Announcement: Preparing for Today’s Network Upgrade
Please follow these instructions to upgrade your nodes to the newest version to ensure the best performance and stability for everyone! It is important you do this for optimal earnings.
Then, press Ctrl + U, and hit Enter. This will upgrade your nodes. Upgrading can take several minutes for each node. Please don’t close the app or stop your nodes during this process.
Your nodes will now stop
Press Ctrl + S to start your nodes again
For CLI Tool Users:
If you’re using the CLI tool, please update and upgrade. Run the update first: safeup update
Then run the upgrade: safenode-manager upgrade --interval 60000
For ALL Users:
Please start your nodes gradually — especially if you plan on running multiple nodes.
Be conservative with CPU allocation to maintain stability
ever since safenode 0.112.5 (some days ago already, right?) there seems to be at least problems in displaying actual bandwidth. only inbound traffic is (much lower than before) outbound is all zeros. after a few restarts sometimes both values are shown as total zeros. there seems to be some traffic though but node and connection numbers much much lower than before with earlier versions.
gnome system monitor shows roughly 30megabytes of traffic both in and out and the traffic after the initual startup of a single node, dies kind of down totally.
there are no remaining remote ESTAB(lished) (tcp) connections active any more after this time it seems.
when stopping the node and starting it again it again creates this initial boost traffic then dies again traffic wise.
I’m confused by the versions. You list safenode as v0.112.5, and when I run safenode-manager status --details, it shows my nodes with “Version: 0.112.2”.
However the documentation seems to indicate that the “latest version” number is higher (even though the version I’m running is consistent with what you’ve posted):
Notice they have been upgraded to the latest version0.122.4
I don’t know if there’s a mistake in the documentation, or maybe there’s something I’m not understanding and safenode-manager is supposed to be an older version than Launchpad?
This is my first time trying to use safenode-manager, it was recommended to me because although I was new to using Launchpad I was having a problem with Launchpad determining my available disk space to be lower (and not allowing me to run nodes because it thought I wouldn’t have enough space).
Edit: I just ran safenode-manager upgrade and the version now shows my nodes with “Version: 0.112.5”, but this is still lower than what the documentation seems to indicate I should have. I wanted to try to look at Github to confirm the version number, but I’m not sure which repository to look at.
every few (tens?) of minutes there is a small burst in traffic for maybe a minute or less and again a few ESTAblished tcp connections of the “safenode” process again and a bit of traffic accumulated.
Yeah, I was not sure if that call was for your own sake, or for networks sake, or both. I’ve heard people’s machines jamming if starting too many nodes at a time. But that’s really the only option on Launchpad.
Yeah you’re right - for your own sake it does indeed make a lot of sense to start gradually I think … (e.g. I do start nodes one by one and check cpu usage, ram usage all the time to estimate if another node is ok to start or if I should wait for a bit longer to start the next node… But yes… A pity launchpad doesn’t do it this way as of now)
… Another parameter to watch would be bandwidth usage… I don’t do that because my servers are very well connected (and my home network is just too slow to run nodes…) … But I do see them maxing out the 1gbit connection when ramping up nodes…
When starting the nodes for the first time the gradual strarting seems really important, as they hog a lot of resources for a long time per node. But when restarting after the upgrade, they seem to start all at once, hogging 100% CPU for a short while, but no hiccups anyway.
To minimise the shunning of nodes it’s still worth staggering the startup. There is several minutes of intense downloads and CPU usage for each node being started.
For me (beefy CPU) its the bandwidth usage that needs time to settle down.
Also i noticed even with 24 cores the CPU jumps from less than 10% to 80% or more for a second or two when starting a node. I can only image the problems that might occur if people on more typical computers try and start 20 nodes at once.
This upgrade has removed the massive upload the system monitor on the PC was showing but not in router b/w monitor. Seems they fixed the issue there. So while there is the massive download b/w used while filling the record_store it seems to be much better behaved during startup. Also able to reduce the interval on the big PC
Four more now, and things in general are more evenly balanced between nodes.
I’m thinking about starting a few more, as there has been no problems either in my machine or on our LAN. But I think the network traffic in Autonomi is still very low, and that problems may arise if it gets more intense…
Announcement: Preparing for Today’s Network Upgrade 
