* Add consoleuser and stamp default profile on known username in migration
* Refactor consoleuser to verify Id, fix seats on linux and default stamp
* Add default profile claim
* Add disable auto-claim of default profile and always fail close
* Add disable auto-claim flag to migration
* Adding timeout to console user on Linux and close library load on darwin
* Fixed failed close test
* Close both Dlopen for darwin
* Replace RegisterFunc with purego.Dlsym to avoid possible panic
* Fix freebsd tty enumeration
* Fix active profile migration logic and add test
* Log defaultClaimDisabled error once
* Guard against panicking console user lookup.
* Fix merge conflict
* Fix broken tests
* Implement OwnsProfile on Server
* (WIP) List profiles based on ownership by Identity
* (WIP) Migrate active_profile
* Fix status and list profiles
* Add profile stamping as active migration
* Add one-shot migration
* Only default profile fail open
* Use restricted write for config json
* Fix stale server config after stamp
* Fix OwnsProfile fallback to active profile
* Fix config concurrent reload during OwnsProfile check
* Move known check to inside stamp owner
* Update client/internal/profilemanager/service.go
Co-authored-by: cubic-dev-ai[bot] <191113872+cubic-dev-ai[bot]@users.noreply.github.com>
* Recover from dup active profiles that cannot be resolved with username.
* Add test for already owned profile during migration
* Improve stamping of fields in the config
* Apply suggestion from @cubic-dev-ai[bot]
Co-authored-by: cubic-dev-ai[bot] <191113872+cubic-dev-ai[bot]@users.noreply.github.com>
* Fix codespell and test comment
---------
Co-authored-by: cubic-dev-ai[bot] <191113872+cubic-dev-ai[bot]@users.noreply.github.com>
GetSessionExpiresAt took the exclusive lock for a plain field read, so
every caller queued behind writers and behind each other. The Android
SessionMonitor polls it from the main thread, and in the captured ANR
that is exactly where the main thread was blocked while hundreds of
peer-list callbacks held or waited on the same mutex.
d.mux is already an RWMutex and the other getters use RLock; this brings
the deadline read in line with them.
