@deafness said: @VirMach Can you help me solve this problem? I can't do any operation, and I can't even have functions such as issuing the ticket
Just offhand, I think you need to have checked the "Payment Agreement" box, the dialog you've got explains what situations won't actually charge you and yours should fall under that.
No, I checked that option
This appears to be working as intended. Stop trying to make a priority ticket, it's attempting to save you from making a $15 mistake.
But now the VPS in my account cannot be used. Their state is offline, and I can't send a ticket to explain the situation. Even in the time I give you a reply here, one of the VPS shows the connection timeout. The following is a screenshot of the timeout. And the timeout of this VPS has not been displayed from the control panel from my purchase. The Invoice of the VPS in my account is:#1344478,#1444057
@VirMach Can you help me solve this problem? I can't do anything on your website, even the most basic tickets cannot be guaranteed. At least you should let VPS start. I paid the latest bill, but now I can't turn on the machine.
@deafness - Please click the "Management" button on the left menu shown in your screen shot above. When the next left menu appears, click "VPS Control Panel". That should open SolusVM, if the VM still shows offline in SolusVM, you can attempt to restart it. If it shows "unavailable" please report the Node name here.
A lot may change on it when released (already thinking about lowering bandwidth amounts or some kind of clear fair use speed cap.) Let me know what location we should send the first one to (not Tokyo.) Disk is currently planned to be RAID1 (enterprise Gen3 7.68TB.) Processor is currently planned to be Epyc 7452 (2.35GHz.) IPv4 may or may not be included at base price. SolusVM may or may not be used.
The idea is lower disk amount, slower disk, slower processor, and just a boatload of RAM at a hyper-competitive price.
(edit) These are at MSRP. I wrote out a long since-deleted (or never posted) message about possibly being able to hit $15 a year special on the base plan (which was actually the 4GB about a week ago.)
Geekbench 5 could be up to 1000 but probably around 600 or so single threaded in real world use.
If your going to be offering 2 Cores, 4GB RAM, 30 GB NVMe, for $15/year, and increments upward from there. I think you should expect a lot of takers. I'll probably buy one depending only on if the location is not some where I already have big VMs long term.
I don' t need more RAM so I won't comment on merits here
But please fix the Purchase button, it's ugly, can't see shit, need some contrast - just steal the button above design.
Thanks! https://i.imgur.com/sPyXAml.png
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
@Jab said:
But please fix the Purchase button, it's ugly, can't see shit, need some contrast - just steal the button above design.
Thanks! https://i.imgur.com/sPyXAml.png
@VirMach said: what location we should send the first one to (not Tokyo.)
Los Angeles! Seattle! San Jose! Any location near Tokyo
@haynhat said:
Anyone has servers working on LAXA009 ? The node is currently locked for more than 2 days (not mentioned in the Network Status)
Yes, I have a VM on node LAXA009. It has also be showing locked in the billing panel, which as noted there means it is under going maintenance. Mine has been up and down over the last couple of days. Currently it has been up for the last 15+ hours.
@haynhat said:
Anyone has servers working on LAXA009 ? The node is currently locked for more than 2 days (not mentioned in the Network Status)
Yes, I have a VM on node LAXA009. It has also be showing locked in the billing panel, which as noted there means it is under going maintenance. Mine has been up and down over the last couple of days. Currently it has been up for the last 15+ hours.
Date
Time
Status
Elapsed Time
20221213
11:04
UP
15 Hours 19 Mins 40 Secs
20221213
10:10
Down
54 Mins 0 Secs
20221213
09:42
UP
28 Mins 0 Secs
20221212
15:46
Down
17 Hours 56 Mins 0 Secs
20221212
07:26
UP
8 Hours 20 Mins 0 Secs
20221211
22:26
Down
9 Hours 0 Secs
20220702
02:01
UP
162 Days 21 Hours 24 Mins 59 Secs
Time is US Central Standard Time
Thank you, @FrankZ. Mine are now up too.
Seem you have server on almost nodes :-)
@FrankZ said: Yes, I have a VM on node LAXA009. It has also be showing locked in the billing panel, which as noted there means it is under going maintenance. Mine has been up and down over the last couple of days. Currently it has been up for the last 15+ hours.
Leaving it locked for a little while longer before I call it "stable enough" and unlock controls.
@VirMach sorry for ask again and again, since got no any reply, but it's the last, please just repy Yes or No, I will tell my friend and don't ask anymore.
My friend ordered this plan at Tokyo before, should have 3GB RAM, but when you recently created it at LA, only have 2GB RAM, will you honor the 3GB RAM or not? thank you.
IIRC VirMach said earlier that they made a change to new purchases of this plan and it now only comes with 2GB RAM when it is deployed. He also said that for purchases made before, when the plan had 3GB RAM, he would honor that but the extra RAM needs to be added manually. Let me try to find the comment so as to find out if you need to ticket for the extra RAM.
