Uptime... who has the longest....

[itNGO]

Just to make it searchable and to have this question documented for others who might ask.

How long can Proxmox VE run nonstop compared to others like VMWa** or Hyper*

600days is nothing special for PVE. Except without Kernel-Live-Patching add on, you are sticked to an older Kernel.

 

[Nollimox]

Just to make it searchable and to have this question documented for others who might ask.

How long can Proxmox VE run nonstop compared to others like VMWa** or Hyper*

600days is nothing special for PVE. Except without Kernel-Live-Patching add on, you are sticked to an older Kernel.

View attachment 53340

That's without updating or upgrading?

 

[itNGO]

That's without updating or upgrading?

We update regulary. Currently it is on 7.4-16. Was original installed with 7.0-4. But we do not reboot so the kernel is still the one, which came with 7.0-4 more or less. It is our Office PVE-Server where we just have Firewall, Fileserver and ERP and some other stuff. Backed by UPS and not reachable from anywhere. It is some sort of scientific test.. how long will it last.... ;-.)

 

[xqlusive]

for others who might ask

why would anyone ask this except you maybe? -.- it will run until the lights go down, something needs replacement, patches to be applied, .. right?

 

[Ramalama]

Just to make it searchable and to have this question documented for others who might ask.

How long can Proxmox VE run nonstop compared to others like VMWa** or Hyper*

600days is nothing special for PVE. Except without Kernel-Live-Patching add on, you are sticked to an older Kernel.

View attachment 53340

You can try to update to pve8 without rebooting now xD

Tbh, i don't think anyone is patient enough of the active members in the forum, to not update his server, including Kernel.

Otherwise you need to find someone that forgot that he has an Proxmox server running somewhere, like setupped and forgot

 

[Dunuin]

Sometimes I see people here asking for support running their PVE 4.X or PVE 5.X. Wouldn't wonder if they could beat the 599 days

But yes, for a security perspective, long uptimes aren't really good thing without live kernel updates.

 

[itNGO]

why would anyone ask this except you maybe? -.- it will run until the lights go down, something needs replacement, patches to be applied, .. right?

Compared to other Virtualization-Products where you simply can not have this expectation.... But people ask us, how stable is it many times.... Just wanted to get some other Uptime-Results to prove or compare.... We for ourself already now... it "can" last forever if you want and have done it right.... Cause Proxmox is until now not a "Global Player" like others are or seem to be...... ;-)

 

[itNGO]

You can try to update to pve8 without rebooting now xD

Tbh, i don't think anyone is patient enough of the active members in the forum, to not update his server, including Kernel.

Otherwise you need to find someone that forgot that he has an Proxmox server running somewhere, like setupped and forgot

It is wisely isolated and protected. As I said, its also a scientific "Sever"...

 

[xqlusive]

Compared to other Virtualization-Products where you simply can not have this expectation.... But people ask us, how stable is it many times.... Just wanted to get some other Uptime-Results to prove or compare.... We for ourself already now... it "can" last forever if you want and have done it right.... Cause Proxmox is until now not a "Global Player" like others are or seem to be...... ;-)

you and the people that ask you are asking strange questions really.. I would say: it is a server virtualization system expected to be running 24/7. the more you get used to it, the more you believe in it

 

[bbgeek17]

Compared to other Virtualization-Products where you simply can not have this expectation....

We've had ESX running for years, until datacenter lost power. We also have Blockbridge storage systems in the field that look like these:

00:58:43 up 790 days, 23:35, 1 user, load average: 16.94, 16.75, 16.68

01:00:25 up 833 days, 5:17, 1 user, load average: 22.71, 22.51, 22.51

01:00:58 up 947 days, 3:45, 1 user, load average: 45.74, 45.07, 45.06

21:06:51 up 1224 days, 4:06, 1 user, load average: 7.10, 7.06, 7.07

This is just a random sample. These systems had software upgrades, but didnt need any Kernel updates, which we are very judicious about.
Any good enterprise system, Proxmox included of course, can achieve impressive uptime if it was installed properly and does its job.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[Neobin]

Sometimes I see people here asking for support running their PVE 4.X or PVE 5.X. Wouldn't wonder if they could beat the 599 days

Recent example:

https://forum.proxmox.com/threads/permission-denied-error-for-vzdump-command.130885/post-574612

 

[Dunuin]

Recent example:

https://forum.proxmox.com/threads/permission-denied-error-for-vzdump-command.130885/post-574612

When looking at the swap and KSM I would bet the memory is pretty fragmented after 4 years of uptime.^^

 

[Darkk]

Pretty cool to see that kind of up time but patches and security updates are more important to me than up time. Plus rebooting the servers is a good way to reset the RAM especially with ZFS.

 

[itNGO]

You can try to update to pve8 without rebooting now xD

Tbh, i don't think anyone is patient enough of the active members in the forum, to not update his server, including Kernel.

Otherwise you need to find someone that forgot that he has an Proxmox server running somewhere, like setupped and forgot

The Lady is on 8.1.3 now and still running....

 

[Darkk]

The Lady is on 8.1.3 now and still running....
View attachment 59019

Nice. Can't imagine what version of the kernel still running on that node since it needs a reboot to load the new kernel.

 

[itNGO]

Nice. Can't imagine what version of the kernel still running on that node since it needs a reboot to load the new kernel.

Linux 5.13.19-1-pve (Tue, 23 Nov 2021 13:31:19 +0100)

 

[sb-jw]

Wow, a collection of admins who are very negligent with the security of their hypervisors. It's neither nerdy nor cool to boast about such uptimes and outdated kernels. Of course there may be exceptions, for example if you use live patching.

I'm sorry to spoil your fun, but as someone who lives information security, I absolutely couldn't resist the comment.

 

[itNGO]

Wow, a collection of admins who are very negligent with the security of their hypervisors. It's neither nerdy nor cool to boast about such uptimes and outdated kernels. Of course there may be exceptions, for example if you use live patching.

I'm sorry to spoil your fun, but as someone who lives information security, I absolutely couldn't resist the comment.

As you can read in this thread.... in our case this is an isolated system....

 

[sb-jw]

Yes, I can read and this may apply to you. For others, the server is open like a barn door on the internet.

From an information security perspective, there is no valid reason at all why a hypervisor that can still be updated free of charge is not updated.
Even if it no longer works, the system belongs behind a firewall in your own network, so that using the systems is associated with maximum pain so that someone can take care of migrating the services. As a risk manager, you simply can't bear that anymore.

By the way, please don't feel attacked because I wasn't addressing you directly, but rather the general population of people who think updates are annoying and prefer not to do so. Basically, my answer may be a wake-up call for some people to update and reboot their system.

Everyone here can do whatever they want with their system. I have now given my 2 cents and leave your Schrödinger's servers to you (am I hacked or not - you don't know) ^^

 

[Nollimox]

please don't feel attacked because I wasn't addressing you directly,

But that was implied...no reasonable persons reading this thread would think "ah, this the solution I had been waiting for."