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.
Comments
ok
Let's go!
Please post a screenshot of the full ticket.
I have been very happy with Linode, I know they did some migrations recently. but I chatted to them and nobody seems to know whats going on.
How long should I give them before I move, or is this long enough. I am shipping my data back to elsewhere.
Yeah fair enough, I guess Linode are feeling the reality of charging less.
Seems Linode is not doing so well these days I have seen a lot more complaints this year compared to previously which was almost 0, the idea of charging for managed backups and having it down for 9 days is quite a lot to swallow, for Linode it should be unthinkable.
Maybe move over to Vultr.
I think you may be right, I wonder if its just me or others too..
and remember to keep your backup yourself no matter who's the provider
How difficult to spot the issue with backup process? Not difficult at all it’s just luck of server administration. As nowadays cPanel validate the backup files and send the report about backup task on daily basis. The rest from any provider is BS!
??
"Once we have everything back up and running, we'll be happy to give you a credit to cover any time your Backups weren't working properly"
I mean, that's pretty damn standard. Don't really see the big issue, or am I missing something?
Linode+managed backups+9 days without service with no real updates.
That is highly irregular for Linode.
Yes! this is because the company has brand image + reputation, on an overall basis. Plus this is not a free service OP is getting.
Who mentioned about cPanel?
Assuming what someone else is dealing with and how easy it must be, in an environment that isn't like yours, is a rookie mistake. Usually something someone does when they have just enough admin work under their belt to think they know it all, but not enough to have actually dealt with a problem you can't just google your way out of. Scale is hard. If you think their admins have gotten them this far with less knowledge than the average cPanel user, that would be quite illogical.
Make sure to ask them to cover the costs while it's not working. I didn't notice that in your screenshot. I'm sure they'd be more than happy to. Make sure you're taking other backups as well, never put all your eggs in one basket
I guess I can see that! I was probably a bit too focused on the "linode charge me"-portion of the post.
I'll admit the responses from linode isn't exactly ideal.
One of my client's Linode have similar backup failure, one of the disk could not backup, I ended up backup the linode to liteserver.
Linode give me $10.
Don't know why, It looks like a small probability event, I have over 100 clients using Linode's backup, only seen this once.
still doesn't matter with cPanel or without, can be setup script which will do exactly the same function - from our experience.
You are hired!
-Bezos.
You can set up a whole fleet of KVM hypervisors to backup guests without ever experiencing an issue now or in the future on any subset of those hypervisors, all without disrupting the service of the KVM guests on servers of minimal or maximum capacity, and without breaking a sweat... just as easily as setting up cPanel backups?
Well you're hired then. Better be able to deliver.
What the fuck man! I hired him first.
You asked for it.
-still Bezos.
If Linode is using something like CRIU whatever can go wrong, will go wrong.
Got a bit more information :
"Some other customers are also affected by this maintenance, not just yourself. We are working to mitigate the issue and certianly hope to have this resolved shortly"
And it is now back online..
Thanks for the comments.
save your backups in the freezer like what @joeri does with his burgers
Well, shit can happen, but to last 10 days is not exactly stellar.
I think that, still, overall, linode delivers better performance than even top rated leb hosts. Whether they will be able to keep this up charging as they do now, is up for debate, people already noticed slips, but if they do not grow worse, it should be ok for them.
Linode is pretty good with credits on downtime but than they also promised there will always be a xen node for those still on them. Still unhappy about that.
I don't see Linode going anywhere. More likely, they're experiencing the pains of growth and adapting to newer technologies. I know the KVM transition was a while ago, but quite frankly KVM comes with it's own set of issues at scale that are unique to it. Those unique issues will never fully disappear, nothing coded by humans will ever be 100% flawless.
Like almost everyone is doing in this market when they reach a certain size, they probably purchased larger servers with intent to fill more capacity per rack space, to help them maintain roughly equivalent revenue stream while lowering prices to compete in the space. This comes with unique challenges too. High tenancy per hypervisor means that if a fix is just "a reboot away" then you have to recognize how many people will be down, for how long, what their value is to you, and whether or not they'll leave over it. Hell, at HG everyone always said you'll only reboot a shared server once, and damn were they right. With that many people on a server it took forever to boot, and in came tickets, tweets, phone calls, and you're suddenly having to propagate this information out to every team so they know how to respond, and even so the next shift knows how to respond when they get to the tickets. Life is hard at scale. I say it a lot, but it's because I don't think it's very evident. I never imagined any of it and thought myself better than the big companies who took forever to do anything. Boy did I learn.
Anyway, reliability of the base platform can often be considered a more pressing matter than fixing an issue for a minority. Some issues could even require an upgrade of QEMU and therefore require that every instance be killed and started new to take advantage of it.
That to say, there's a lot to consider at scale. Every action has a reaction, and you have to consider the weight of every action even if it would have seemed so simple "back in the day." Literally everything is easier when you're not huge, and Linode is pretty huge.
I like how NetCup has handled this. They put a notice in your container controller which says that there is an update which addresses , and you need to shutdown and restart to receive these fixes. Until that time, your container runs on the old system.
Yeah, this is truly the best way to handle that situation, assuming not an average setup where the update is an escalation vulnerability.
Ze Germans vill not have ESCALLATIONS! Ze machine be rebooten und showin das blinkenlites. SCHNELL! SCHNELL!