We have noticed that since 20-09 we have heavy drainage of cycles on some of our canisters. Is it possible to check what exactly is draining cycles? Before that date our canisters used like 0.05 T cycles per day. Now it is 0.5T cycles per day (10 times more) and it is ongoing usage.
Some examples:
ICPunks:
ICats:
Given usage of 0.5T per day, the cost of running canister is at about 20$ per month, it is quite an amount of money. We have not experienced increased usage of our marketplace or the number of transactions on canister itself.
If it is some kind of drainage, how can we check what is happening and how can we prevent it?
Back when we launched PetBots we had a huge cycle leak that we couldn’t identify. Ultimately, I was able to identify the problem after installing the CanisterGeek logging utility and tagging each update method and nested methods. This resulted in a log output that showed we were calling another service 100s of times a minute.