this post was submitted on 13 Dec 2024
75 points (87.1% liked)
Firefox
18037 readers
333 users here now
A place to discuss the news and latest developments on the open-source browser Firefox
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
My laptop with a non-critical service: Uptime: 9 weeks, 5 hours, 34 minutes
I've had Debian VMs run for long periods of time without me touching them. They normally would have high uptime unless it automatically reboots to apply a kernel update. The key is these are virtualized servers. You should absolutely avoid running to long without a reboot. The longer you wait the greater the chance of something breaking on the next boot. There is also the issue of memory fragmentation but that's not really an issue these days.
I just have docker containers serving up some self-hostable service for myself.
I don’t think I’ve seen or heard of issues not rebooting for too long recently. Aside from not getting security updates or bug fixes, what would be some problems that could happen if a system has been running for too long?
It might not come back up after power loss.
Also you do want security updates. It is a bad idea to not install them.
Could you elaborate on it not coming back up after a power loss? Assuming these services can get restarted after booting without the need for a user login, why and how would a previous long uptime lead to a possible failure of these services to be spun back up? I apologize if these questions sound dumb and have obvious answers, but I genuinely do not know, and it’s why I’m asking.
And I’m not in any way trying to say I don’t want security updates. I’m asking that aside from security updates and bug fixes, are there any downsides to a long uptime? Please treat the question as one of curiosity.
It can happen because of simple things such as a hardware failure or because the kernel was removed 3 weeks prior. Its unlikely but it always will come at the worse time.
Also rebooting after any update makes sure that all services have been restarted and are using the newest libraries.
I’m sorry but I fail to see how these problems would be tied to having a long
uptime(note the inline code block, as I mean the output of that command instead of uptime in an SLA, which is typically described as high or low instead of long or short). I have yet to find mentions where long uptime leads to higher chance of hardware failures as of recent. If some critical library or the kernel was removed some weeks prior to a reboot, I don’t think long or short uptimes would change your encounter of these issues.And security patches are good, I agree. But there are instances where you don’t need it, eg in an airtight infrastructure, meant just for internal users, of which has no access to the Internet. You fall back to more traditional approaches to security in such cases.
As far as whether a service is properly restarted due to library updates, in a containerized environment, you wouldn’t have issues with library version mismatches, or missing libraries, or any sort of failure to restart due to dependencies getting changed without human attention (note that you can automate container updates, but you are then putting trust into whoever is publishing that container).
I’m not sure if it’s a lack of understanding of what my question is asking, or some other reason, but if you would please take the time to carefully read my questions and answer more appropriately and with clarity, that would be much appreciated.