@FrankZ said: @hotsnow - Unless I am misreading the VirMach statement I would close the other ticket and make a priority ticket mentioning the RAM issue in the title.
@Virmach can we make priority ticket mentioning the RAM issue? thanks :'(
@Virmach can we make priority ticket mentioning the RAM issue? thanks :'(
A lot may change on it when released (already thinking about lowering bandwidth amounts or some kind of clear fair use speed cap.) Let me know what location we should send the first one to (not Tokyo.) Disk is currently planned to be RAID1 (enterprise Gen3 7.68TB.) Processor is currently planned to be Epyc 7452 (2.35GHz.) IPv4 may or may not be included at base price. SolusVM may or may not be used.
The idea is lower disk amount, slower disk, slower processor, and just a boatload of RAM at a hyper-competitive price.
(edit) These are at MSRP. I wrote out a long since-deleted (or never posted) message about possibly being able to hit $15 a year special on the base plan (which was actually the 4GB about a week ago.)
Looks interesting... definitely worth a buy when it comes out.... location, I prefer the US east coast... NY preferred
Would love to see the missing 64gb on the slider
Also, the slider is not very intuitive on dark mode on mobile.
I am running out of disk space on my mostly 'static files' VPS - 1 core, 1 GB RAM, 35GB disk. I could move my shit to bigger one... but those have 3+ cores and it it will total waste to host static files there... so I guess I am buying another one when we get "NOT Black Friday sale" ?
EDIT: Wait, I expected this years ago and bought some crazy thing on BF few years ago - 52GB disk, 512 MB RAM, 524 GB bandwidth - totally enough for my mostly 'static files' via CDN... But do I really want to move this shit for extra 17GB or disk?!
FirstWorldProblems
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
@LinuxWinMAC said: @VirMachMy vps traffic abnormal problem has been solved, thank you for your timely handling, I finally dare to start my vps! 😂
There is a new problem. This vps was assigned 2 ipv6 addresses before, but now the second one is invalid. The traffic problem is probably caused by this ip address. I don’t know if it can be fixed?
thanks.
2022-12-12 08:48 | 146.9 GB of 500 GB Used / 353.1 GB Free
2022-12-12 21:15 | 156.93 GB of 500 GB Used / 343.07 GB Free
2022-12-13 18:25 | 156.93 GB of 500 GB Used / 343.07 GB Free
@VirMach All issues have been resolved and I have closed ticket #496617. Thanks again!
@FrankZ said: @deafness - Please click the "Management" button on the left menu shown in your screen shot above. When the next left menu appears, click "VPS Control Panel". That should open SolusVM, if the VM still shows offline in SolusVM, you can attempt to restart it. If it shows "unavailable" please report the Node name here.
@FrankZ No, it does not work. When I click on the control panel, it shows the picture below
Contabo replacement only if ticket E.T.A was 1 working day rather than 1 week
Okay so I've thought about bandwidth levels a little bit more, and it's going to change to this instead:
100Mbps = 32TB.
So it'll be 100Mbps Unmetered (32TB) and then a "full speed" (10-20Gbps) amount based on 100/10,000 = 1/100th of 1Gbps or 3.2TB (also known as 10%.) So what this means is we'll basically just do 10Gbps fair use, give 32TB for that plan for example, but reserve the right to cap it to 100Mbps after 10% usage. Depending on what software we use we'll most likely only want to measure outgoing only (incoming will be unmetered, otherwise we'll double the figures, so 3.2TB = 6.4TB.)
This also will scale with cost instead of RAM amount. Rounded off a bit to look nicer. This means:
Short version for BW: it means 48GB, 64GB, and 96GB plans have their bandwidth reduced a bit and we just defined the "fair use" part for the bandwidth levels.
I added a Fair Use Guidelines section which basically covers the level of overprovisioning. Short version is that IPv4 will basically be dedicated unless otherwise noted, it covers the bandwidth amounts I mentioned above, disk space is currently planned LVM but may be thin LVM, physical RAM will be oversold 2:1 ratio, and CPU will be oversold about 8:1 ratio. Seems to be "industry standard" for these type of plans and we'll of course manage it responsibly and cut it off early if alerts go off.
Everything is technically worked out to where we can lock everything off at about half the amount, as in 0 oversold RAM, and 4:1 ratio for CPU instead, and we don't take a loss (still profitable.) So anything in between that and max is just extra profit and will be done cautiously as a bonus.
@Flying_Chinaman said:
Los Angeles! Seattle! San Jose! Any location near Tokyo
I might actually end up doing Tokyo first just to see what worst case scenario of this build ends up looking like for us. But if we do any specials for it, I'll be sure to advertise it per the 10% fair use. So for example, a "special" version of 2GB plan might look like:
$2/MO $30/2YR 2GB RAM 15GB RAID10 NVMe
200GB @ 10Gbps / 2TB @ 6.4Mbps (Perhaps a forced selection, but then we might just add like 1TB @ 1Gbps option too.) 1 Core @ 25% Special fair use 1 IPv4 & /64 or /48 IPv6
Might as well make this post longer...
