Just an update on /SWAP activating automatically while only 4 nodes are running on this dev Notebook (Acer Aspire i5 10th gen 8 core 16 thread, 8GB RAM, 256GB SATA SSD with 90GB free storage capacity).
I noticed on the Ubuntu 24.04.2 LTS build that /swap activated itself overnight, when there was none in use when I checked at 12:00am midnight EDT last night before going to bed…
Anyone have any insights, why /SWAP all of a sudden woke up to service this small 4 node configuration?
Nothing else was running…, although I need to check syslog to see if there was any backdoor local weirdness going on (I don’t think so, although maybe tre was system invoked EXT4 defrag going on or somehting like that…)
this is an edge condition that will affect earnings negatively,
If the root cause of /swap being turned on is indeed the OS File/System ie ‘defrag’ or similar OS process hi prioritized process, then the antnode processes are at the mercy of the OS resource manager/scheduler dynamic duo, unless the antnode asks for higher priority during quote asks…
If the OS is not the root cause of /swap activating, then what might it be?
Can anyone quickly rule out the antnode code no using or not invoking /swap?
I’m miffed at the moment why /swap magically turned on over night, there was lots of memory and the 4 nodes on average were using 8% CPU resource steadily , with very little block copy and write activity (log updates mainly at 1.4kbit to 2.7kbit periodically, maybe every 5 sec with this 4 antnode set up, one such log update write)
Just to clarify that to solve the issue you’ll need to choose not to use UPnP when you create nodes, in Formicaio you will see this checkbox when creating nodes, you seem to need to uncheck it.
“Try to use UPnP to open a port in the home router and allow incoming connections.If your router does not support UPnP, your node/s may struggle to connect to any peers. In this situation, create new node/s with UPnP disabled.”
but I did as you said and started from scratch with a new dir and 12 new nodes - so far the nodes seem to be working. The modem-router still has UPnP checked - so the quote says I SHOULD use UPnP but it doesn’t work when I do that?!
Maybe your internet provider is using CGNAT, which means UPnP won’t work. If that’s the case, you can ask your internet provider for a public ip, which allow you to use UPnP. Public ip does not need to be static, dynamic is enought.
Some providers give a public ip for free just if you ask, but that might not be with all providers.
I’m trying to convert from Launchpad to Formicaio on a Linux host. When I first launched it, the app failed because it couldn’t find the ‘migrations’ folder. So, I created an empty one for it. Now, I’m getting the following errors:
Sqlite query error on settings: error returned from database: (code: 1) no such table: settings. We'll be using defaults.
Nodes auto-upgrading setting enabled?: false
Sqlite query error: error returned from database: (code: 1) no such table: nodes
[2025-03-24 15:01:48.598422499 UTC] No active nodes to retrieve metrics from...
When you unzip the released .zip file you will see it contains the migrations scripts/files and the files required for the web frontend, just unzip it and run the app from within the unziped folder, look at this instructions.
I guess I found the reason for formicaio sometimes reporting “unable to fetch info”. It’s because some outgoing tcp connection has reserved the metrics port when the node has been created.
Last time it happened, I checked all tcp connections with “netstat -a -t -n”. Found out that there was one outgoing connection from the same port which I chose for the metrics port for the new node. I assume it had been used with formicaio for fetching metrics from some other node. When creating an outgoing connection, the port is randomly selected.
By looking at the screenshot it looks like you are running on Windows and were installing Podman…? if so, you could instead download the released .zip package for Windows and simply run the formicaio.exe without the need of Podman, if you preferred it.
The option to add is --no-upnp by itself if starting antnode directly. Do not follow with true or anything. Its the args processing library they use sets an option to true when its specified
EDIT: antctl uses --no-upnp as well, so it must be something in Formicaio that is needed.
But looking at the teams replies it seems that there is a bug actually in antctl where it does not pass the --no-upnp to the antnode as it starts it
So if using antctl then you cannot fix it.
Thus go back to version 0.3.11 of antnode.
Maybe Formicaio needs an option to set the version to use for antctl and antnode rather than just always upgrading to latest which is what antctl does automatically
Formicaio version 0.5.0 has been released and it’s ready for installation across all supported platforms in native mode.
Docker containers built with this new version, including for UmbrelOS, will be were also published in the next few hours.
This release is meant to introduce changes required to make it compatible with latest antnode v0.4.0. Note you cannot run previous/older versions of antnode with this new Formicaio version, therefore make sure to upgrade any existing nodes instances to antnode v0.4.0.
I had a hard time getting my nodes up and running, but got the sequence figured out, empirically . They would not start and run stably until I started each individually, and then watched till the upgrade arrow came up. If left at that point, they would close after a few seconds. But if I clicked on the upgrade arrow right away, the node would restart and run stably. Had to go through this on each node, but that’s been 10+ hours and they’re still doing fine. Hope this helps.
