summaryrefslogtreecommitdiff
path: root/deps
Commit message (Collapse)AuthorAgeFilesLines
* Add documentation for the paginationrabbitmq-managenet-2615Gabriele Santomaggio2020-12-141-8/+78
| | | | fixes: https://github.com/rabbitmq/rabbitmq-server/issues/2615
* rabbitmq-upgrade(8): add missing commandsMichael Klishin2020-12-131-1/+22
| | | | References rabbitmq/rabbitmq-website#1109
* Add chunk header parameter to send_file callbackArnaud Cogoluègnes2020-12-091-1/+1
|
* Merge pull request #2674 from rabbitmq/rabbitmq-server-2673-followupLuke Bakken2020-12-081-0/+1
|\ | | | | Pass RABBITMQ_NODENAME via Windows service environment
| * Pass RABBITMQ_NODENAME via Windows service environmentrabbitmq-server-2673-followupLuke Bakken2020-12-071-0/+1
| | | | | | | | | | | | | | | | Without this change using anything other than `rabbit` or the `rabbitmq-env-conf.bat` file will result in `erlang_dist_running_with_unexpected_nodename` Follow-up to #2673 cc @dumbbell @michaelklishin
* | Fix failure tests in stream pluginArnaud Cogoluègnes2020-12-081-3/+3
|/
* Merge pull request #2673 from ↵Luke Bakken2020-12-071-0/+1
|\ | | | | | | | | luos/allow_configuring_rabbitmq_base_multiple_times2 Windows service: allow overriding base service location
| * allow configuring multiple rmq bases for multiple servicesLajos Gerecs2020-12-051-0/+1
| | | | | | | | | | | | Currently RABBITMQ_BASE is always dynamically picked up from the environment. This change would fix it at the time of configuration of the service allowing multiple RabbitMQ services to be configured.
* | Send shutdown message to non network/direct connectionArnaud Cogoluègnes2020-12-041-1/+5
| | | | | | | | | | | | | | | | Connections to the stream plugin does not have a type, so they can trigger some function_clause errors. This was the case when trying to close a connection from rabbit_connection_tracking module. The function now falls back to a simple gen_server call to the connection process for connections without a type.
* | Wait longer for replicas to come up in stream pluginArnaud Cogoluègnes2020-12-043-2/+2
| |
* | Handle errors in crashing_queues_SUITEkjnilsson2020-12-034-5/+22
| | | | | | | | | | | | | | | | | | | | As the connection may crash during the previous declaration and a caught error would be returned in amqp_connection:open_channel/1 that wasn't handled previously. Exactly how things fail in this test is most likely very timing dependent and may vary. Also fixes mqtt test where the process that set up a mock auth ETS table was transient when an rpc timeout was introduced
* | Handle undefined args caserabbitmq-server-2668Luke Bakken2020-12-021-0/+1
| | | | | | | | Fixes #2668
* | Add test that fails prior to the change for #2668Luke Bakken2020-12-022-1/+38
| |
* | Use ?assertEquals hereMichael Klishin2020-12-011-2/+2
| |
* | WordingMichael Klishin2020-12-011-2/+2
| |
* | Remove unused code line.Mirah Gary2020-11-301-2/+0
| |
* | Change per-object endpoint to `/metrics/per-object`.Mirah Gary2020-11-305-6/+6
| | | | | | | | This conforms with other http endpoints.
* | Document `/metrics/per_object` endpoint.Michal Kuratczyk2020-11-301-1/+8
| | | | | | | | | | | | [#175264017] Signed-off-by: Mirah Gary <mgary@pivotal.io>
* | Deleted unnecessary patterndcorbacho2020-11-301-2/+1
| | | | | | | | :registry matches any registry, it's only a variable
* | Add /metrics/per_object endpointMichal Kuratczyk2020-11-304-13/+54
| | | | | | | | | | | | | | | | | | Regardless of the value of `return_per_object_metrics`, this endpoint always returns per-object metrics. This allows scraping both endpoints at different intervals or scraping per-object metrics only during debugging. Co-authored-by: Mirah Gary <mgary@vmware.com>
* | Convert AMQP 091 timestamp to millisecondArnaud Cogoluègnes2020-11-302-3/+3
| | | | | | | | For start offset in stream queue.
* | Support timestamp offsets for stream consumersdcorbacho2020-11-302-0/+86
| |
* | Remove stream field from publish commandArnaud Cogoluègnes2020-11-304-43/+35
| | | | | | | | | | Only the publisher ID is necessary now the publisher declaration is mandatory before publishing.
