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
This can happen for variety of reasons, even a kernel/systemd upgrade within your VM causing it to detect the interface slightly differently. Also, if your VM was migrated to a different HV, the hardware layout or underlying network card bus changed, this could've caused the OS to change the name.
Systemd assigns these names using a order of rules:
1. Names provided by the BIOS/firmware (
eno1- Ethernet + onboard)2. Names provided by PCI Express hotplug slot numbers (
ens3- ethernet + slot (PCIe slot number))3. Names based on the location of the hardware reported by PCIe controller (
enp1s0- Ethernet + PCIe bus)4. Names based on MAC address (
enx79c3d4...- Ethernet + MAC address)So ens3 = slot 3 (hotplug slot 3), while enp1s0 = PCIe bus 1, slot 0 (hardware address 01:00.0).
It seems in your case, Linux wasn't able to get the slot number directly, so it fallen back to PCIe bus location of network card.
There's a mitigation which most providers use which prevents this behavior and restores the old interface naming scheme which tends to work better for virtual machines. You can enabled it by adding those two parameters to kernel command line:
Yeah, i was thinking something along the lines of a lone
apt-get upgrade(the i will 100% reboot at some point but not now variant) in combination with a node restart but then it could probably also be some kind of change to the emulation settings or some control freak panel messing around with/etc/network/interfaces. I just love blaming systemdGlad to hear that.
You should have done this before opening a ticket, for any unmanaged server, at any provider, even with a 1 minute average response time. Not sure why "promo machine" is an argument here. Unmanaged is unmanaged, and troubleshooting yourself should be the first step before contacting support.
I assume from the excellent service and response times you have received on all of your other services, you are making an assumption that only this one is a promo service. But all four services in your account, are in fact, promotional services and zero of them are standard ones.
I am glad you at least agree to the fact that you have received much better support than what was advertised when you purchased these servers, that you even forgot these were promo services. Hopefully we can expect a separate review for that great experience (or do you only leave reviews when they are negative)?
that's really good support.
Classic PEBKAC (Problem Exists Between Keyboard And Chair) case here.
Honestly, @bxk6rbd, blaming the provider for a 5-day "outage" when it was just a local network interface naming mismatch (ens3 vs enp1s0) is a bit rich. We live in the era of LLMs—if you had spent 30 seconds pasting your ip a and NETWORK UNREACHABLE logs into an AI, you would have fixed this before your first coffee.
This is the reality of Unmanaged VPS: The provider guarantees the pipe to your door, not the plumbing inside your house. If the host says the network is fine and you can't even ping your gateway from the console, 99% of the time, it's your OS acting up.
That said, @hosthatch, you guys aren't entirely off the hook for the PR mess. While "Unmanaged" means we shouldn't expect hand-holding, a simple UI improvement would save everyone's time:
Just add a "Estimated Response Time: 48-72h for Promo Plans" notice on the ticket submission page.
People don't mind waiting for a $18/year deal, but they hate waiting in a "black hole." A bit of transparency goes a long way in preventing these "crying wolf" threads on LET.
No offense but i am not entirely sure whats more scary: People not checking if their fucking interface is up, has an IP, ... or the idea that it would take an AI to do these kind of very basic diagnostic steps.
Sounds like Gemini slop.
Agreed. Even from my personal experience, HostHatch's support has been always great.
You shouldn't have even needed those instructions. Like he said, it's an unmanaged service. If the panel allowed you to boot into a custom ISO, you should have all the information required to solve the issue. If you don't know how to do that, you should pay for a managed service so that you can demand that the provider do things that you could have done yourself.
Please understand that support is not free. It takes time out of the provider's work if they're a small provider, and it costs money (real, significant money) to hire tech support if they're a bit larger. For a cheap promo VPS, it's very often the case that the provider takes a net loss the moment you open even one single ticket.
@hosthatch has provided you with more support than you needed. He even explained, step by step, how to boot into a rescue ISO. Even if he had said nothing more than "This is an unmanaged service and the node is functioning correctly. Our panel allows you to boot from custom ISOs to troubleshoot your configuration problems.", it would be more than what you should expect. Nothing the provider said was rude, even while you were rushing him to fix your own problems for you.
For an unmanaged service, you should only open a ticket for a small number of reasons:
Struggling to realize that you have to boot into a rescue ISO is most certainly not one of the above.
My hosthatch Singapore VPS operates normally.
If you haven't already, I'd encourage you to change the root passwords on all your servers and record them, even if you use key-based authentication. In case you need it for something like this.
Before I created the first ticket, I confirmed that the inability to access the machine was due to a network issue, and that the problem was not caused by me.
I haven't received "excellent service and response times you have received on all of your other services". In reality, I rely very little on the VPS; I don't do business with these, and these machines only slightly improve my cyber life. The resource usage I used on these machines would make you service providers happy, but you've made me unhappy with your foolish methods! The only reason I'm getting increasingly angry is that I clearly felt this deliberate disregard and delay while waiting! You are service providers! You serve customers! We customers can't and don't want to remember your various terms and conditions. If you think customers should remember any particular terms, then you should constantly remind them in the control panel! Don't act like royalty!
Look at the ticket I gave you. Three years ago, the Singapore VPS had network problems. You acknowledged the problem and resolved it. Then, when it came time to renew, I was too busy and completely forgot. I only remembered when the VPS became completely unusable after the deadline, and then I asked you to restore the machine. Otherwise, I didn't bother you at all. In the last three years, it's you who have been bothering me from time to time with machine migrations! For a long-term customer like me, with minimal resource usage and who hasn't bothered you for three years, instead of showing a little more care, you deliberately create unpleasantness. I think these are the more important points, and everyone on LET needs to see them!
