way that this connector was structured already depended only on the resource ID. This
PR just makes it completely independent of the resource state term, and also treats an
already absent pool as a success when stopping it.
2023-06-07 14:34:22 -03:00
260fae296b
feat(gcp_pubsub): generate jwt tokens on demand without workers (5.1)
...
Fixes https://emqx.atlassian.net/browse/EMQX-9603
Rather than relying on a JWT worker to produce and refresh tokens, we
could just produce then on demand when pushing the messages to GCP
PubSub. That can generate a bit of extra work (as multiple processes
might realize it’s time to refresh the JWT and do so), but that
shouldn’t be much. In return, we avoid any possibility of not having
a fresh JWT when pushing messages.
2023-06-06 13:19:24 -03:00
46393343e2
chore: use `timeout_duration` types for timer fields
...
Fixes https://emqx.atlassian.net/browse/EMQX-10020
2023-06-05 11:46:38 -03:00
99796224d8
refactor(resource): rename `request_timeout` -> `request_ttl`
...
See
https://emqx.atlassian.net/wiki/spaces/P/pages/612368639/open+e5.1+remove+auto+restart+interval+from+buffer+worker+resource+options
2023-06-01 13:01:53 -03:00
1e25ebb64c
test(gcp_pubsub): attempt to fix flakiness
...
https://github.com/emqx/emqx/actions/runs/5125118728/jobs/9218520994?pr=10887#step:8:309
```
=CRITICAL REPORT==== 30-May-2023::19:19:34.887082 ===
"check stage" failed: error
{assertMatch,[{module,emqx_bridge_gcp_pubsub_SUITE},
{line,1066},
{expression,"? of_kind ( gcp_pubsub_request_failed , Trace )"},
{pattern,"[ # { reason := Error , connector := ResourceId } | _ ]"},
{value,[#{connector =>
<<"bridge:gcp_pubsub:emqx_bridge_gcp_pubsub_SUITE0005FCEE15534E9CD4CD02004CF10000">>,
msg => gcp_pubsub_request_failed,query_mode => async,
reason => {closed,"The connection was lost."},
recoverable_error => true,
'~meta' =>
#{gl => <0.17903.2>,
location =>
#Fun<emqx_bridge_gcp_pubsub_connector.19.19548918>,
node => 'test@127.0.0.1',pid => <0.19724.2>,
time => -576460610660164}}]}]}
Stacktrace: [{emqx_bridge_gcp_pubsub_SUITE,
'-do_econnrefused_or_timeout_test/2-fun-2-',3,
[{file,
"/__w/emqx/emqx/source/apps/emqx_bridge_gcp_pubsub/test/emqx_bridge_gcp_pubsub_SUITE.erl"},
{line,1066}]},
{emqx_bridge_gcp_pubsub_SUITE,do_econnrefused_or_timeout_test,2,
[{file,
"/__w/emqx/emqx/source/apps/emqx_bridge_gcp_pubsub/test/emqx_bridge_gcp_pubsub_SUITE.erl"},
{line,1022}]}]
```
2023-05-31 10:19:55 -03:00
324459990f
Merge branch 'release-50' into merge-r50-into-v50-20230524
2023-05-24 12:54:15 -03:00
4565acc600
fix: handle `infinity` timeout option in `ehttpc` (r5.0)
...
Fixes https://emqx.atlassian.net/browse/EMQX-9987
2023-05-24 10:33:54 -03:00
7d798c10e9
perf(buffer_worker): flush metrics periodically inside buffer worker process
...
Fixes https://emqx.atlassian.net/browse/EMQX-9905
Since calling `telemetry` is costly in a hot path, we instead collect
metrics inside the buffer workers state and periodically flush them,
rather than immediately as events happen.
2023-05-22 09:11:23 -03:00
383fec9dfc
docs: fix invalid links
2023-05-15 11:18:09 +08:00
90cf1ade74
chore: bump application versions
...
* emqx_connector 0.1.22
* emqx_bridge_gcp_pubsub 0.1.1
* emqx_bridge_kafka 0.1.2
* emqx_bridge_pulsar 0.1.1
* emqx_ee_connector 0.1.12
2023-05-02 17:29:22 +03:00
Kjell Winblad