Appreciate the rapid response!
As you can see, no QR Code. I tried resetting the 2FA in admin account settings but that didn’t help.
If I can help with the map icons and port disabling I’ll happily do so; I'll DM you.
The searchable fields request is linked with a desire to integrate with our CRM to help non-tech staff determine customer outages etc.
Loving your work so far by the way.
Amm0 corebit install script.
RoS version 7.22rc2
Health Check fails.
failed with exit code 1, tries 75/3, output: Connecting to 127.0.0.1:3000 (127.0.0.1:3000)
wget: server returned error: HTTP/1.1 404 Not Found
CMD-SHELL
wget --no-verbose --tries=1 --spider http://127.0.0.1:3000/api/health || exit 1
Hopefully there's a fix soon.
Thank you
I have posted an update to Docker now, that should fix it. Can you test?
working!
good, output: Connecting to 127.0.0.1:3000 (127.0.0.1:3000)
remote file exists
Am interested in this but would like to know how links to devices are discovered and how they are handled with topology changes?
My gold standard for a network visualiser (which I have yet to find) is one that can automatically find all OSPF neighbours and how they connect, automatically map them out, place them on a geographical map by GPS coordinates automatically, display ALL links between neighbours (that includes multiple OSPF adjacencies i.e. backup links), show bandwidth on all of those, and then importantly it shows when an OSPF adjacency goes down and isn't dependent on the interface state (because ethernet can be up but there is a break in the link).
All of it must be dynamic so that if the topology changes it updates by itself, I'm not just giving myself another job with a network monitor that requires constant adjustment
Does this application utilise OSPF states at all? If it's just using CDN how does it handle link changes. Will it discover and show multiple links between neighbours? And if one goes down and the neighbour disappears from i.e. a VLAN interface, will it show as down on the map or does it simply remove the link?
Thanks for outlining your expectations — that helps a lot.
What you’re describing is essentially an enterprise-grade routing visualisation platform: full OSPF adjacency modelling, multiple parallel neighbour relationships, routing-aware topology reconciliation, automatic geographic placement, dynamic path validation, and continuous state tracking independent of interface status.
That class of functionality exists — but it typically lives in the network assurance / digital twin segment (e.g. IP Fabric, NetBrain, Forward Networks). Those platforms are priced per device and per year, and even relatively small networks can end up costing significantly more annually than CoreBit’s entire license.
CoreBit is intentionally positioned in a different segment.
It is designed as a practical topology mapper and monitoring platform for SMB and mid-size environments, not as a full routing control-plane modelling engine.
Today:
Topology discovery is primarily based on L2/MAC forwarding data and interface relationships.
It focuses on physical and logical connectivity visibility.
It does not currently parse or visualise OSPF neighbour states as topology truth.
It does not attempt to build a routing “digital twin” or perform full path simulation.
The goal is to provide dynamic, low-maintenance topology visibility without the complexity, operational overhead, and cost of enterprise assurance systems.
That said, I am exploring the possibility of adding a routing overlay layer (OSPF/BGP session awareness) as an optional feature — starting with neighbour health visibility rather than full LSDB modelling.
If your requirement is complete routing-aware automatic topology modelling at enterprise scale, CoreBit is not trying to compete directly in that category, at least not today.
Let me know a bit more about your network size and design — I’m happy to clarify what would be realistic within this segment.
Understandable and I hear you. Unfortunately even those incredibly expensive paid tools don’t really do the job properly. And i’m not actually interested in a full routing topology overview, just simple point-to-point link states
Let me explain with a practical (and very real world for me) example of a simple yet scalable topology involving just 2 routers. And explaining why most tools utterly fail at this and where it could be done much better
Lets start with RouterA and RouterB which connect together - but not directly. In between them are a pair of LHG 60G radios connected to Ether1. These are invisible to both routers at Layer3+, but are there at Layer1 and Layer2. Now having a program that can map out a Layer2 network and show those devices being present is very useful, but i’m mostly interested in network topologies at Layer1 and Layer4
The problem we discover is the link goes down in rain, but of course none of the Ethernet links do, they are in the Up and Running state. It’s the 60ghz wireless interface (invisible to both RouterA and RouterB) that goes down. More crucially they can actually be flapping, or it might even be Up but its so bad that effectively no traffic is going across the link. Here lies the biggest problem with most network monitoring software - they concern themselves with the state of the Interface, not whether or not its actually passing traffic or a viable path. Ergo they don’t report this link as being down or degraded in any way since as far as RouterA/B are concerned, they have an active Ethernet interface so happy days
This comes back to ideally monitoring the OSPF/BGP neighbors. It’s not the only way and a cheaper/easier implementation would be to use CDP/LLDP/MNDP and see if a neighbor merely exists on an interface. It’s not as fast or accurate but if there is an interruption to data flow then a neighbor will eventually timeout and disappear
