Cycles being drained. How to understand what's behind it?

My team is handling an EXT fungible token canister and the cycles are quickly running out. It is consuming almost 7T cycles per day, which is 2-3x the amount consumed by other similar canisters. Does anyone have any suggestions on how to break down the cycle consumption of a canister to try to understand what is consuming the most amount of cycles?

We had a similar issue a while back that we solved, if you search the forum you should find a thread about it

2 Likes

@jonit is probably referring to this thread Canister burning through cycles, did you already take a look at that?

4 Likes

Thank you for directing me toward that thread. I will have a look!

Thanks for directing me toward that thread. I will have a look!

To save you some time:

I recommend calling initCap if you haven’t already. This will initialize the CAP event service. If you don’t do this your canister will trap when it attempts to report a transaction to CAP. Your canister will catch this error and try again, hundreds of times within a few seconds, and will continue this behavior until you either restart the canister or initialize the CAP service using the command I mentioned in the first sentence.