FAQ

General

Where should I go for support?

If, after carefully reading the Wiki and the rest of this FAQ, you can visit our Discord. Make sure to read the rules channel carefully to avoid getting banned.

For bugs or support requests, feel free to also open an issue on our Github - just make sure to use the correct template and to fill out all the information!

What scanner should I use?

We have support a wide variety of scanners and forks - including Map’A’Droid and RealDeviceMap. PA relies on information from your scanner via webhook - as a result, certain features may be supported by certain scanners.

How can I get request features?

Feel free to open an issue on our Github - just make sure to use the correct template and to fill out all the information!

Discord is NOT a good place to request new features - it’s highly likely your request will be missed or ignored. If you want to make sure you get a response, please use the Github page.

Do you accept Pull Requests?

Yup! Feel free to open one against our Github - if is a significant change, it is probably best to consult the team before sinking too much time in.

How else can I support PA?

Besides submitting PRs or helping with the Wiki, you can support Deadly (the original creator and maintainer of PA) via Patreon. To buy someone on the team a drink, use the !support command in the PA Discord for the relevant links.

Startup Errors

PA throws a`ValueError` error about a JSON file?

First, make sure you haven’t used Notepad or TextEdit on your configuration files - they change the encoding on your file and cause issues.

Second, make sure your file is in the correct JSON format - Google “JSON Formatting” and use a `JSON Editor https://www.jsoneditoronline.org/`__ to verify you are in the right format.

Error about a ‘module named parse’?

PA only supports Python3 (version 3.7+), and not Python2.

ModuleNotFoundError: No module named X

After updating PA, make sure you always check for updated requirements. Run the command pip3 install -r requirements.txt --upgrade from the PA folder.

Additionally, ensure you are using Python3 and pip3 instead of Python2 utilities.

Runtime Errors

DTS reporting “Unknown”

If you see a DTS being replaced by “?”, “???”, or “Unknown”, that usually means that your scanner didn’t send enough information to PA for it to determine its value. This is usually one of two reasons - the scanner might not know the value, or else it doesn’t support the DTS.

Gym Name DTS reporting “Unknown”

Some scanners do not send relevant gym details through each webhook - to compensate up for this, PA caches information between different webhooks. However this cache can take a while to build up. You can also use the file cache, which will save and load it’s contents to a file in between runtimes.

Why didn’t I get an alert for this Event that is on my Map?

Make sure that your filters are set up exactly as you want them - in particular, make sure you understand the “is_missing_info” restriction. You can use the webhook tester (python3 tools/webhook_test.py) to recreate an event to verify they are working correctly.

Finally, check your both PA and your scanner for any errors related to webhooks. If your scanner doesn’t send the info, PA can’t process it.

Error 10053 (Windows) or Error 53 (Linux)

This error means that PA tried to use a connection that has already been closed. It often happens when the server PA is running on is overloaded. Check your CPU and Disk IO - if it is spiking when you get these errors, your server is under heavy load. There some things you can still do:

  • If on Windows, use Linux (Even a VM will see huge improvements).
  • Optimize PA and your scanner setup (see below).
  • Run PA, your scanner, or your database on separate machines if possible.
  • If all else fails, upgrade your hardware.

How can I optimize PA?

In terms of efficiency Rules are better than Managers, and Managers are better than instances. As such, reducing the number of instances by using Managers and reducing the number of Managers by using Rules.

This holds true with one caveat - due to limitations with the Python GIL, each PA process is limited to a single processor core. If you wish to take advantage of multiple cores, running instances may increase your throughput. Do not follow any advice to use a load balancer - PA is a stateful application, and this will cause problems. Instead, pair instances of your scanner with instances of PA that make logical sense.

Finally, you can lower the ‘concurrency’ setting for PA to a lower value - this limits the number of connections PA can handle at once. Lowering this value will reduce overhead and increase response times.

If you need additional advice, feel free to ask in our Discord.

How can I optimize my scanner?

For the most part, you should consult the maintainer of your scanner for tips. Generally, you want to follow the same logic and work towards consolidating instances.

For Map’A’Droid in particular, there are default values you can adjust (particularly around sending webhooks): + webhook_excluded_areas - list of area names to exclude elements from within

an area to be sent to a webhook.
  • webhook_max_payload_size - Split up the payload into chunks.