Distance limit on Network Hubs?

[Tyrasuki]

Hi all,

I'm trying to do a little experiment where I run a simulated ISP for a village, but it seems after about 5 or 6 network hubs chained together, a computer cannot reach the primary router, nor other connected clients.

I'm not sure if this has to do with physical distance, or perhaps with a internal limit of the amount of hubs that can be daisy-chained?

Manually adding an ARP entry for v4 doesn't seem to work, and the ipv6 ULU (fe80::/64) ranges seem to be unreachable as well.
I've also set the TTL for all ping attempts to 128. since the documentation mentioned one hub counts as one hop.

Any clues? :)

I've attached some screenshots of the hubs and their layout as well as the connected devices.

TOPOLOGY AND RESULTS

[ajh123]

I'm also curious if there is a distance limit, becuase I'm geting similar problems with my ISP too.

I have a computer that is within a 40 block radius of my ISP router and it is not able to receive DHCP packets, but a computer that is next to the ISP router can.

[PatriikPlays]

The TTL seems to be decremented on connectors as well, I'm not sure if that is intended or not, but you can get your setup working by increasing admin.virtual_network.ethernetFrameTimeToLive in config/oc2-common.toml to account for this.
Assuming default TTL (12):

[PatriikPlays]

Oh, connectors can have 2 connections and act like a hub. I see why they decrement the TTL now, but the default TTL seems low.

[ajh123]

Thanks, gald someone could help.

[CoolCat467]

So in the end in summary, the solution is to raise the time to live value of packets instead of leaving it at its default value

[ajh123]

So in the end in summary, the solution is to raise the time to live value of packets instead of leaving it at its default value

Yes, since each network connector decreases the TTL, increasing the TTL will mean that the packet can travel further. So, using a TTL of 9999 (or more) should be enough for large networks.