* Add test for #4752: Nil pointer dereference in waitForPluginLoad
This test demonstrates the bug where waitForPluginLoad() panics with a nil
pointer dereference when a plugin fails during startup before the reattach
config is set.
The test creates a runningPlugin with reattach=nil and closes the failed
channel, simulating a plugin that fails in startPluginProcess before
initializePlugin is called.
* Fix#4752: Add nil check for reattach in waitForPluginLoad
Adds a nil pointer check before accessing p.reattach.Pid when a plugin
fails during startup. If the plugin fails before the reattach config is
set (e.g., in startPluginProcess), reattach will be nil and the code
would previously panic when trying to log the PID.
The fix checks if reattach is nil and logs an appropriate message in
each case, preventing the panic while still providing useful debugging
information.
* Add test for #4799: Race condition in rate limiter map access
This test demonstrates the race condition in getUserDefinedLimitersForPlugin
where the userLimiters map is read without mutex protection while being
concurrently written by other goroutines.
Run with: go test -race -v -run TestPluginManager_ConcurrentRateLimiterMapAccess ./pkg/pluginmanager_service
* Fix#4799: Add mutex protection for rate limiter map access
Protected all accesses to m.userLimiters map with RWMutex:
- getUserDefinedLimitersForPlugin: Added RLock for map read
- getPluginsWithChangedLimiters: Added RLock for map iteration
- handleUserLimiterChanges: Added Lock for map write
- refreshRateLimiterTable: Added RLock for map iteration
- setRateLimiters: Added RLock for map read
This prevents data races when the map is concurrently read and written
by multiple goroutines.
* Update PopulateChildren.PopulateChildren to take plugin instance into account
* Fix GetNewConnectionStateFromConnectionInsertSql to use connection names not map of connections
- Execute RefreshConnections asyncronously
- Add connection_state table to indicate the loading state of connections
- Optimise RefreshConnections by cloning connection schemas
- Add locking to ensure only a single instance of RefreshConnections runs
- Start executing queries without waiting for connections to load, add smart error handling to wait for required connection
- Optimise autocomplete for high connection count
- Autocomplete and inspect data available before all conections are refreshed
- Update file watcher to respond to CHMOD, so thaat it pickes up deletion of file contents
Closes#3394Closes#3267
