summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* Finish an incomplete comment sentenceMichael Klishin2019-07-021-1/+1
|
* Remove a test we decided againstMichael Klishin2019-07-021-3/+0
| | | | | | | | | | | Integration tests in rabbitmq_auth_backend_oauth2 cover more ground. In addition we would have to introduce a dummy backend that supports state updates which @acogoluegnes and I decided against. So all the testing currently happens in the only backend plugin that actually supports credential expiration and state/secret updating.
* Avoid a case clause by renaming a binding, spotted by @lukebakkenMichael Klishin2019-07-011-4/+4
|
* Merge branch 'master' into oauth2-credential-expiration-supportArnaud Cogoluègnes2019-07-015-332/+841
|\
| * Update rabbitmq-components.mkJean-Sébastien Pédron2019-06-281-0/+9
| |
| * Update erlang.mkJean-Sébastien Pédron2019-06-281-331/+756
| |
| * Merge pull request #2043 from rabbitmq/qq_metrics_tweakGerhard Lazu2019-06-252-0/+7
| |\ | | | | | | Qq metrics tweak
| | * Set ra wal_max_size_bytes default if not setkjnilsson2019-06-241-0/+6
| | | | | | | | | | | | To match it in unit with the rest of rabbit.
| | * pass metrics keykjnilsson2019-06-201-0/+1
| | |
| * | Merge pull request #2041 from luos/fix_cleanup_of_exclusive_queues_masterMichael Klishin2019-06-211-1/+69
| |\ \ | | |/ | |/| add test for cleanup of exclusive queues in case of restart or partition
| | * add test for cleanup of exclusive queues in case of restart or partitionLajos Gerecs2019-06-211-1/+69
| |/ | | | | | | | | | | | | | | | | | | | | When using different than client-local queues exclusive ones can be placed on a different host than the client. When there is a partition or restart these queues are stuck because they are not cleaned up in the nodes as the connection is still alive when the node goes offline, however the queue is dead in the partition. These queues can not be redeclared or consumed on 3.7.x. Related: #2039
* | CosmeticsMichael Klishin2019-06-291-1/+7
| |
* | Return a complete connection recordMichael Klishin2019-06-281-2/+2
| | | | | | | | Corrects a typo in 97f3f52.
* | Remove a debug log line that's meant to be temporaryMichael Klishin2019-06-281-1/+0
| |
* | Update user authn/authz state correctlyMichael Klishin2019-06-281-1/+1
| | | | | | | | Spotted by @acogoluegnes.
* | Error reporting for invalid secrets (tokens) WIPMichael Klishin2019-06-271-4/+10
| |
* | Propagate updated user/authn/authz state to channelsMichael Klishin2019-06-242-8/+31
| | | | | | | | | | | | | | | | | | Channel processes use it for authz backend checks. The record is identical to that of connection process (and originates from it), so a simple cast to update the field is appropriate: any errors would be handled and communicated back by the connection itself. Per discussion with @acogoluegnes and @kjnilsson.
* | WIP: credential expiration support prototypingMichael Klishin2019-06-225-4/+61
|/
* Delete server if add_member failskjnilsson2019-06-181-1/+1
| | | | Addressing a TODO of that effect.
* Replace spaces by underscores when forging module nameArnaud Cogoluègnes2019-06-141-1/+3
| | | | | | | | | | | | Plugins can provide a module to be called to extract extra information at authentication time. The module name is based on the protocol used. Some protocols have a space in their name, like "Web MQTT", so spaces in such cases must be replaced by underscores to come up with a valid Erlang module name. [#166271318] References #1767
* Merge pull request #1982 from rabbitmq/rabbitmq_prelaunch_ipv6Luke Bakken2019-06-131-0/+1
|\ | | | | Workaround for IPv6 EPMD
| * Add RABBITMQ_SERVER_ADDITIONAL_ERL_ARGSGabriele2019-05-311-0/+1
| | | | | | | | | | With this patch it is possible to pass variables to prelaunch state.
* | Merge pull request #2034 from rabbitmq/rabbitmq-server-2030D Corbacho2019-06-132-0/+15
|\ \ | | | | | | Make default worker pool size configurable
| * | Make default worker pool size configurableMichael Klishin2019-06-132-0/+15
|/ / | | | | | | Part of rabbitmq/rabbitmq-server#2030.
* | Merge pull request #2033 from rabbitmq/rabbitmq-cli-330Michael Klishin2019-06-131-1/+16
|\ \ | | | | | | Schema information in mnesia for cli command
| * | Schema information in mnesia for cli commandDiana Corbacho2019-06-131-1/+16
| | | | | | | | | | | | [#164550205]
* | | amqqueue*: Uniformize APIJean-Sébastien Pédron2019-06-132-14/+277
| | | | | | | | | | | | | | | | | | Now, both modules export the same set of functions. This will help with the backport to `v3.7.x` and thus the life of plugin developers who will not have to have one copy of their plugin per RabbitMQ version.