Also means probably for same theoretical Tokyo special that may or may not exist:
@Flying_Chinaman said: Contabo replacement only if ticket E.T.A was 1 working day rather than 1 week
Working on finally resolving this. As always not required but would be extremely helpful if people stopped making like 3 tickets for their $1.50 pending order.
ETA around December 30th it looks like for everything to be normal again.
(edit) And yes this is very much aimed toward being a Contabo replacement and I want to make sure people understand this. It means similarly oversold but we're going to obviously try our best to make it better. As in, more idlers!
@FrankZ said: @deafness - Please click the "Management" button on the left menu shown in your screen shot above. When the next left menu appears, click "VPS Control Panel". That should open SolusVM, if the VM still shows offline in SolusVM, you can attempt to restart it. If it shows "unavailable" please report the Node name here.
@FrankZ No, it does not work. When I click on the control panel, it shows the picture below
Looks like another issue, if you used the button and it broke make sure to mention that in ticket
@VirMach said: I kind of want to do Singapore and Minneapolis.
UK? Somewhere, anywhere - London, Manchester, Aberdeen (that'd be a major coup!)
In stasis until the shitposting stops/abates.
Than=compare;then=sequence:brought=bring;bought=buy:staffs=pile of sticks:informations/infos=no plural. It wisnae me! A big boy done it and ran away. || NVMe2G for life! until death (the end is nigh).
@VirMach said: I kind of want to do Singapore and Minneapolis.
UK? Somewhere, anywhere - London, Manchester, Aberdeen (that'd be a major coup!)
Here I'll list of all the locations I've thought about, but it's definitely not going to go as fast as we want and we definitely have to end up skipping some of these locations until things get back on track.
North America:
Minneapolis
Toronto
Las Vegas
Hawaii (just so I can keep going to Hawaii to power cycle servers)
APAC:
Singapore
Europe/Middle East:
Istanbul, Tel Aviv, Dubai, or Bucharest
Zurich/Bern/Geneva
London/Dublin/Manchester or somewhere in France (Isle of Man would be cool.)
Stockholm
Madrid/Lisbon (Gibraltar would be cooler.)
South of the Equator:
Anywhere. Currently F5'ing but honestly Sydney/Johannesburg/Sao Paulo all seem like they'd be pretty abysmal the more I look into it. Sydney was the closest "winner" except I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand. (edit) Perhaps some smaller island version of this like New Caledonia would be cool and it could be called our "French" location.
Maybe: (But too close to an existing location.)
Vancouver, Ashburn, Vancouver
Maybe: (But meh.)
Oregon or Idaho
Kansas City
Oklahoma City/Tulsa
Houston/Austin
Utah
Maybe: (It'd just be cool to have but probably not a good idea, an extension of my Hawaii dreams.)
Costa Rica/Puerto Rico/Bahamas (Even worse version of this with less use cases: Bermuda.)
Ukraine/Russia/Georgia/Kazakhstan
Hong Kong
Kochi/Kerala (India)
Malta/Ibiza
Iceland
Maybe: (Haven't really thought of it but maybe this is the ultimate mediocre for everyone location.)
@VirMach said: I kind of want to do Singapore and Minneapolis.
UK? Somewhere, anywhere - London, Manchester, Aberdeen (that'd be a major coup!)
Here I'll list of all the locations I've thought about, but it's definitely not going to go as fast as we want and we definitely have to end up skipping some of these locations until things get back on track.
North America:
* Minneapolis
* Toronto
* Las Vegas
* Hawaii (just so I can keep going to Hawaii to power cycle servers)
APAC:
* Singapore
Europe/Middle East:
* Istanbul, Tel Aviv, Dubai, or Bucharest
* Zurich/Bern/Geneva
* London/Dublin/Manchester or somewhere in France (Isle of Man would be cool.)
* Stockholm
* Madrid/Lisbon (Gibraltar would be cooler.)
South of the Equator:
* Anywhere. Currently F5'ing but honestly Sydney/Johannesburg/Sao Paulo all seem like they'd be pretty abysmal the more I look into it. Sydney was the closest "winner" except I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand.
Maybe: (But too close to an existing location.)
* Vancouver, Ashburn, Vancouver
Maybe: (But meh.)
* Oregon or Idaho
* Kansas City
* Oklahoma City/Tulsa
* Houston/Austin
* Utah
Maybe: (It'd just be cool to have but probably not a good idea, an extension of my Hawaii dreams.)
* Costa Rica/Puerto Rico/Bahamas (Even worse version of this with less use cases: Bermuda.)
* Ukraine/Russia/Georgia/Kazakhstan
* Hong Kong
* Kochi/Kerala (India)
* Malta/Ibiza
* Iceland
I miss my UrPad Iceland server. Man that was a good server.
