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
Got an email a couple days ago from Terabit with a new IP and credentials for a Debian 13 VPS that I never setup (I don't use Debian 13). Uptime Kuma started chirping that my Enzonix VPS, Debian 11, was not responding.
Logged in to Terabit, no servers in my account.
Logged in to Enzonix and there's my old server. Didn't try to access it, because... don't care at this point.
Tried the new server credentials from the email, they don't work.
Good job guys, "moving" mhmm. At this point, just keep the money and stop emailing me, I won't be renewing anything.
@wadhah My vps is not responding. i can't restart it from panel. Ticket #8995792 opened for a day now but no answer. can you check it out pls ?
Looks like Terabit can't deliver on its promises lol. This “migration” has turned into a total drama.
yeah idk everyone was into open communication until the conversations got difficult and now i guess we're being babies about it
And here I thought the folks getting migrated to poverty spec Xeons were getting the rough end of the stick. Guess it always could be worse.
Still no answer from support. this is getting ridiculous
Hi there,
I understand your frustration with the migration process and timeline. We sent out service migration notices, but I think there is something important that needs to be conveyed:
The sheer size of the data at Enzonix is large, and the overselling is significant. If you're unaware, we agreed to take over the clients at Enzonix because Alex wanted out and it showed the signs of impending doom -- it was going to deadpool if we didn't intervene. We also wanted to acquire WitherHosting from him, and we were set to until he changed his mind on that.
Since we took over the clients, everyone who has been on our infrastructure has been getting free time from our perspective. We didn't migrate the billing yet and did not fully intend to until all data was migrated over (not to mention, the import process from Blesta isn't fantastic), and billing going to Enzonix was not distributed to us -- and yes, Alex did ask us to foot the server bills, and we still footed the server bills out of pocket despite him obtaining the revenue from the service, which largely runs on our infrastructure except a few nodes remaining.
We tried to do this in the very simple approach, taking our time a bit longer than most acquisitions would, because of the size of data and time needed to migrate files.
When we took over the systems, we were given very short timelines for a chunk of nodes needing to be migrated, some were only a few days away from expiring and we had to step in and migrate ASAP, so we moved them to the infrastructure we had on hand -- there was no time to acquire additional builds to deliver equipment. At the same time, we went out and bought another company in the same facility we're inside of (largely hardware purchase) which we migrated and started deploying on as well to perform more emergency migrations without having to wait for hardware lead times. We have spent tens of thousands of dollars to buy more equipment (and recently, inflated pricing equipment) and to continue to deliver on the deals despite not receiving any revenue for it still.
To give you perspective, the vast majority of Enzonix's revenue is what I'll call "dollar revenue". It's large specs, in one case someone holding over 700GB of NVMe for less than $24/year. It's unsustainable, and we've still kept to our original promise of honoring the deals that are sustainable -- and in this case, even the unsustainable ones.
Another example, is the Enzonix cPanel shared hosting, the specifications of it and price being $0.99/month, simply is unsustainable given the cost of the accounts is ~$0.35, once you add payment processor fees in, there's no money to be made there, even if it's reselling NameCrane, there's no profit and it does operate at a loss.
If it's not clear already, the only reason Enzonix sustained these deals was because Wither, the other brand, supported Enzonix.
Thanks for your understanding and taking the time to read this.
You did too much in a short time. The loss is heavy if you would go down shortly.
I am the one who has the unsustainable deal that Terabit is talking about, and I really appreciate their work on it. It's clear that they are not profiting for it and they are doing all they can to make it work. I really thank them for that. @wadhah and all Terabit staff has been extremely kind to me even when I was somehow impatient, and they are doing everything to get this solved as soon as possible.
Everyone needs to calm down and wait a little bit, and it will get solved as soon as possible. I'm not a Terabit employee, but i feel sorry for them cause they are working hard and people are accusing them of not delivering promises, which they are trying the hardest they can -- I even worry about their mental state!
So, again, let's calm down. I am a customer and I'm having some issues as well, but I am pretty sure that they are going to solve it as soon as they can!
this seems like complete bullshit and idk who's more at fault. them selling you hot shit knowingly which i thought they were better people than this or you guys for accepting the deal without good DD on what you were getting. seems like they offloaded garbage onto you guys. good on you guys for keeping a promise but Jesus christ.
I'd hope and assume you did actually receive some kind of financial compensation for taking on that burden in that case, otherwise I'd have to wonder why you'd even consider that rather than agreeing with enzonix on an announcement to let people know the situation and offer a special refugee deal with migration instead. That way you'd have got paid while also helping people on the transition.
Honestly, the way it has been done doesn't really help anyone. The old IP just being yanked, having to migrate to a new IP just a couple of months before renewal time anyway (assuming they were mostly BF deals from last year). Especially coupled with unclear notification on timescales meant that it was impossible to plan a migration between IPs.
