* Update to Go 1.23.3
- Update to latest major version of Go 1.23.3 from 1.22.8.
- Update github.com/sasha-s/go-deadlock to address deadlock timer
issue we were seeing.
- Fix one of our tests to only reset the member variable we change
instead of the entire Opts parameter to avoid a data race during
testing.
* Add workaround for MSSQL TLS certificate container issue
* Add a core test logger to help capture the MSSQL container output
- I believe the if t.Failed prevents the logging of the container
logging as when executed the test isn't considered failed yet.
- Use a test core logger so that we can capture the container output
all the time and get it from the captured log files when the test
fails
* bump image tag to 2022-latest
---------
Co-authored-by: JM Faircloth <jmfaircloth@hashicorp.com>
- Get better test failure error messages by not shadowing the errors
when we are attempting to start the MSSQL docker container, so
we can fail the tests with the proper error message that is occuring
instead of mssqlhelper.go:60: Could not start docker MSSQL: %!s(<nil>)
* add inline cert auth to postres db plugin
* handle both sslinline and new TLS plugin fields
* refactor PrepareTestContainerWithSSL
* add tests for postgres inline TLS fields
* changelog
* revert back to errwrap since the middleware sanitizing depends on it
* enable only setting sslrootcert
* adds sslinline option to postgres conn string
* for database secrets type postgres, inspects the connection string for sslinline and generates a tlsconfig from the connection string.
* support fallback hosts
* remove broken multihost test
* bootstrap container with cert material
* overwrite pg config and set key file perms
* add feature flag check
* add tests
* add license and comments
* test all ssl modes
* add test cases for dsn (key/value) connection strings
* add fallback test cases
* fix error formatting
* add test for multi-host when using pgx native conn url parsing
---------
Co-authored-by: Branden Horiuchi <Branden.Horiuchi@blackline.com>
* timeout 'testListenerConnFn' waiting on the server connection after 3 secs
* return the invalid upstream error so the library knows not to stop listening/serving
* update go-proxyproto to use fork/tag
* test that fails before library and code update, but passes afterwards
* improve: extract logic to select either locking.DeadlockRWMutex or locking.SyncRWMutex out of CreateCore and into their own functions
* add copyright header for new files
* move new files to helper/locking package
* adjust names of helper functions moved to locking package
* Add new func RetryUntilAtCadenceWithHandler to testhelpers.go.
* Correct godoc comment for RetryUntilAtCadenceWithHandles.
Co-authored-by: Steven Clark <steven.clark@hashicorp.com>
---------
Co-authored-by: Steven Clark <steven.clark@hashicorp.com>
We have many hand-written String() methods (and similar) for enums.
These require more maintenance and are more error-prone than using
automatically generated methods. In addition, the auto-generated
versions can be more efficient.
Here, we switch to using https://github.com/loggerhead/enumer, itself
a fork of https://github.com/diegostamigni/enumer, no longer maintained,
and a fork of the mostly standard tool
https://pkg.go.dev/golang.org/x/tools/cmd/stringer.
We use this fork of enumer for Go 1.20+ compatibility and because
we require the `-transform` flag to be able to generate
constants that match our current code base.
Some enums were not targeted for this change:
When creating database connections, there is a race
condition when multiple goroutines try to create the
connection at the same time. This happens, for
example, on leadership changes in a cluster.
Normally, the extra database connections are cleaned
up when this is detected. However, some database
implementations, notably Postgres, do not seem to
clean up in a timely manner, and can leak in these
scenarios.
To fix this, we create a global lock when creating
database connections to prevent multiple connections
from being created at the same time.
We also clean up the logic at the end so that
if (somehow) we ended up creating an additional
connection, we use the existing one rather than
the new one. This by itself would solve our
problem long-term, however, would still involve
many transient database connections being created
and immediately killed on leadership changes.
It's not ideal to have a single global lock for
database connection creation. Some potential
alternatives:
* a map of locks from the connection name to the lock.
The biggest downside is the we probably will want to
garbage collect this map so that we don't have an
unbounded number of locks.
* a small pool of locks, where we hash the connection
names to pick the lock. Using such a pool generally
is a good way to introduce deadlock, but since we
will only use it in a specific case, and the purpose
is to improve performance for concurrent connection
creation, this is probably acceptable.
Co-authored-by: Jason O'Donnell <2160810+jasonodonnell@users.noreply.github.com>
* helper/metricsutil: Update usage of expfmt
The previous exposed constants are no longer exposed and we can use expfmt.Format instead.
https://github.com/prometheus/common/pull/576
* reodered imports
---------
Co-authored-by: Matthias Loibl <mail@matthiasloibl.com>
* add metrics for secret sync clients
* changelog
* remove enterprise tag from changelog
* fix test and make clearer what it's testing
* replace with underscores
* Add Logger to BackendConfig
* EntryFormatter use logger and recover panics
* Added TODO to consider
* Add 'name' to entry formatter
* Add test for the panic
* Fix NoopAudit with update params
* emit counter metric even when 0
* Fix vault package tests
* changelog
* Remove old comment during test writing
* add gosimport to make fmt and run it
* move installation to tools.sh
* correct weird spacing issue
* Update Makefile
Co-authored-by: Nick Cabatoff <ncabatoff@hashicorp.com>
* fix a weird issue
---------
Co-authored-by: Nick Cabatoff <ncabatoff@hashicorp.com>
