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.
FranTech (BuyVM.net) loses all data - "FRIED NODE"
This discussion has been closed.
Comments
He was paying $60/month for a VPS and if the legend is true then it would technically be managed on account of it being over $30/month, correct?
Fran fucked the drive before OP could take backups. Fran should restore from his backups.. Oh wait..
@user37489 I understand you are trying hard to make your case and possibly just troll. But feels like you have a knowledge gap. In the end if it was Microsoft that lost your data or AWS you wouldn't say a thing. I suggest you start reading Term of Services and get creative with a recovery plan. No other provider EVER , GUARANTEED will try to give you credits, refunds or even (care) dedicate time to read your rant , IF you agreed with their ToS (which you did when you signed up) which 100% exclude them from any liability for losing your data and even more if you don't even pay for backups and Nowhere says it will be backed up.
Managed != free backups. Backups are a paid addon for all plans, big and small.
Managed means a user can ticket and request help.
The user had the option to buy backups at order time, he didn't.
Again, wrong. This node was built in this configuration a year ago. Been left alone since its been fine. We didn't pick a random node to start screwing with or something.
Francisco
Price is irrelevant. Ask the provider about what they do or do not guarantee (other people would say read TOS but I don't do that, ever). Archive this answer and if they under deliver, litigate them (in court if the money is worth it, otherwise forum pitchforks will suffice).
Expensive dells disappearing sounds more like a you problem
My problem, is it that your NVMe drive failed, and you didn't even notify me.
Here's my personal TL;DR
@user37489, an utter tech-idiot (maybe smart in other areas but clearly and obviously moronically dumb in IT) who, to make it worse, also is of the "money no concern, I can pay" breed, actually does not care about his data, what he really cares about is offloading his boiling emotions like a spiteful bitch in the last stage of separation.
@Francisco a respected and well liked (by many) provider has f_cked up to some degree, albeit within a total normal frame (could have happened at most providers) and then makes the classical mistake of attempting to rationally discuss with OP and even lays out some cookies for her.
Francisco, find a way to avoid any and all confusion! Lay everything out in a way that even Windows corporate users can grasp. Example: you don't make/have automatic backups of VMs? No problem, many providers don't - but spell it out clearly and unmistakably, don't give anyone a chance to say "but I thought there were backups".
Customer can trust a provider even without auto-backups if it's clear and there is no confusion.
Anyway, customers should make their own backups, even if the provider also does.
Not really. After all Francisco didn't gamble; he did what many providers, small and large, do (PCIe risers/carriers) and even with a relatively reputable brand. Otoh I of course get your point; before changing a node one should make a backup, yes.
I'm definitely not in that gang. In my books BuyVM is among the better providers but I wouldn't and don't shill for them nor blindly defend them.
Thanks for confirming your image with me.
Your Windows Excel urgently needs recalibration.
You fucked it up by running experiments on production equipment and selling that to someone for $60/month. Why do you keep making excuses and blaming everything else and everyone else?
It already is though. You have it on check out mentioning it isn't free, and our main page mentions it's a paid addon too.
.....No node was changed until after the drives were dead. The node was the exact same setup, in the exact same U's, since it was installed last year.
Only once we declared things dead did I pull another node off our spare pile and rack that up.
Prove it's experimental. Here's your big chance, make a thread, ping all of your favorites, see who is using onboards or risers/PCI-e NVMe's. This isn't some 'wacky' idea.
Francisco
@dahartigan i dont know who has the biggest gap, you or the OP.
How is it a test when it was a production node and nobody was fucking with it?
Nodes hardware varies all the time, it doesnt mean is a test. @Francisco I just hate you used that word. It just gives cancer more sugar.
Yes, I know that, thank> @jsg said:
This. And up until I started this thread, they probably loved this idiot as a customer since I always paid my bills, never needed support, and barely used the darn thing.
What the fuck were you "testing" (your words my man) then? I think you should just stop because you're just embarrassing yourself at this stage.
Dear Fran shill, go back over the OP and read where Fran said it was a test to use the riser card in production.
