Closed markbrown4 closed 4 days ago
Describe the bug
We've started seeing inconsistent errors with various plugins(gcloud kubectl postgres) on different machines, starting around ~2 days ago.
One of the errors that we've been seeing is that the postgres database directory isn't specified. e.g.
mise exec -- pg_ctl -o "-F -p 55432" restart > pg_ctl: no database directory specified and environment variable PGDATA unset
It seems specific to --shims, this workaround resolves the issue.
--shims
# ~.zprofile eval "$(/opt/homebrew/bin/brew shellenv)" - eval "$(mise activate zsh --shims)" + eval "$(mise activate zsh)"
# mise.toml [tools] postgres = "15.4" [plugins] postgres = "https://github.com/smashedtoatoms/asdf-postgres.git#4592b6406292d09626124982823fa0e49b56a764"
Have there been any changes recently to shims or zsh activation that could explain this kind of issue?
After upgrading to today's release 2024.11.23 these issues appear to be resolved.
2024.11.23
Describe the bug
We've started seeing inconsistent errors with various plugins(gcloud kubectl postgres) on different machines, starting around ~2 days ago.
One of the errors that we've been seeing is that the postgres database directory isn't specified. e.g.
It seems specific to
--shims
, this workaround resolves the issue.Have there been any changes recently to shims or zsh activation that could explain this kind of issue?