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.
Best way to manage and monitor 40+ servers?
I am curious what people will suggest regarding simplifying the management of a large amount of servers. Updating, modifying configuration, uploading new files...
In my case they will all run the exact same thing.
https://commando.io/ looks like the most simple solution but its way too expensive for so many servers.
The only alternative looks to be Ansible but I feel like its overkill for the basic management I want to do.
Also what about monitoring? I think Observium would work, but configuration is more complex than some other solutions.
Comments
Ansible for management. Might take a little to wrap your head around but it is incredibly powerful.
The above link is a good introduction into Ansible using a cluster of Raspberry Pis to demonstrate, but it's exactly the same for any server.
ansible , puppet etc.
I know ansible is good but I typically use puppet for my configurations on all my servers.
Nagios to monitor, it can de difficuktnto setyp but once its done its pretty powerful.
As for management ive never had to mansge 40+ at one time so cant hwlp there sorry.
Was the for a particular reason or just because that's the one you ended up using?
I was introduced to puppet first at the corporation where I was interning at. I also like and understand the syntax better and prefer their method of abstraction.
It could just be because it is what I have the most experience with / was introduced to first.
This is a very good topic, any one for recommendation on other software to manage dozen of server?
Personally, I recommend SaltStack for provision, but seems Ansible is more popular.
Depends how unique of each server. If they are pretty much the same, you can build a customized linux image, and setup PXE boot with KickStart.
I say it's a mistake to look at tools until you define your service needs. You say they'll all be running the same thing, but does that mean you want to manage them all from one central server, or does that mean you want them all to propagate any changes amongst themselves? Same goes for monitoring: they could all reasonably monitor each other.
Imo, [sensu](https://github.com/sensu/sensu] and ansible. Cockpit if you want to manage / monitor it without having to SSH each time, etc (only if you're running a RHEL-based system)
Excellent topic, very handy. I am just moving to the stage where I have VPS management issues as any more than 6/8 starts to be time consuming.
I wrote my own management tool, if anyone is interested, I'll open source it and throw it on github.
It really depends what your server specs are and what you're doing with it. A low end LES.com server might not be worth worrying about.
For monitoring, Zabbix