"Harvester did not respond" log warning?

I am running a full node on one windows 10 machine with the gui and a local harvester on a separate w10 machine using the CLI. I updated both to version 1.2.0 yesterday and I am seeing plots passed filter from both machines in the gui and the harvester logs show times under a half second for found proofs. The problem is, I keep getting this warning in my logs on my full node machine
"2021-07-09T18:59:31.464 farmer chia.farmer.farmer : ERROR Harvester did not respond. You might need to update harvester to the latest version"
Any idea what that means and should I be concerned? I googled it but could only fine people asking what it was and no real answers. Any help would be greatly appreciated. Thanks.

1 Like

It will probably take a while till the problem is assessed being such a new version.

I’ve seen others report lots of errors since updating to 1.2.0

1 Like

Yeah maybe it is a new warning introduced with the current version. I also keep getting “respond plots came too late” but I can’t find much on it either.

1 Like

same issue here.
Since I got over 1000 plots, one harvester with 445 plots is getting this warning and going offline.

Update to at least 1.2.1, it might be this issue:

  • Fix harvester cache updates. Prior to this commit the farmer called the request_plots every second for each harvester as long as they failed to respond properly. Since the rate limit was 10/minute this lead to hitting the rate limit if the harvester didn’t responds for 10 tries in a row for whatever reason. This commit changes the behavior to always keep track of request attempts even if they end up in a timeout to really only re-try every 60s no matter what.

I’ll try that, thanks. I’m still trying to confirm but I was thinking it’s from me deleting my old plots while the gui is running so that I can replace them with portable ones.