@VirMach said: South of the Equator:
* Anywhere. Currently F5'ing but honestly Sydney/Johannesburg/Sao Paulo all seem like they'd be pretty abysmal the more I look into it. Sydney was the closest "winner" except I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand. (edit) Perhaps some smaller island version of this like New Caledonia would be cool and it could be called our "French" location.
@AuroraZero said: I miss my UrPad Iceland server. Man that was a good server.
If there's any recommendations for serious datcenters there it could be a reality assuming the whole lower power costs thing they have going there if it hasn't been "ruined" by an influx of crypto stuff (aka retaliatory regulations.) Since we could just ship off super beefy stuff there.
This is just me making a wild guess though but I'm assuming Iceland would be one of those places where the citizens have access to 1Gbps fiber for like $20-50 a month but 1Gbps at a datacenter might cost $5,000 a month.
@VirMach said: South of the Equator:
* Anywhere. Currently F5'ing but honestly Sydney/Johannesburg/Sao Paulo all seem like they'd be pretty abysmal the more I look into it. Sydney was the closest "winner" except I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand. (edit) Perhaps some smaller island version of this like New Caledonia would be cool and it could be called our "French" location.
@AuroraZero said: I miss my UrPad Iceland server. Man that was a good server.
If there's any recommendations for serious datcenters there it could be a reality assuming the whole lower power costs thing they have going there if it hasn't been "ruined" by an influx of crypto stuff (aka retaliatory regulations.) Since we could just ship off super beefy stuff there.
This is just me making a wild guess though but I'm assuming Iceland would be one of those places where the citizens have access to 1Gbps fiber for like $20-50 a month but 1Gbps at a datacenter might cost $5,000 a month.
Sorry no idea man. I just remember the server was in Iceland. Never really looked into coloing anything there. I would imagine as you said though.
You'd please a lot of folks, with an "off-shore" location i.e. Hawaii, isle of Man, Guernsey, Cook Islands, Malta, Fuerteventura.
You may even get a lot of new recruits, for hands-on approach!
In stasis until the shitposting stops/abates.
Than=compare;then=sequence:brought=bring;bought=buy:staffs=pile of sticks:informations/infos=no plural. It wisnae me! A big boy done it and ran away. || NVMe2G for life! until death (the end is nigh).
@AlwaysSkint said:
You'd please a lot of folks, with an "off-shore" location i.e. Hawaii, isle of Man, Guernsey, Cook Islands, Malta, Fuerteventura.
You may even get a lot of new recruits, for hands-on approach!
I'm kind of liking this idea, probably until I see the prices. It's too late I have to commit now, I'm the proud owner of VPShore.com and islev.ps (not to be confused with v.ps.) Actually just kidding I didn't register islev.ps even though iSlev sounds kind of cool, we don't have the kind of budget for $59 a year domains but I did also grab isle.host to add to my C-tier list of domains we'll never use. Then later we'll buy the PREMIUM (apparently) domain hostisle.com for $2500 because it sounds/looks like hostile and register some shady company name like Hostile Host Pty for our ASN because it'd match all the abuse reports we have to end up dealing with and make it easier for company to blanket blocklist.
Spain --> Canary Islands instead.
France --> New Caledonia instead.
UK --> Isle of Man & Gibraltar instead (I can't pronounce Guernsey or maybe don't like how it comes out so we'll skip that.)
Okay like only 90% of that was a joke, we'll see how far the other 10% gets us.
@VirMach said: I can't pronounce Guernsey or maybe don't like how it comes out..
Maybe Jersey is more appropriate, given how frrreezing it is here, this week!
Guernsey sounds very like girning - a Scottish word for a teething, whining bairn (baby).
In stasis until the shitposting stops/abates.
Than=compare;then=sequence:brought=bring;bought=buy:staffs=pile of sticks:informations/infos=no plural. It wisnae me! A big boy done it and ran away. || NVMe2G for life! until death (the end is nigh).
@VirMach said: I added a Fair Use Guidelines section which basically covers the level of overprovisioning. Short version is that IPv4 will basically be dedicated unless otherwise noted, it covers the bandwidth amounts I mentioned above, disk space is currently planned LVM but may be thin LVM, physical RAM will be oversold 2:1 ratio, and CPU will be oversold about 8:1 ratio. Seems to be "industry standard" for these type of plans and we'll of course manage it responsibly and cut it off early if alerts go off.
Fined by me with the CPU oversell and bandwidth cap if the price was fair enough.
But oversell mem and dik is a new level even for VirmArch
Frankly, oversell memory is a recipe for disaster, guest OS always tend to fully occupy allocated memory as preload cache.
Knew memory in OVZ can be oversell, yet that's KVM right?