Overall, I used to have a good impression of you, otherwise I wouldn't have bought 4 machines from you. However, you've destroyed that goodwill yourself. Not only do you require customers to remember that PROMO machines only enjoy low-priority services (your actions demonstrate this), but your technical skills are also inadequate. You made a customer's system to inexplicably change its network interface name, causing the machine to completely lose internet access (the machines were set up without automatic updates or restarts enabled, only the ports I needed were open, and I didn't plan to touch them for five or ten years; how could they change the interface name on their own? I've never encountered this with other companies).
I just saw my new ticket requesting a refund of the remaining balance, and they refused. Speechless.
Well no shit they refused. Did you read their refund policy which you accepted when you purchased the service?
Notably, "being annoyed by slow ticket response time and regretting sending $100" is not included.
He made the interface name change? You say his technical skills are inadequate, but you installed your system with an interface naming mode specifically designed to change if the underlying network card changes, and migrations across nodes is not an uncommon occurrence. If you didn't want it to change, you should have enabled old-style interface names.
That's on you and shows that your technical skills are inadequate, not his. Perhaps you are not yet capable of running a VPS correctly and would benefit from paying for a managed service.
lmao @ op
No, you have not. Your interface naming scheme changed. No one currently knows why this happened.
If you have indication that there was some drastic change to the emulation settings (like a very different network adapter being emulated) or the virtualization environment messing around in /etc/network/interfaces you would have at least one leg to stand on here but as is nothing in this direction has been provided.
LOL
Even if it was because of e.g. a VM migration to a node with a slightly different configuration, OP should have 1) figured this out himself or 2) enabled old-style interface names. @oloke explained to him exactly how to do that.
The sad thing is that customers like this increase the price of VPSes across the board because providers have to take into account losses from idiots making unnecessary tickets.
Read what I wrote in an earlier post:
There is no intentional delay. There is just an entitled MJJ trying to strong-arm an unmanaged provider into providing managed VPS services despite using a low-end promo that likely doesn't even cover the cost of answering a single ticket and then complaining that they aren't making a custom exception to their refund policy.
It's not like there was some secretive and slimy policy hidden in their ToS. It's as simple as having lower priority in the support queue. That's not uncommon for cheap providers because you aren't paying them enough to hire a huge number of support staff who are willing to answer basic questions that you could have asked Google.
That's a shame. It means you were a perfect customer for three years, only opening tickets when it's genuinely warranted.
A long-term customer paying peanuts who started nagging the provider's support staff on an unmanaged service to get them to fix a problem that you already had the power to fix. They are not your teachers. Their job is to set up a hypervisor with a good network and slap a VM on there for you to run. As long as the node is running and the network is connected, their job is done. If you want them to do more, like resetting your password for you or auditing your configuration to warn you that you're using new-style interface names which may change, pay extra for a managed service.
You're making a fool out of yourself. What you're showing LET is that HostHatch goes above and beyond what they need to do, using their own time (and thus money) to hold your hand and give you a step-by-step explanation of how to boot into a SysRescue ISO and enter a fucking chroot. So thanks for showing us that they're so solid that the only problem you've had with them in the last three years was purely an ID-10-T error.
You need a managed service. You clearly aren't equipped to operate a VPS on your own.
You literally said you're speechless because a provider isn't making an exception to their refund policy just for you. So who's acting like royalty now? It's like you walk into a buffet and your jaw hits the floor because because your mind can't even begin to comprehend that the cook isn't putting food on the plate for you. Grow the fuck up.
First one ticket, then spamming tickets in a row, then asking for a refund. Classic MJJ move. Here we go again.
btw why do I keep seeing this every single year?
And that part was actually already me being extra fair. I very much doubt there would be much network related configuration in relation to emulating an average VPS able to cause a fallback in the naming scheme.
I mean, when looking at the fallback clusterfuck systemd's geniuses describe it seems that it suddenly wasn't possible anymore to determine the PCIe hotplug index. That's not just some slight difference.
At the very least the originating node would need to emulate a plugable PCIe based adapter while the receiving one (for whatever reason) doesn't. Maybe qemu's default would be to emulate an onboard adapter (i somewhat doubt there to be so much variance but maybe ...). I don't think something that fundamental would just suddenly change among the usually rather uniform infrastructure of a commercial host.
The whole idea is just very hypothetical.
I've had the network naming thing happen in the past after system updates, both vps and my home machine. Without any hardware changes. I put the grub fix in place a long time ago to avoid that.
This is a perfect example of someone who is knowledgeable enough to use an unmanaged service, unlike OP.
Happened to me on Debian too. Interface name changed after a kernel update and the VM went dark. Rescued it via VNC and moved on.
The OP's technical knowledge appears insufficient to manage an unmanaged server; he lacks the ability to understand and handle a Linux system. It is recommended that he opt for a managed server instead.
Wouldn't the VPS have had to be remade when migrating from the old system to the new one? It seems this is the first time using it in almost three years.
Could this be it?
GRUB_CMDLINE_LINUX="net.ifnames=0 biosdevname=0"Have you tried reading this topic?
) excellent post that you just skipped, totally.
@oloke typed (or semi-generated?
https://lowendtalk.com/discussion/comment/4791003/#Comment_4791003
Yes, as referenced by @oloke at the top of this page.
https://lowendtalk.com/discussion/comment/4791003/#Comment_4791003
Hello, do you still need your server? Or would you like to transfer it to me?
What a waste of time reading this thread.
Yaa, even my ticket not replied the problem is already solved
if its problem on hosthatch network, they will solved ASAP, then i just closed my ticket