Node shows offline periodically

so far as i can tell this last part of the log looks fine…

when it successfully uploads and downloads it’s most likely fine…

it kinda makes me wonder if it’s the dashboard that shows it wrong… but i’m not aware of that being something it does…

a couple of days ago there was a case where somebody was running two nodes on the same storagenode identity, which created a similar behavior of the nodes switching between being online and offline.

i didn’t suggest that because i got the sense that you only had one node and so assumed it was most likely another power management related thing, which is a bit more common for new nodes…

is this your second node?

1 Like

I only have one node now. I did create a node on another pc about 2 weeks ago and it got banned for reasons unknown to me. This current node is on a dedicated pc where I generated a new id although I am using the same email address.
Also the node went back online after a couple of hours and remains online now.

lets hope one of the potential fixes you have implemented have dealt with the issue, since it’s not a potential node killing issue, i would just wait and see if it happens again, before spending more time on it.

using the same email is fine

1 Like

When you see this, and refresh your page, does it fix the problem? Last contact of 2h 28 mins in your screen shot makes me think that you have left the page open. When you leave the dashboard page open, the last contact starts increasing. It won’t go back to zero until you refresh the page. After a certain amount of time, this increasing last contact makes the node status offline. I really don’t think there is anything wrong with your node. Try not leaving the dashboard page open, and only view it when you want to check on things.

Looking at the code on github, the page is designed to show status “Offline” if the last pinged time is more that 120 minutes. So anytime you see your node is status offline, if the last contact is greater than 120 mins (2 hours), this is (probably) not a true offline, and is just an artifact of how the page is coded.

—edit— See update below

1 Like

really? i haven’t seen mine go offline in the dashboard because of last contact time out…
think i’ve been pretty far up tho… maybe 48-72 hours

infact i actually think that it would change to offline if it losses the connection to the node maybe… so yeah maybe if the computer one is using goes offline or switches between networks, the page may loose connection or timeout and then the 2hour mark comes into effect…

sounds interesting… haven’t seen the effect of this, 2 hour limit… but this machine is also online when it’s booted, so maybe thats it… maybe its a browser thing.

i just know i haven’t seen the 2hour limit

good suggestion tho

1 Like

To be fair, I actually just tested this. I opened the page and waited 2 hours. It still showed online, but I have seen this before. Then I remembered that it’s when I close the browser with the page open, and then have the browser automatically restore open tabs on start. I think this causes the page to be loaded from cache and not actually refresh the information from the node. Then the page shows > 2 hours last contact and node offline. So this may not be as much of an issue as I thought.

1 Like

Rather embarrassed but I have discovered the root of my problem and want to share it just on the off chance someone else may experience this in the future. Using a site called uptimerobot I found that periodically it said the port was closed and when I looked it appeared to still be open. Looking more closely I saw that my Noip DDNS address had been updated with a new IP that didn’t match my actual IP so that when the DNS name was used to look for the open port it showed up as closed. I corrected the Noip entry and everything was OK for awhile. Again a few hours later uptimerobot showed a problem and again the Noip IP address had been updated incorrectly. I reasoned that there must be another Noip update client installed somewhere with my credentials that is periodically updating my DDNS entry with the wrong IP address. I changed my Noip logon password online and in my update client and haven’t had the problem since and the Node now shows online all the time. I have no idea where it is located or how the rogue Noip update client got ahold of my logon credentials but somehow it did and now it can no longer make any bogus changes to my DDNS entry.
Thanks for all of your quick and valuable responses since I made lots of useful changes to my Node that may have eventually given me heartache.

3 Likes