Konto erstellenAnmelden

Frühere Suchanfragen

Keine früheren Suchanfragen

Suchoptionen

Nur verfügbar, wenn angemeldet.
nrw.social ist einer von vielen unabhängigen Mastodon-Servern, mit dem du dich im Fediverse beteiligen kannst.
Wir sind eine freundliche Mastodon Instanz aus Nordrhein-Westfalen. Ob NRW'ler oder NRW-Sympathifanten, jeder ist hier willkommen.

Verwaltet von:

Serverstatistik:

2,8 Tsd.
aktive Profile

nrw.social: ÜberStatusProfilverzeichnisImpressum/Datenschutz

Mastodon: ÜberApp herunterladenTastenkombinationenQuellcode anzeigenv4.3.7

#netdata

0 Beiträge0 Beteiligte0 Beiträge heute
Fortgeführter Thread
Öffentlich
immibis
I shut down #netdata because I didn't like the push to make it cloud, the fact it always tries to load UI from their servers first (not mine) with no way to turn that off, and the relative lack of customizability. If I want to see the CPU on all nodes there seems to be no way to do that apart from adding all the individual CPU charts to a custom dashboard. And I couldn't find a way to force a graph to start at 0. It seems a bit hard to collect custom metrics too.

Things I do like are how well it works out of the box and collects a lot of metrics I didn't know existed.
Öffentlich
Mika

#Netdata has saved my butt - I'd only set it up to get temp monitoring for my #homelab, but today I woke up with an alert notifying me that one of my #Proxmox node's #ZFS pool has degraded.

Opening up my Proxmox web interface, checking the main "Summary" and "Disks" tab, nothing would suggest that anything is out of the ordinary - even the failed disk is showing that it had passed its #SMART test. It is only when I dug deeper into the "ZFS" section that I could see that one of my #SiliconPower NVME disk (avoid at all cost btw) has failed and the ZFS pool has degraded.

Fortunately I've an #ADATA SX8200 Pro NVME lying around that I will use as the replacement disk, I've never done this outside of #TrueNAS's web UI, so I'll have to do so while referring to a guide. I'll also link to my guide on setting up Netdata, and will probably write up a guide on how to replace a failed disk in a ZFS pool on my repo after I've done so myself.

🔗 https://github.com/irfanhakim-as/homelab-wiki/blob/master/topics/proxmox.md#monitoring

▶️ https://youtu.be/IQA7aTezrVE

Öffentlich
Aires

Can anyone recommend self-hosted observability alternatives to #Netdata? I really liked the UI, automatic alerts, and how easy it was to set up, but no way am I connecting it to the cloud just so I can see processes and logs. Should I just give in and learn #Grafana? This is for a 3-node homelab, so I don't need anything enterprise-y.

#Linux#Observability#Homelab
Öffentlich
Lanie Carmelo

Hi #SelfHosted community. I've figured out a lot of my setup. I now have a new domain, laniesplace.us, just for #HomeServer stuff. It's set up through Porkbun with Dynu for #DDNS. I've now got #Traefik, #TailscaleVPN, #Linkding, #Forgejo, #Dokuwiki, Code-Server, #Portainer, #Netdata, #Watchtower, #Cockpit, #Pihole, #MiniFlux, #TheLounge, #Filebrowser, #UptimeKuma, and the #Homer dashboard service installed. I'm now trying to set up #Authelia so I can have single sign-on to my services. For some, it's working now, but I can't seem to get Linkding to work no matter what I do. This is on a #RaspberryPi 500 with 8 GB RAM and a 512 GB SD card, running #Stormux, which is based on #ArchlinuxARM. Can anyone help? I'll reply to this post with all my relevant config files in separate posts. What's happening is this: Linkding is supposed to be available at bookmarks.laniesplace.us. When I go there, I see a 401 unauthorized error and a link to sign into Authelia. Once I sign in, though, it redirects back to the page with the 401 error. I've been trying to figure this out for hours with no luck. Files will be in replies to this post.
#SelfHosting #Linux #HomeLab #RPi #RaspberryPi500 #RPi500
@selfhost @selfhosting @selfhosted @linux

Öffentlich
Mika

I've implemented a Parent-Child #Netdata setup for my #Proxmox based #homelab. I don't get how Netdata get its data for the #GoSensors section (which I can only assume is what is documented as #Linux Sensors on their docs, cos they don't really tell)? I don't have lm-sensors installed on any of my nodes, yet it reports some temps under GoSensors. Problem is, one node has more metrics to report (under GoSensors) than the other, significantly - and none of them have temps of what I want like my #AMD CPUs (1700, 5600G).

I'm not even sure anymore if Netdata even requires/supports metrics coming from lm-sensors cos it's not mentioned in their docs at all, only seen it being mentioned by their team on their forums. Installing it and after detecting some new sensors, adding them to /etc/modules and restarting the kmod service, also restarting the Netdata service, no new (Sensors) section or (temp) metrics appear on the dashboard.

Netdata's documentations are insanely poor. It fools you into thinking it's good with just how many things are documented, but a lot of them are outdated or straight up inaccurate, and there are a lot of missing holes in their docs in general. Also, for wtv reason, you'll find A LOT of dead links thrown out by their team or their users when you scour through their forums, #Reddit, etc. Do they change their URL every month or something? ​:blobfoxcat:​

🔗 https://learn.netdata.cloud/docs/collecting-metrics/hardware-devices-and-sensors/linux-sensors

learn.netdata.cloudLinux Sensors | Learn NetdataPlugin: go.d.plugin
EntdeckenLive-Feeds
Über