FTL v1.3.1

Improved garbage collection algorithm, added forward destinations over time data analysis (required for pi-hole/web#441)

FTL v1.3

Provide 24 hours of consecutive data for the dashboard if possible. We parse both files, pihole.log and pihole.log.1 to achieve this. Note that this will not work if the user flushed the Pi-hole log manually at some point. In order to ensure that only data within the given time interval of 24h is kept, FTL will collect and discard queries that are older than 24h at every full hour.

In addition, FTLs threads are now named, so that it can be seen more easily which thread is consuming how much CPU resources:

The individual client listener threads are labeled with the corresponding client IDs (the own ID can be queried by connected clients using >clientID).

FTL v1.2.1

Update multi-threading version of FTL to spawn new client connection handling threads in detached mode so that they can cleaned up after themselves when the clients disconnect. This will free otherwise dead memory.

FTL v1.2

New: FTL uses multiple threads to optimize workload.

You will be able to see the following tree structure:

• Main process: Catching signals and handling them, launching the Log parser and the Socket listener
  • Log parser: Waiting for and subsequently processing log changes
  • Socket listener: Waiting for and subsequently processing incoming socket connections, launching Client threads for every connecting client
    • Client threads: One client thread per concurrently connected client waiting for incoming requests and answering them

Note that FTL ensures thread safety, i.e. new log data will not be processed while communication with clients is currently ongoing. Similarly, client requests will be queued if new log data is currently being processed.

FTL v1.1

• Remove any 24h limitation from the overTime data (use Unix timestamps)
• Fine-tuning of wildcard blocking detection
• Memory optimizations

FTL v1.0

First official release of FTL