* | Include publisher state managementArnaud Cogoluègnes2020-11-301-36/+80
| |
* | Enforce publisher ID uniquenessArnaud Cogoluègnes2020-11-301-36/+52
| |
* | Add query publisher sequenceArnaud Cogoluègnes2020-11-303-3/+55
| |
* | Declare publisher in stream testArnaud Cogoluègnes2020-11-301-6/+20
| |
* | Start adding publishing dedup support for streamsArnaud Cogoluègnes2020-11-305-19/+120
| |
* | Include publish dedup in stream protocolArnaud Cogoluègnes2020-11-303-5/+50
| |
* | Add failing direct connection safe call timeout test/mockAyanda-D2020-11-271-1/+18
| |
* | Add erlang client safe_call_timeouts testsAyanda-D2020-11-262-3/+102
| |
* | Add timeouts to direct connection establishmentAyanda-D2020-11-263-7/+16
| | | | | | | | | | | | | | | | | | operations. At some point, these have to timeout to avoid having lingering connection processes attempting remote connections under unstable conditions. Improve handling of potential call timeout exceptions in amqp_gen_connection.
* | Ensure safe amqp client call timeouts on connectionAyanda-D2020-11-263-1/+40
|/ | | | | | | | | | | establishment. This guarantees that the effective call timeouts are always safe, i.e. granting enough time to the underlying network operations, which must always timeout first in case of any unexpected lingering operations leading to timeouts. This eliminates the chance of leaking connection processes when call timeouts elapse, while underlying remote cconnection establishment call was still taking place.
* AMQP 1.0 Java tests: bump dep versionsMichael Klishin2020-11-231-2/+2
| | | | (cherry picked from commit 9963bb5b15166574d920f57bf1b05e575b436e5c)
* AMQP 1.0 Java client tests: target a reasonably modern lang versionMichael Klishin2020-11-231-2/+2
| | | | (cherry picked from commit 539b8d8b416b7295f0f7c7da137b59fc756b401e)
* QQ: Emit release cursor for empty basic getsqq-basic-get-snapshotskjnilsson2020-11-192-3/+14
| | | | | | Else an application that polled an empty quorum queue frequntly using basic.get would never result in a snapshot being taken and results in unlimited log growth.
* Merge pull request #2636 from rabbitmq/rabbitmq-stomp-gh-153-monorepoMichael Klishin2020-11-192-1/+66
|\ | | | | Test stream queues with STOMP
| * Test stream queues with STOMPrabbitmq-stomp-gh-153-monorepodcorbacho2020-11-172-1/+66
| | | | | | | | | | | | Stream queues require prefetch count and manual acknowledgment (cherry picked from commit 45d3c3fecbf9f9782851bf580620a50bb4692a45)
* | Merge pull request #2582 from rabbitmq/rabbitmq-amqp1.0-gh-104-monorepoMichael Klishin2020-11-192-0/+4
|\ \ | | | | | | Test stream queues with AMQP 1.0
| * | Test stream queues with AMQP 1.0rabbitmq-amqp1.0-gh-104-monorepoLuke Bakken2020-11-162-0/+4
| | | | | | | | | | | | | | | | | | | | | Original PR https://github.com/rabbitmq/rabbitmq-amqp1.0/pull/104 cc @kjnilsson cc @dcorbacho
* | | Ensure pg2 group is created only when links are created and removed with themrabbitmq-server-2497dcorbacho2020-11-181-2/+8
| |/ |/| | | | | | | | | The queue decorator is called even when no links are present, which caused the pg2 group to be created. At the same time, deleting a link wasn't deleting the queue group. Closes #2497
* | rabbit_common: Remove the rabbitmq-github-actions Erlang.mk pluginJean-Sébastien Pédron2020-11-175-184/+3
| | | | | | | | This is unused after the switch to the "monorepository".
* | Merge pull request #2627 from rabbitmq/stream-purgeMichael Klishin2020-11-171-0/+2
|\ \ | | | | | | Hide Purge option for stream queues
| * | Hide Purge option for stream queuesstream-purgedcorbacho2020-11-171-0/+2
| | |
* | | Remove unused .github directoriesJean-Sébastien Pédron2020-11-1769-23848/+0
| | | | | | | | | | | | | | | They were valid until the switch to the "monorepository" when everything was merged into a single Git repository.
* | | Move issue and PR templates to monorepo rootMichael Klishin2020-11-172-64/+0
| | |
* | | Wording and additional usage docsMichael Klishin2020-11-171-2/+8
| | | | | | | | | | | | (cherry picked from commit 3e3ef927b2471ec90899f449919bbdfa74bc5311)
* | | Allow "all" as a feature flag nameMichal Kuratczyk2020-11-172-1/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | `rabbitmqctl enable_feature_flag all` enables all currently disabled feature flags. While https://github.com/rabbitmq/rabbitmq-cli/issues/455 asked for `--all`, I decided to just use "all" for consistency with `rabbitmq-queues rebalance all`. Closes rabbitmq/rabbitmq-cli#455 (cherry picked from commit af9e877848cfe4434de697d756e64014d180a945)
* | | Check if member processes are alive in metadata commandArnaud Cogoluègnes2020-11-172-73/+93
| |/ |/| | | | | In case the Mnesia record is stale.
* | Merge pull request #2504 from rabbitmq/qq-credit-modePhilip Kuryloski2020-11-163-47/+45
|\ \ | | | | | | Use correct credit mode x-credit