Simulation stopped working after #1971

[mmdiego]

I've updated my local copy of zenbot after some months and simulation stopped working after this commit.
So I checked out a fresh copy of zenbot and had the same results. If I rolled back to the commit just before #1971, which is ea7929f it starts working again.
I'm testing it using:
zenbot.js sim binance.BNB-USDT --strategy=macd --period=1m --days=1
Maybe it's something related with the exchange?
I'm using node v8.3, but the same happens with v10.

[jorisw]

That’s really weird mmdiego, as I’m using the same exchange as you are, also used both Node 8.3 and 10, and the commit you referenced, I wrote in order to fix the problem.

I am running on that commit with a simulator that never stops.

I could only suggest debugging the portion of code that I made changes to in my PR. And I’d be curious to know how many other users chime in with this problem.

[mmdiego]

I'm trying to debug it, however I´m not an expert in Node.
I've added many prints to see what was happening inside getNext(). I found that it's getting stucked after the second time it's called.

However, I also found that the behavior depends on the "limit" param. I've changed that value to 10 while debugging to simplify the execution. But then I also changed it to 5000 and it started working correctly! However, if I change the simulation to run on BTC-USDT, it started getting stucked again.

I will continue debbugging and I'll post my console output with the prints I've added, but in the meanwhile, can you try your simulation lowering down the limit param?

Thanks!

[jorisw]

That is very helpful info, mmdiego. I’ll look into it asap.

[james-ingold]

Having the same issue, but if I revert to a previous #1971 sim.js, then I get the issue that @jorisw was originally fixing. 😄
I think it will work if you just change this line (294 sim.js) =>
setImmediate(async () => await getNext())
to
return getNext()

[jorisw]

I think it will work if you just change this line (294 sim.js)

Could you try that, @mmdiego ? I can't get my simulator to fail (yet), with limit 10. And I never quite understood the reason for the setImmediate call there, so I left it in.

[mmdiego]

Yes, the change proposed seems to fix the problem. It works even with different values of the "limit" param.

[mmdiego]

I didn't have enough time to test the fix, I've only checked that at least simulations are working. However, I noticed simulations run extremely slow compared to the previous code using setImmediate(). It was also weird that I got slightly different results running the same simulation, so I'm not sure yet if it's the right fix.