Snapserver expects the arguments `--tcp.bind_to_address` and
`--http.bind_to_address` instead of the `--tcp.address` (and http
equivalent) versions.
This caused the process to listen on `0.0.0.0` (for TCP and HTTP
sockets) regardless of the configuration value. It also never listend on
the IPv6 address `::` as our module system made the user believe.
This commit fixes the above issue and ensures that (at least for the TCP
socket) that our default `::` does indeed allow connections via IPv6
(to localhost aka ::1).
Sanitizers don't seem to be present on aarch64-darwin/macOS 12 (Monterey), so they are removed from the aarch64-darwin tests.
Switching from nativeBuildInputs to buildInputs and adding cc to the deps list caused some strange error messages to go away.
Although sanitizers can catch and prevent undefined behaviour during runtime, it has a significant impact on performance. They also cause issues on macOS where they can make compilation fail. The future goal is to instead utilize static analysis to prevent undefined behaviour as makeBinaryWrapper evolves.
On macOS, /tmp is a symlink to /private/tmp. When performing cd /tmp, and checking cwd - it won't match since it follows the symlink.
This caused test breakage on macOS but not Linux. Instead, use a folder which is not a symlink, and consistent across Linux and macOS.
* libflux: 0.124.0 -> 0.139.0; as specified in influxdb 2.1's release notes
* UI: 2.0.8 -> 2.1.2; matching what influxdb 2.1.1 requires
* Add checks in influxdb2's build to verify that libflux and the UI are at
the version specified by upstream.
* The CLI got split into a separate repository for reasons to do with influx's
cloud service, and its version has also been decoupled from the main influxdb
repository. Link it back into the influxdb2 derivation, for compatibility
with the previous derivation versions.
While it is a fact of life that aarch64-darwin is built on Hydra, it has
never formally been elevated from the Tier 7 state it was originally
assigned in RFC 0046. Since platform Tier status is not only
descriptive, but also normative, a consensus to commit to supporting
aarch64-darwin would need to be reached.