But taking on a massive liability in buying more machines seems foolish given what was provisioned. I personally have gone from an epyc to Xeon and now have less than 1/3 of the old GB score. Unless there's a guarantee of migration to decent spec hardware, there's just no point keeping it even if I was cheap originally.
It's a shame as my other VPS with you is good, so I know you can do it.
Another way to see it is an act of goodwill and gain 1000s of customers.
Even if a small % of customers stick around to buy other services layer its a big win.
I dont know the actual customer acquisition cost but inam sure they have had that in the calculations for the long term.
Our intention once the dust settles (all migrations are done, billing moved over) is to look at moving VPSes over to EPYC and Ryzen, but in order to sustainably do that we realize that some changes need to be made.
For transparency, each hypervisor we deploy we aim for a minimum of $250-300/month in revenue for Ryzen. For EPYC and the beefy Xeons, we aim for about $500-650/month per hypervisor as we stack these with 1TB+ memory, 6-8x 4TB Ent SSDs, etc. and at the density scale it works great for our use case.
At the current cost of Enzonix clients ($/mo in revenue), if we put them all on Ryzens we'd fill ~20 Ryzens, and we'd stand to make only ~$89-103/month per hypervisor on average. It doesn't meet our density requirements and effectively burns half a cabinet of equipment. That doesn't account for the underlying bandwidth costs either, of which Enzonix clients did draw a surprising number of bandwidth, and with our addition of Cloudflare Magic Transit for all prefixes, new 100G ERA-IX ports, more 100G T1 circuits coming up and other changes, it does get costly, but again it's the cost of doing business so we reasonably eat it.
I completely agree Xeon's aren't fantastic to be landed on, and it's why we will be moving them off in due time. If we had more time from the original impending doom, we would've moved slower and gotten more equipment on hand first -- but time was limited and we had to move fast.
The decision we've made internally is to acquire some more 7443s and 7763's to meet our density requirements without burning excessive power and cabinet space.
Any of the VDS lineup, will end up on Ryzens without a doubt once the dust settles. After all is done, if space permits we'll permit the migration between CPUs. We just racked another 4x 5950Xs and two more 7900's in Montreal which go into production likely Tuesday after the initial burn-in testing is completed, and we put another two 7900s in Kansas, while Amsterdam is behind on equipment right now as we're waiting on our shipment of 768G of DDR4 64G DIMMs we ordered -- the boxes are there, just no memory in them yet. We also upgraded the NICs in all the nodes to be 2x100G where bandwidth exists to do as such, one of which is Amsterdam, the other is Montreal (Kansas is waiting on another 100G GTT port, ETA Jan 6th).
In our case, we just didn't want to see them lose their host/data/etc., so @wadhah and I jumped into action to keep the lights on at any cost, even if that meant numbers of sleepless nights doing data migrations, and far too many 'ffs' moments.
Well hope it goes well for you. I do share @ascicode 's worry that if you scale up too quickly and there's a big exodus of people who weren't happy, there is a huge risk of being left with a ton of hardware that's basically impossible to fill, and negatively impacting your existing business. I'm sure you've run a few different projections though.
I just think you guys need to at least post frequent updates and answer all doubts here to at least calm down the nerves of users. This will stop at least a little bit the exodus Ralf said
Germany service went down. Enzonix panel still shows the old IP and I got no migration notice prior to downtime, which was promised.
Should we guess that the host machine rent/colo contract expired at 01.01, while nobody bothered to migrate the services anywhere, after all?
Mine also went offline around 9:20PM...
Can confirm, offline here too (69.42.220.xxx).
Was to be expected :P
rip
YES
And congratulations—not only have you lost service, but now you have to listen to the merchant's plea: “We're having a tough time, please be patient.”
Both of my VPS are down as well. I think I have seen Curtis said they planned migration in Jan. Taking into the account the speed, it will probably be Feb at the best.
I buried those two machines in my head, cannot be asked to constantly chase the provider.
enzonix got off to a promising start. i had various services that were running well (more or less; less the free s3 in germany).
when the takeover was announced, thank god i canceled everything right away, because my gut feeling about it was already pretty bad at the time, and it turned out to be true. what a mess.
@AlexPads would you mind addressing this, so at least we know what to expect? Are our vpses staying offline until the migration?
Same-same, the very first set of migrations triggered a bunch of my personal flags and I noped out.
Seems back up at the old location. Someone managed to get in touch with the DC and promise to pay for another week? Take your seats according to the purchased tickets, for the second act of the clownage.
One of my VPSs is back online. Cannot be asked to check it and the second one, whatever happens will happen.
The nodes are back online and VPS will be migrated as soon as Terabit has the new hardware configured for them.
If it happens again before the migration, can we expect to be resolved as well?
soon™
Today I check my smokeping, and I'm like wow
never saw anything like that before, what happened?