Disk oversell 50% initially, serious? I'd rather purchase secondhand older nvme like Kingston PCI3.0 to cut cost. Guess you will do both
No more 2 years commitment for an experimental oversell plan. BRING BACK ANNUAL BILL Only wonna pay biannually for storage plan and GCV EPYC 6666
Hope my remain days at VirmArch does not end up like the early-stage CloudCone and shitty PacificRack.
@VirMach said: I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand.
How do you feel about 17 or 18 hour non-stop flights? Plus connections if you're not on the west coast. There's also 19 or 20 hour flights from Dallas.
@VirMach said: I kind of want to do Singapore and Minneapolis.
UK? Somewhere, anywhere - London, Manchester, Aberdeen (that'd be a major coup!)
Here I'll list of all the locations I've thought about, but it's definitely not going to go as fast as we want and we definitely have to end up skipping some of these locations until things get back on track.
North America:
Minneapolis
Toronto
Las Vegas
Hawaii (just so I can keep going to Hawaii to power cycle servers)
APAC:
Singapore
Europe/Middle East:
Istanbul, Tel Aviv, Dubai, or Bucharest
Zurich/Bern/Geneva
London/Dublin/Manchester or somewhere in France (Isle of Man would be cool.)
Stockholm
Madrid/Lisbon (Gibraltar would be cooler.)
South of the Equator:
Anywhere. Currently F5'ing but honestly Sydney/Johannesburg/Sao Paulo all seem like they'd be pretty abysmal the more I look into it. Sydney was the closest "winner" except I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand. (edit) Perhaps some smaller island version of this like New Caledonia would be cool and it could be called our "French" location.
Maybe: (But too close to an existing location.)
Vancouver, Ashburn, Vancouver
Maybe: (But meh.)
Oregon or Idaho
Kansas City
Oklahoma City/Tulsa
Houston/Austin
Utah
Maybe: (It'd just be cool to have but probably not a good idea, an extension of my Hawaii dreams.)
Costa Rica/Puerto Rico/Bahamas (Even worse version of this with less use cases: Bermuda.)
Ukraine/Russia/Georgia/Kazakhstan
Hong Kong
Kochi/Kerala (India)
Malta/Ibiza
Iceland
Maybe: (Haven't really thought of it but maybe this is the ultimate mediocre for everyone location.)
Guam
+1 for Dubai, Singapore.
+1000 for Kochi. I'm from Kochi.
It's even worse on the backend. Whoever made it (not me) did very poorly. But that's how it'll be... it's a feature! You're not supposed to put the slider that far, if you do then it's just a failsafe so the slider doesn't break off.
Also, the slider is not very intuitive on dark mode on mobile.
Okay this part I can admit I don't care enough to fix, I'm anti-anti-blinding lights.
@FrankZ said: @deafness - Please click the "Management" button on the left menu shown in your screen shot above. When the next left menu appears, click "VPS Control Panel". That should open SolusVM, if the VM still shows offline in SolusVM, you can attempt to restart it. If it shows "unavailable" please report the Node name here.
@FrankZ No, it does not work. When I click on the control panel, it shows the picture below
Looks like another issue, if you used the button and it broke make sure to mention that in ticket
@VirMach But my account can't send you a ticket, as I mentioned before when I send a ticket the button is not available, I made sure I checked the button where I agree to pay, so now I can't send a ticket on your site to let you know the problem and I can't use the vps, I hope you can help me to solve this problem, it's been 4 months since this problem.The id of this vps is 634681
And on your site, I use Troubleshoot with this result
I also don't see any notification that CHIZ002 is offline on your Server Status page.
@nutjob said: @VirMach said: I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand.
How do you feel about 17 or 18 hour non-stop flights? Plus connections if you're not on the west coast. There's also 19 or 20 hour flights from Dallas.
I don't mean to get political but I guess I will, just a tad. I wouldn't feel very welcome based on the last couple years. I know that's not the best thing to base it off but even getting detained or whatever you want to call it, for a few days, would be an absolute no-go for me right now after you add in the flight times. And then next time something similar happens, trying to figure out how to plan travel would already be a nightmare but doubly so it seems with that region since they obviously are an island nation and want to be cautious. As far as islands go though I think we'd have better luck with all the others since I assume they rely more on tourism (might as well add in that as far as the semantics of islands go, everything would technically be an island then but it has that vibe so I apologize if I'm using the wrong term.)
On top of that, we would also have to deal with the high power costs, and high bandwidth costs, and high labor costs, except it seems like there'd be no upside since no one really outside of Australia benefits (and again, that'd be like saying "no one outside of the US benefits from US hosting" except at least Tokyo comes with the better connectivity to pretty much everywhere in that region.)
I will say though I wasn't even thinking about the flight being that long, for some reason in my head it was the same as going to Tokyo even though that's wrong and even then I already had that mindset.
TLDR if I get a signed and backed by gold pass from the prime minister of Australia that I can come whenever, maybe an ounce of gold every day it's delayed due to government policies, I'd be more keen.
