New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Good tool to manage all these servers....?
So in a sense we all made provisions in the Easter thread, but was wondering if anyone had a good suggestion of a software/dashboard to keep tabs on all our precious vps/servers?
I know I could use each provider's but mostly look for something more central, provider agnostic...
Comments
Me too. But most of the users buy and just let it idling
Do you mean...
(a) monitoring
(b) job control (e.g., RunDeck)
(c) configuration management (e.g., ansible)
?
Like hosting your own monitoring? Monit, Munin, etc, etc, etc..
Thinking monitoring ... monit looks kind of cool
I guess he is searching for monitoring and management. Like a complete dashboard for everything. Thats what im looking for
Yeah,
The best analogy I can come up with right now is something of a webui where I can keep track of tasks and possibly do a few basic tasks... but mostly see which provider disappeared overnight
You're already here, though..
Add all of them to Newrelic , if a server goes down , it gets hidden in the control panel. Could get tedious if you have a lot of servers though.
PRTG?
Observium
The developer is an asshole. Use LibreNMS (a fork of Observium) instead.
Start with Ansible, then everything else on top.
Redhat should be a good steward(ansible current owner/maintainer) . Python3 path is guaranteed I think.
I never quite understood why people have so many VPS servers, I mean, I appreciate things scale up, but when it gets to this point and you need advice on managing them, you went too far
For the sake of management at a certain scale you are so much better off getting a couple of dedi's and installing proxmox or vmware and then you have central management and flexibility.
Personally I think if you have 8+ your at that stage.
I just have a handful of cheap ones in critical IXs near to me for my browsing traffic.
Singapore, Pune, Amsterdam, London to switch VPN between, whenever my state-run ISP's routes get certifiably STUPID.
But yeah, all web and self-hosted are now consolidated on two cheap dediboxes in Proxmox cluster + couple failover_IPs.
+Holding on to a dediserve resource-pool and a couple of InceptionHosting BF KVMs for aux serviceVMs.
The logical final stage of LET user evolution is close.
didn't know that, care to explain? will have a look into librenms then ;-)
Ansible with a UI, Monitoring LibreNMS is pretty good.
Offtopic but who do you use in Pune, LeapSwitch?
Quadhost openvz in Pune(LeapSwitch Network).
I waited for a Yearly-special /coupon-code.
There's i-83 if you can manage with shared IPV4.
Get at least 2 of those, to make sure if something dies at one side, it does not affect your Monitoring.
Get internal Monitoring aka resource Monitoring like LibreNMS.
Keep track of each box, when it expires, login details..
Try to automate Updates, Like a tool which can deploy with a single click updates on all of these boxes. DO NOT run that process complete automated, it may break your system.
Its not witchcraft.
Take a look at ISPConfig. Open Source and it will handle 1 to 100's of linux VPS's or dedicated servers all thru a single control panel. Makes it easy to stay on top of things.
You are right - if all of them are in geo proximity. But some are expressly looking to have different providers and different locations.
Btw. I didn't say that (under certain limits) you are right for politeness; I mean it and I live it. I have a dedi with some VMs on it. But I also want/need different providers/locations for quite some things.
mostly https://asyd.eu/ but also Jenkins currently (not so nice to use tho).
runcloud does what I need
Figuring things out is part of the fun, isn't it? We all have to start somewhere.
Are people better off? I mean, one bad patch/command and the box is hosed. Most providers don't have on demand console access for dedis, and they also don't have shared storage for HA failover. Shared storage can be fixed, but paying people to provide the infrastructure is sort of the point of VPSes.
Dedis are more flexible, but having dedis also doesn't fix the central problem of managing and monitoring the servers and applications. They're just one more thing.
You're not wrong, but there are reasons for sticking with VPSes.
One of the more interesting things is hybrid setups. VMs with dedicated resources, but they're still able to take advantage of being a VM.
True, I was coming to it from the perspective of managing the VPS inventory rather than the services they run, for example proxmox or esx you have all your virtual servers in 1 GUI then.
Everyone has different levels of need though I guess, for some many VM's with many hosts with many logins, many invoices and many VPS management portals is preferable to having 1 or maybe 2 dedicated servers with 1 login and 1 web UI.