GoZ Relayer auto-update runner

I have made an automated script to update the relayer clients with tunable config, error handling and failure handling. Its been working for me very well, but I can't assure it handles every case

Please make a pull request if have any feature additions. But it shouldn't make the script less reliable than it is right now. Thanks!

Quick Start

1. Update commands in script.sh. Use docker-compose exec -T in case you are running via docker-compose.

COMMAND1="rly tx raw update-client gameofzoneshub-1a your-chain-id xxxxxxxxx"
COMMAND2="rly tx raw update-client your-chain-id gameofzoneshub-1a xxxxxxxxx"

2. Run the script

bash script.sh

You can see the output for the first run. Now sleep easy during GoZ.

For failure handling and ccall alerts, read below.

How to use

1. Update the config section and handle failure to your preference

   START_DELAY - How long to wait before first run. Useful if you've already run the commands recently

   TIMEOUT - How long to wait before retrying a command

   INTERVAL - What is the interval between success runs. For GoZ, something like 88m is ideal

   MAX_TRIES - Maximum failed attempts before it gives up calls you

   COMMAND1 - Command to update your relayer client 1. Use docker-compose exec -T when using docker-compose exec

   COMMAND2 - Command to update your relayer client 1

   TZ - Timezone for logging time

   Important: My handle_failure function calls me on failure using twilio voice API so that I can look into it. You can sign up for free or do something else here to be notified via slack, email, etc.

   But do not run without this as you have no way to know if something has failed

2. Run the script in tmux or use something like systemd to keep it running

   $ ./script.sh

How it works

It runs the two commands in a loop

rly tx raw update-client chain-id-1 chain-id-2 client-id-src

and

rly tx raw update-client chain-id-2 chain-id-1 client-id-dst

and uses jq to parse the output

If there is an error or the height value is 0 in the response, it considers it a failed response

On failure it gives a TIMEOUT and runs it again till MAX_TRIES

if it still failes, it will hanlde failure and halt

Fearures to be added

• Check balance on each run and handle low balance (Feel free to make a PR)

Contact

• You can reach me on telegram with id @sub4sh or email me at subash@audit.one