Now expanding a bit more, we got sick of the weather killing the link so we install a 5ghz backup radio between RouterA and RouterB on Ether2 that never goes down in the rain but its nowhere near as fast. Herein lies the 2nd problem with most software. They either don’t acknowledge direct backup links, or they simply can’t draw it. What i’d like to see on a map is 2 discrete lines between RouterA and RouterB, each corresponding to their relevant interfaces. And some people might think well does it really matter as long as they are still technically up and working? Yes it does because we have our 3rd development in how this network is evolving:
An obstruction happens (building in the way, tree grows, mast falls over, whatever), meaning the 60ghz radio is now permanently interrupted and does not link, yet the 5ghz is still able to. We have a permanently degraded connection between routers operating at ~150mbps 24/7 instead of operating at ~1gbps 99% of the time. Nobody has been aware of it because the link is still ‘Up’ due to the 5ghz maintaining an uninterrupted connection
So what i’d like to see is a program that can dynamically build and show adjacencies between routers and display all of the links between them (in this case Ether1 and Ether2). I’d be happy with a piece of software that will automatically display a link if a neighbor ever pops on an interface, and if it disappears it still remembers that a neighbor ‘should’ be there on that interface but isn’t. Thus draws the link as red/broken rather than automatically removing it, so its a very clear visual indicator that something is wrong. Allowing me to then go and manually delete that link if it was either drawn in error, interfaces have changed or the link has been decommissioned
In an ideal world i’d like to see an industry supported extension to OSPF/IS-IS that can utilize transit metrics and not just interface state (i.e. bandwidth, latency, packet loss) and it can steer traffic dynamically through an entire network in the best possible way rather than just relying on fixed costs. Until such a pipe dream ever happens i’m still reliant on manually intervening and having to periodically look over the network and make sure we don’t have any heavily degraded or failed links. Anything that can make that job easier is a very welcome addition to the tools arsenal
Hi Millenium7
Without wanting to steal Claus’ thunder for Corebit which as he says is positioned differently, have you tried NetXMS to achieve what you want?
It has powerful NXSL scripting language and agents that can execute commands interpret the output and then use that output to set the status of your device or link or whatever.
NetXMS natively does discovery as you want in your post, but the scripting opportnity goes much further.
Or you could replace the LHG60’s with a pair of Ubiquiti Wave LR or Wave Pro’s that have more gain (and therefore rain resiliency) and built-in 5GHz backup and Ethernet carrier-drop in case bandwidth goes below a certain threshold. I hope MikroTIk has carrier-drop planned for their new Peraso-based radio announced at MWC. It appears to have everything else.
In CoreBit I have mapped out the connections between VLAN interfaces on routers (one VLAN per radio link) so that I can monitor throughput at least on those connections, and they can be drawn parallel to each other on the map so you can see what’s going on with regards to bandwidth utilization.
I haven’t played with it for a few weeks (been pulled in other directions), but what I think would be helpful in this scenario is a way to have the system throw an alarm if a particular link’s utilization goes below a certain threshold, or above a certain threshold, and you could set times of day for those thresholds (like 4-6 timeslots). For example, if the LHG60’s are normally pushing 500-700Mbps during peak hours (4pm-10pm), and you see “0-100”, it should go into alarm. Or if the backup sees 500K most of the time, but suddenly spikes to 150Mbps, it should go into alarm.
(Honestly, it could simply start with primary links should always see traffic, say >2Mbps, and backup links should see hardly any, like <500K.)
I do run NetXMS and it has its purposes but its not good for visual maps and is extremely script heavy
I’m looking at CoreBit not as a full monitoring platform replacement but a great visualizer to run in addition, as I want something that can automatically draw readable and nicely presented maps
Changing out radio’s doesn’t change the key principle i’m trying to convey, which is that there is a viable need for more than a single link between 2 routers and having them drawn and alerted when something happens that impacts the network. Having all adjacencies shown on a map and alerted if there is interruption is a huge value add
The alerting thresholds are 1 way but I can think of a few drawbacks. Some people use multiple VLAN interfaces and bonding to get unequal cost load balancing, i.e. 5x VLAN interfaces on linkA and 2x on linkB so both actually get utilized in appropriate proportions. Traffic Engineering can also be used to selectively offload certain types of traffic through alternate paths so the usage pattern can be highly varied, i’m patiently waiting for MikroTik to implement Segment Routing or MPLS EVPN with FRR to make this viable and scalable
I think the simplest implementation without any kind of custom scripting or manual work defining parameters is to use the neighbors table. if RouterB exists on ether1 and ether2 then draw both links. If RouterB suddenly disappears from ether1 then the link remains but shows as broken
That's fine. You can still set a threshold per VLAN. If any VLAN isn’t pulling its weight, you want to know.
Yeah, there’d have to be some kind of “remembering” mechanism, and a way to “forget” links that no longer exist.
Latest addition is Rack elevation management, and device documentation in ANSI/TIA-606-B compliant documentation format.
Racks can be placed on canvas like devices, and is synced with devices in same map, so clicking a device will highlight it in the rack.
You can add images of the racks, and zoom on areas for details.
New device documentation window, joins all device information, and adds extra device Documentation in forms of Images and custom text.
What is missing?
Please come with suggestions to next feature.