@Flying_Chinaman said: Fined by me with the CPU oversell and bandwidth cap if the price was fair enough.
But oversell mem and dik is a new level even for VirmArch
I have some bad news for you if you actually use Contabo and you believe they don't oversell.
Frankly, oversell memory is a recipe for disaster, guest OS always tend to fully occupy allocated memory as preload cache.
Knew memory in OVZ can be oversell, yet that's KVM right?
This isn't as true as you think on so many levels but I'll spare you the long read. I will say something though, outside of any pretty much "guaranteed loss" low quantity plans, any KVM Ryzen host that has up until today sold any 1GB / 1 IPv4 / plan for less than $17 per year is either (1) overselling memory to some level, (2) lying to you with they claim otherwise, or (3) going to genuinely go out of business. Remember I said outside of "guarantees loss" low quantity, as in a provider can obviously do this if they just make a lot of money from full-price plans and have extra space (except if they have extra space it's most likely from selling regular plans, and the node is full, except 128GB assigned RAM equals like 40GB actually used memory.)
Even someone who doesn't know what they're doing can easily oversell RAM on KVM.
And what you said about guest OS and cache, it's just not how it works but it's complicated. We'd have to do a very long lesson that no one will read.
We have 100% oversold memory when we switch from 256GB dual E5 to 128GB Ryzen. There's was other way to do, but we carefully measured everything out server by server before moving. Those servers are pretty much perfect usage levels. Where we did screw up was for Tokyo Ryzen, due to how the semi-automated system activated orders, there's one or two nodes that ended up with too many high RAM plans and even then our highest, out of ALL servers, in terms of memory usage, is around 80GB out of 128GB actual RAM used, with around 8GB free no matter how you calculate it, even with cache usage, etc. And that node has been stable for half a year, 100% uptime, and closely monitored.
Trust me this isn't the long version.
@Flying_Chinaman said: Disk oversell 50% initially, serious? I'd rather purchase secondhand older nvme like Kingston PCI3.0 to cut cost. Guess you will do both
Where did you get the 50% disk oversell from? We definitely wouldn't oversell disk by that much, that's where I would agree with you that it's dangerous. One of the reasons we did KVM Lite the way we did instead of thin LVM was to avoid it altogether. So I guess think of it this way instead, what we could do is do 15GB dedicated, or where you start at 10GB and can extend up to 20GB. Something like that.
And like it says on the page what's advertised for disk is 100% guaranteed. So if you run out of disk in that scenario, and the server runs out of disk, which would be rare, we'd move you to one with disk and give you the 20GB.
I think you underestimate how many people don't use their disk. When we do backups and anything related to grabbing the actual disk sizes, it ranges from maybe 30-65% actual usage, average.
But anyway, I don't think we'll be doing thin LVM unless we go with a panel that does a really great job and even then I don't think we'd actually be overselling it.
@Flying_Chinaman said: No more 2 years commitment for an experimental oversell plan. BRING BACK ANNUAL BILL
Okay just because you said this if we do Epyc Tokyo, it's going to be 3 year for the 2GB. That's a good idea. $45/3Yr. Then 4GB $40/2YR maybe, and if you want 1YR and don't want to be part of the experiment, 8GB will be 1Yr still.
Or we could do 2GB for $15/YR plus $15 setup fee instead of $30/2YR if you want. Let me know. We can do whatever you want, but it will add up to $30. Even 6 months for $7.50/6Months + $22.50 setup.
I almost went there maybe a year or two ago, it's hard to tell how time has transpired recently. But I didn't probably because of my stupid jobs kicks trashcan and frowns while crossing arms.
@FrankZ said: @deafness - Please click the "Management" button on the left menu shown in your screen shot above. When the next left menu appears, click "VPS Control Panel". That should open SolusVM, if the VM still shows offline in SolusVM, you can attempt to restart it. If it shows "unavailable" please report the Node name here.
@FrankZ No, it does not work. When I click on the control panel, it shows the picture below
Looks like another issue, if you used the button and it broke make sure to mention that in ticket
@VirMach But my account can't send you a ticket, as I mentioned before when I send a ticket the button is not available, I made sure I checked the button where I agree to pay, so now I can't send a ticket on your site to let you know the problem and I can't use the vps, I hope you can help me to solve this problem, it's been 4 months since this problem.The id of this vps is 634681
And on your site, I use Troubleshoot with this result
I also don't see any notification that CHIZ002 is offline on your Server Status page.
You're facing at least two issues right now simultaneously. Then a third issue is that you have multiple accounts, and a fourth issue is that you're flagged for any of the following, one or more: (1) trying to force migration to Tokyo but it's very close to the cutoff time for definitely so maybe not, (2) creating tickets about tickets and making migration request and another one about your initial issue that did get fixed at some time, and ISO request, and migration request all together while another request was still being fulfilled, and then doing it again after you specifically got a message saying not to do that. And then ticket requesting refund for failed migration while opening more migration requests. And then replying to a ticket about your IP changing that says not to use this ticket for other issues, and you use it for other issues. Oh and you had two tickets for the first issue I think, and you didn't follow the guidelines. Then you replied to two tickets like that where you also ignored the guidelines being re-iterated. Yeah looks like it's definitely for #2. Oh looks like you have another ticket open you replied to same thing that's completely also unrelated.