* | | include/amqqueue.hrl: Fix indentationJean-Sébastien Pédron2019-06-131-9/+9
| | |
* | | Merge pull request #2031 from rabbitmq/rabbitmq-shovel-60Michael Klishin2019-06-131-0/+25
|\ \ \ | |/ / |/| | Update log configuration tests for rabbitmq/rabbitmq-common#328
| * | Update log configuration tests for rabbitmq/rabbitmq-common#328Michael Klishin2019-06-121-0/+25
|/ /
* | Head message timestamp: make the test more robustMichael Klishin2019-06-111-4/+16
| |
* | Reference a public IPv4 that has a PTR recordGerhard Lazu2019-06-101-9/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Public IPv6 192.240.153.117 is managed by OVH US and has a PTR 117.153.240.192.in-addr.arp record that resolves to www.rabbitmq.com. It would be nice to move to an IPv4 & IPv6 that are managed by our wider org. dig 117.153.240.192.in-addr.arpa PTR +short www.rabbitmq.com. whois 192.240.153.117 | grep -i Organiz Organization: OVH US LLC (OUL-16) cc @michaelklishin @trevormarshall
* | Merge pull request #2019 from ↵Michael Klishin2019-06-075-46/+71
|\ \ | | | | | | | | | | | | rabbitmq/rabbitmq-server-1767-protocol-specific-ctx-in-authn-authz Propagate protocol-specific context to authorization
| * \ Merge branch 'master' into ↵Michael Klishin2019-06-0716-81/+224
| |\ \ | |/ / |/| | | | | rabbitmq-server-1767-protocol-specific-ctx-in-authn-authz
* | | Handle rejected queues on confirm.queue_confirm_refactorDaniil Fedotov2019-06-043-29/+34
| | | | | | | | | | | | | | | | | | | | | When one queue confirms a message and another queue failed without confirming it, the message should be rejected. Fixes some races between confirms and rejects.
* | | Switch msg_id_to_channel from gb_tree to maps.Daniil Fedotov2019-06-042-14/+12
| | |
* | | Reject a message before discarding it from BQ.Daniil Fedotov2019-06-041-3/+10
| | | | | | | | | | | | | | | discard/4 confirms a message, which can cause races between confirms and rejects.
* | | Merge pull request #2022 from rabbitmq/ra-metricsMichael Klishin2019-06-042-0/+13
|\ \ \ | | | | | | | | Add channel info for pending raft commands
| * | | Add channel info for pending raft commandskjnilsson2019-06-042-0/+13
| | | | | | | | | | | | | | | | [#165184345]
* | | | Update rabbitmq-components.mkGerhard Lazu2019-06-031-1/+1
| | | |
* | | | Type fixes to keep dialyzer happyDaniil Fedotov2019-05-292-3/+3
| | | |
* | | | Merge pull request #2018 from rabbitmq/rabbitmq-cli-302Michael Klishin2019-05-292-2/+26
|\ \ \ \ | | | | | | | | | | 'check' function for health check of virtual hosts
| * | | | 'check' function for health check of virtual hostsDiana Corbacho2019-05-282-2/+26
| | | | | | | | | | | | | | | | | | | | [#163451547]
* | | | | Merge pull request #2015 from Invizory/dead-letter-rejected-publishedDaniil Fedotov2019-05-298-34/+130
|\ \ \ \ \ | | | | | | | | | | | | Add an option to dead-letter rejected publishes
| * | | | | Add remaining tests for `reject-publish-dlx`Arthur Khashaev2019-05-283-15/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add tests for `reject-publish-dlx` overflow strategy for all corresponding tests for `reject-publish` strategy. The tests are grouped to avoid code duplication. See #1443
| * | | | | Add overflow tests to `confirms_rejects_SUITE`Arthur Khashaev2019-05-281-13/+35
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change splits the tests into groups for both the existing `reject-publish` overflow strategy and the new `reject-publish-dlx`. The corresponding queue names have been suffixed to make the tests in the groups independent and thus parallel. See #1443
| * | | | | Add an option to dead-letter rejected publishesArthur Khashaev2019-05-274-6/+52
|/ / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add `reject-publish-dlx` overflow strategy, which is similar to `reject-publish` strategy, but also dead-letters rejected messages. Closes #1443
| | | * | Be more tolerant in rabbit_channel:get_amqp_params/2Arnaud Cogoluègnes2019-06-071-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The function can be called outside of any AMQP connection (e.g. from the management plugin), so trying to get any protocol-specific information from the connection state does not make much sense. References #1767
| | | * | Propagate protocol-specific context to authorizationArnaud Cogoluègnes2019-05-295-45/+69
| |_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Propagate protocol-specific context like MQTT client ID to authentication and authorization backends. The propagation was already there for topic authorization. Resource check needed propagation and a new argument in the backend function. Virtual host check just needed propagation, as there was already a context parameter (for peer address). A context parameter and propagation was already there for login-related functions. Unfortunately the context API is not fully consistent, due to historical reasons: proplists are the historical data structure, maps are used whenever possible. References #1767
* | | | Improving confirms test stability and error reportingDaniil Fedotov2019-05-271-3/+10
| | | | | | | | | | | | | | | | | | | | | | | | Add process messages to the error message. Try harder to kill queues.
View Lorry Controller Status