Experience of 0 devices out of box

[bilogic]

1. I installed using docker using https://docs.netalertx.com/DOCKER_COMPOSE/#baseline-docker-compose
2. I expect basic stuff to work out of the box, e.g. it should show clear errors/alerts or devices on my subnet
3. But I saw Plugin: ARPSCAN in red on top right corner, clicking it brings me to logs, and nothing specific or clear
4. After searching, I see Need help, network discovery across multiple subnets #1342
5. So, I went into docker's bash nmap -v -sn 192.168.1.0/24 runs successfully, i.e. shows IP addresses that are used
6. arp-scan 192.168.1.0/24 also shows a list of devices
7. Bottom line, the experience isn't great... what else am I missing?

On further searching, I see https://www.reddit.com/r/selfhosted/comments/1ol9qzf/netalertx_alternatives/

1. When docker/docker compose is recommended, it sets my expectation that things are just going to work
2. So perhaps "slapped together" in a clearer sense means, it doesn't
3. In a typical basic setup, I would think there is only 1 interface that holds the IP address 192.168.1.0/24, so it should be possible to figure out the interface and subnet to run nmap and arp-scan against
4. I do understand this can come across as very critical, but my aim is to improve the experience for first time users of this project

[jokob-sk]

Hey,

It does sound way too critical for something I manage in my free time, so I'd appreciate a more positive tone, really, don't want to help out if you go and dig out negative comments. Especially network- discovery tools and docker are not a match made in heaven.

Happy to improve with concrete examples what to improve.

That out of the way, please read: https://docs.netalertx.com/SUBNETS/?h=subne

If you still can't get it to work (which heavily depends on your network setup from my experience), please open a Setup issue and provide necessary logs, instead of digging through reddit comments and copy pasting negative statements.

Once this is sorted, happy to take your feedback on board on how we can auto detect more complex setups, in my network, out of the box, all devices are discovered within 5 minutes.

Looking forward to your contributors on how to make the on-boarding experience better, for users like you and others.

[bilogic]

Thanks, I thought it is easier to be upfront and direct instead of beating around the bush, nonetheless I appreciate your effort.

tried to run cron jobs

run every 5 minutes

1. I created this discussion after exploring for about 30 minutes
2. And spent up to 1 hour trying to get devices to show
3. It is now about ~3 hours in, finally devices are showing (I have not changed any configuration in the UI)
4. The only thing I did was try to run cron jobs
5. As they are configured to run every 5 minutes, it is puzzling why nothing showed up in the initial hours
6. I could teardown everything and restart again to figure out why, but in my situation where everything (arp-scan/nmap) was OK and there is only 1 eth0 (ignoring all loopbacks/virtual interfaces), is it unrealistic to expect devices to show after 15 minutes of a first time setup?
7. Or did you mean I must add the SCAN_SUBNETS = ['192.168.1.0/24 --interface=eth0'] in the UI?

[bilogic]

I have tore down everything and started a fresh container right now, let's see when devices start to appear

Also want to mention that I had to strip out these 3 lines in the docker-compose.yaml, it was giving some error Error response from daemon: unknown log opt 'max-size' for journald log driver

    logging:
      options:
        max-size: "10m"         # Rotate log files after they reach 10MB
        max-file: "3"           # Keep a maximum of 3 log files

[bilogic]

I'm seeing the devices now, for some reason, it takes about 30 minutes before devices show

[jokob-sk]

Glad to hear it's working. Let me know what you think is necessary to make the on boarding easier.

Depending on the network size a scan can take substantially longer wwhich also depends on how many plugins are enabled.

Without seeing logs I cant diagnose further (that's why I asked you to submit an issue as there is a template to collect crutial information to diagnose and help, but alas here we are).

Unfortunately your notes a lre all over the place so not sure what you tried and what is your config.

If you want to provide structured feedback, please open an issue so I have all necessary context.

Closing as resolved.

[bilogic]

1. I'm referring to an "out of box" setup, I basically made 0 changes and did not enable any additional plugins
2. It is a small /24 network (255 devices), arp-scan and nmap on the CLI completes in under 10 seconds

Ending arp-scan 1.9.7: 256 hosts scanned in 1.929 seconds (132.71 hosts/sec). 16 responded
Nmap done: 256 IP addresses (13 hosts up) scanned in 2.24 seconds

3. I have a few ideas off the top of my head, but without understanding this project better, those ideas can come across as irresponsible

[jokob-sk]

As mentioned, please open an issue.