As for your current two main issues not including the multi-account or ticket creation issue (which isn't an issue, you're just not allowed to use priority department anymore)
You re-installed OS during maintenance, I assume maybe for the 2nd time since these were fixed but maybe it was missed.
You cropped the image to where I thought it's an important issue but it's actually not, but I've fixed that one so you can click control panel button and then spam operations there, it won't make a difference.
Anyway I just went ahead and moved you to the staging area, you need to use redeploy button now.
If you make any more tickets about this or talk about it here, the best I can do for you at this point is just a pro-rated refund and cancellation maybe.
The price for getting your issue resolved here in this manner instead of just making one single proper ticket instead of like what 6 replies in random tickets or wrong departments or multiple tickets merged/closed is that your 2nd account that wasn't properly flagged is now also flagged.
Okay @Flying_Chinaman here are the final changes to address your biennial and oversell concerns. Above is "before" and below is "after" and we can give you credit for the changes if you desire, maybe Flying_Chinaman special. I know I joked about it in my last post but this is more realistic (and still not good in my opinion but it gives you what you wanted.)
For all plans, 2 year optional, but setup fee waived. 2 year price is 2x annual. Guaranteed no oversell on the RAM since we just cut it in half. CPU also reduced so we can basically guarantee same AUP. Disk stays almost the same and also guaranteed, not oversold but just to be careful reduced a bit. Bandwidth amounts lowered to guaranteed amounts at full port speed.
We might add customization option since you want us to bring that back, prices for some customization on top of above prices:
@VirMach said: @Flying_Chinaman said: Fined by me with the CPU oversell and bandwidth cap if the price was fair enough. But oversell mem and dik is a new level even for VirmArch
I have some bad news for you if you actually use Contabo and you believe they don't oversell.
Knew my disk IO / IOPS will be capped. but don't know their memory & disk oversell.
Main problems for me is their raised renew price, increasingly unreliability(CPU steal surge, network congestion and so on).
Okay @Flying_Chinaman here are the final changes to address your biennial and oversell concerns. Above is "before" and below is "after" and we can give you credit for the changes if you desire, maybe Flying_Chinaman special. I know I joked about it in my last post but this is more realistic (and still not good in my opinion but it gives you what you wanted.)
For all plans, 2 year optional, but setup fee waived. 2 year price is 2x annual. Guaranteed no oversell on the RAM since we just cut it in half. CPU also reduced so we can basically guarantee same AUP. Disk stays almost the same and also guaranteed, not oversold but just to be careful reduced a bit. Bandwidth amounts lowered to guaranteed amounts at full port speed.
Whaaagh, I meant simply annual bill without repercussion. So I could jump off the bus if grabbed GCV EPYC 6666 or things go wrong
Whaaaaaaaaaaaaaaaagh screw ya, that's the illusion of choices, how typical of ya!
Even MJJ with peanut sized brain will pick first one.
Comments
@VirMach Can you help me solve this problem? I can't do anything on your website, even the most basic tickets cannot be guaranteed. At least you should let VPS start. I paid the latest bill, but now I can't turn on the machine.
The ID of that VPS is: 634681
@deafness - Please click the "Management" button on the left menu shown in your screen shot above. When the next left menu appears, click "VPS Control Panel". That should open SolusVM, if the VM still shows offline in SolusVM, you can attempt to restart it. If it shows "unavailable" please report the Node name here.
Bring Back Customization
Offer expired, henceforth you are still chad @FrankZ the funny to me
Ontario Dildo Inspector
What "customization" is people talking about?
From the pre BF limited VirMach existing clients sale a few week ago.
Super early sneak peak at our Epyc plans:
https://virmach.com/cheap-high-ram-vps/
A lot may change on it when released (already thinking about lowering bandwidth amounts or some kind of clear fair use speed cap.) Let me know what location we should send the first one to (not Tokyo.) Disk is currently planned to be RAID1 (enterprise Gen3 7.68TB.) Processor is currently planned to be Epyc 7452 (2.35GHz.) IPv4 may or may not be included at base price. SolusVM may or may not be used.
The idea is lower disk amount, slower disk, slower processor, and just a boatload of RAM at a hyper-competitive price.
(edit) These are at MSRP. I wrote out a long since-deleted (or never posted) message about possibly being able to hit $15 a year special on the base plan (which was actually the 4GB about a week ago.)
Geekbench 5 could be up to 1000 but probably around 600 or so single threaded in real world use.
