Now that 4.1.0 is out, 4.0.x binary releases will
no longer be available publicly.
(cherry picked from commit cf98ba5a8c)
Conflicts:
.github/workflows/release-4.0.x-alphas.yaml
Key changes:
- endpoint variable to handle scraping multiple endpoints
- message size panels (new metric in 4.1)
- panels at the top of the Overview dashboard should be more up to date
(they show the latest value)
- values should be accurate if multiple endpoints are scraped
(previously, many would be doubled)
- Nodes table shows fewer volumns and shows node uptime
(cherry picked from commit c0368a0d24)
This release contains some logging improvements to avoid logging
large stack traces during normal operations such as rolling restarts.
(cherry picked from commit bd96f86dc4)
When debug logging is enabled, we log something at each log level
to test if logs are emitted. I don't think this is particularly useful,
but it's certainly annoying, because I constatnly need to filter
out these logs when searching if any errors happened during tests.
(cherry picked from commit 589e0b578c)
Co-authored-by: Michal Kuratczyk <michal.kuratczyk@broadcom.com>
This ensures that quorum_queues shuts down _before_
coordination where khepri run inside.
Quorum queues depend on khepri so need to be shut down first.
(cherry picked from commit 274f12f063)
[Why]
The CLI sometimes crashes early because it fails to configure the Erlang
distribution.
Because we use two CLI commands to watch the start of RabbitMQ, if one
of them fails, the Make recipe will exit with an error, leaving the
RabbitMQ node running.
[How]
We use a shell trap to stop the node if the shell is about to exit with
an error.
While here, we retry the `await_startup` CLI command several times
because this is the one failing the most. This is until the crash is
understood and a proper fix is committed.
(cherry picked from commit 3a278e7e7c)
[Why]
Many tests do not clean up their connections if they encounter a
failure. This affects subsequent testcases negatively.
(cherry picked from commit 97da746160)
[Why]
Khepri already managed retries if needed, we can just use a timeout.
Note that the timeout was already bumped to a more appropriate 5
minutes, which also matches what we had with Mnesia. However, with 10
retries by default, it meant that this timeout at the end of `init/1`
would thus be 5 * 10 = 50 minutes.
(cherry picked from commit 67d9d89f1a)
This allows restricting access to the /api/index.html and
the /cli/index.html page to authenticated users should the
user really want to. This can be enabled via advanced.config.
(cherry picked from commit 400e8006e5)
Move leader repair earlier in tick function to ensure more
timely update of meta data store record after leader change.
Also use RPC_TIMEOUT macro for metric/stats multicalls to improve
liveness when a node is connected but partitioned / frozen.
(cherry picked from commit 27ef97ecd7)
This should address crashes like this in (found in user's logs):
```
exception error: no case clause matching
[[{connection_details,[]},
{name,<<"10.0.13.41:50497 -> 10.2.230.128:5671 (1)">>},
{node,rabbit@foobar},
{number,1},
{user,<<"...">>},
{user_who_performed_action,<<"...">>},
{vhost,<<"/">>}],
[{connection_details,[]},
{name,<<"10.0.13.41:50142 -> 10.2.230.128:5671 (1)">>},
{node,rabbit@foobar},
{number,1},
{user,<<"...">>},
{user_who_performed_action,<<"...">>},
{vhost,<<"/">>}]]
in function rabbit_federation_mgmt:format/3 (rabbit_federation_mgmt.erl, line 100)
in call from rabbit_federation_mgmt:'-status/3-lc$^0/1-0-'/4 (rabbit_federation_mgmt.erl, line 89)
in call from rabbit_federation_mgmt:'-status/4-lc$^0/1-0-'/3 (rabbit_federation_mgmt.erl, line 82)
in call from rabbit_federation_mgmt:'-status/4-lc$^0/1-0-'/3 (rabbit_federation_mgmt.erl, line 82)
in call from rabbit_federation_mgmt:status/4 (rabbit_federation_mgmt.erl, line 82)
in call from rabbit_federation_mgmt:to_json/2 (rabbit_federation_mgmt.erl, line 57)
in call from cowboy_rest:call/3 (src/cowboy_rest.erl, line 1590)
in call from cowboy_rest:set_resp_body/2 (src/cowboy_rest.erl, line 1473)
```
(cherry picked from commit 6513d028e3)
AMQP Java client uses Netty 4.2, QPid JMS uses Netty 4.1. This commit
forces the use of Netty 4.2 (which is backward-compatible with 4.1).
(cherry picked from commit 12d094bdb3)
Michael Klishin
Loïc Hoguin
Michal Kuratczyk
Michael Klishin
dependabot[bot]
Karl Nilsson
David Ansari
mergify[bot]
Jean-Sébastien Pédron
Marcial Rosales
Arnaud Cogoluègnes