Sure, I usually try to be transparent on things and that'll bite me at times. We post pictures of builds, what software we use, how we build things, etc.
We'll get into discussions with other users on how to implement certain features (people have experience from their jobs, etc). Our community has a lot of amazing members.
NVMe throughput. Again, the onboard slots on the X470D4U boards are limited so you can't hit the full throughput of NVMe drives. Not sure how I'm embarrassing myself, i'm not the one claiming that there's companies that will reball flash chips in your datacenter.
Francisco
If I recall correctly, the "test" was that this was just a non-standard configuration on how they already setup their nodes.
Its not like someone actually peeked inside the servers garden and started sticking their pokey bits in it.
@dahartigan FYI, using risers is normal. Replace the word with "different configuration". Is like using different drives or using a new CPU in the market. It ran for a year as a live server. So when top tier providers buy the latest CPU and make nodes with it and deploy to production, you think is an experiment or a test? Is not...
So why sell that to someone? Did you tell the customer that he would be getting put on a node that's going to underperform and is running in a configuration that will destroy the drive ultimately?
You're selling shit for the price of gold.
What about the other poor souls who had their servers on those drives who are too scared to come forward in this thread?
@dahartigan and @default are just butthurt about Fran pointing out some inconvenient truths in the cociu saga and are continuing this thread, not because they have a point, but just because they can.
Regular maintenance activity going bust is more common than someone might think.
The whole thread stems from a feeling of trust/reliance that was not honored.
Unfortunately, there are still some things in this world I can' trust a commercial vendor for and they are (non-exhaustive list) backup, replication, encryption, and asset/wealth management. Not for any amount of money.
I've lost data with AWS and Linode and if it was up to them, I would have nothing to recover my data from (I had my own replication so it's all good).
Going over the BuyVM website again. Everything about it lured me into thinking it was trust worthy, but turns out the operations are bush league.
My fault for trusting them.
You really should ask around and see how many users put their NVMe's on risers, or use u.2 slots. EVERYONE does. This isn't some big experiment. There's absolutely no reason in mine, or probably anyones, that it would crap out.
You really should ask around. You know enough hosts, go ask them if they ever use risers for their M.2's and you'll find plenty. Many large datacenters do for the reason I gave. Go ask, make a thread about it.
Scared of what exactly? We've helped most users already in ticket. If they had a problem they're welcome to voice their annoyance.
Francisco
Sorry, my bad.
His wording was clumsy but his doing was not. I know of diverse providers, ranging from relatively small ones to ones that are internationally known and respected and really large, doing the same; most in fact do it using dirt cheap risers/carriers and only very few use e.g. HP or Dell ones. Frankly I'm positively surprised to see that a low-end providers at least uses Supermicro stuff.
@Francisco
A propos "clumsy wording": How about apologizing to dahartigan? You treated - and threatened - him far worse than OP (to whom you threw cookies) but AFAIK you didn't follow up on those legal threats.
Nah, I know that you know my beef is with providers attacking other providers - and if they don't do that, they don't expose themselves to the same.
If someone is a spelling nazi, they need to watch their spelling on everything they type or get called out... it's more along those lines.
@stevewatson301 Makes sense.
Wording - I'll take that lump. I'll widen my vocabulary in the future, or just nix details.
As for Haritgan, oh my friend, that's where you're wrong
If only you knew.
Francisco
Yes you're completely right. I just feel like an idiot for trusting them.
Anyways, thanks to some recommendations from people here, I've now got external backups of new Windows VPS.
@Francisco is digging himself a hole to China here..
This is the way.
@user37489 great, so now we can move on and close the thread. Thanks! Best of luck and make sure you have a recovery plan in place.
I'm yet to see a fuckup on that scale, which is a whole datacenter getting involucrated without communication and replacement.
We know that it really wasn't, they just reduced the business "footprint" by a whole lot and giving the appearance of issues was a ruse to prevent people from asking service credit as refunds. If your man @cociu was a bit honest, you know, you wouldn't have to have providers point out these inconvenient truths to you.