If your going to be offering 2 Cores, 4GB RAM, 30 GB NVMe, for $15/year, and increments upward from there. I think you should expect a lot of takers. I'll probably buy one depending only on if the location is not some where I already have big VMs long term.
#Bring Back CustomizationHoly moly, too good to be true, it worths 35 buck at GCV.
Ontario Dildo Inspector
ok, I'll admit you made me lol with that one.
I don' t need more RAM so I won't comment on merits here
But please fix the Purchase button, it's ugly, can't see shit, need some contrast - just steal the button above design.
Thanks! https://i.imgur.com/sPyXAml.png
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
Los Angeles! Seattle! San Jose! Any location near Tokyo
Ontario Dildo Inspector
Anyone has servers working on LAXA009 ? The node is currently locked for more than 2 days (not mentioned in the Network Status)
Yes, I have a VM on node LAXA009. It has also be showing locked in the billing panel, which as noted there means it is under going maintenance. Mine has been up and down over the last couple of days. Currently it has been up for the last 15+ hours.
Thank you, @FrankZ. Mine are now up too.
Seem you have server on almost nodes :-)
Will you consider SGP as a new location?
Leaving it locked for a little while longer before I call it "stable enough" and unlock controls.
@Virmach can we make priority ticket mentioning the RAM issue? thanks :'(
Looks interesting... definitely worth a buy when it comes out.... location, I prefer the US east coast... NY preferred
Would love to see the missing 64gb on the slider
Also, the slider is not very intuitive on dark mode on mobile.
@VirMach
Contabo replacement only if ticket E.T.A was 1 working day rather than 1 week
Ontario Dildo Inspector
First world problem:
I am running out of disk space on my mostly 'static files' VPS - 1 core, 1 GB RAM, 35GB disk. I could move my shit to bigger one... but those have 3+ cores and it it will total waste to host static files there... so I guess I am buying another one when we get "NOT Black Friday sale"
?
EDIT: Wait, I expected this years ago and bought some crazy thing on BF few years ago - 52GB disk, 512 MB RAM, 524 GB bandwidth - totally enough for my mostly 'static files' via CDN... But do I really want to move this shit for extra 17GB or disk?!
FirstWorldProblems
Haven't bought a single service in VirMach Great Ryzen 2022 - 2023 Flash Sale.
@VirMach All issues have been resolved and I have closed ticket #496617. Thanks again!
@FrankZ No, it does not work. When I click on the control panel, it shows the picture below
Okay so I've thought about bandwidth levels a little bit more, and it's going to change to this instead:
So it'll be 100Mbps Unmetered (32TB) and then a "full speed" (10-20Gbps) amount based on 100/10,000 = 1/100th of 1Gbps or 3.2TB (also known as 10%.) So what this means is we'll basically just do 10Gbps fair use, give 32TB for that plan for example, but reserve the right to cap it to 100Mbps after 10% usage. Depending on what software we use we'll most likely only want to measure outgoing only (incoming will be unmetered, otherwise we'll double the figures, so 3.2TB = 6.4TB.)
This also will scale with cost instead of RAM amount. Rounded off a bit to look nicer. This means:
Short version for BW: it means 48GB, 64GB, and 96GB plans have their bandwidth reduced a bit and we just defined the "fair use" part for the bandwidth levels.
I added a Fair Use Guidelines section which basically covers the level of overprovisioning. Short version is that IPv4 will basically be dedicated unless otherwise noted, it covers the bandwidth amounts I mentioned above, disk space is currently planned LVM but may be thin LVM, physical RAM will be oversold 2:1 ratio, and CPU will be oversold about 8:1 ratio. Seems to be "industry standard" for these type of plans and we'll of course manage it responsibly and cut it off early if alerts go off.
Everything is technically worked out to where we can lock everything off at about half the amount, as in 0 oversold RAM, and 4:1 ratio for CPU instead, and we don't take a loss (still profitable.) So anything in between that and max is just extra profit and will be done cautiously as a bonus.
I might actually end up doing Tokyo first just to see what worst case scenario of this build ends up looking like for us. But if we do any specials for it, I'll be sure to advertise it per the 10% fair use. So for example, a "special" version of 2GB plan might look like:
$2/MO $30/2YR
2GB RAM
15GB RAID10 NVMe
200GB @ 10Gbps / 2TB @ 6.4Mbps (Perhaps a forced selection, but then we might just add like 1TB @ 1Gbps option too.)
1 Core @ 25% Special fair use
1 IPv4 & /64 or /48 IPv6
Might as well make this post longer...
Also means probably for same theoretical Tokyo special that may or may not exist:
Probably no setup fee unless it gets annoying.
Working on finally resolving this. As always not required but would be extremely helpful if people stopped making like 3 tickets for their $1.50 pending order.
ETA around December 30th it looks like for everything to be normal again.
(edit) And yes this is very much aimed toward being a Contabo replacement and I want to make sure people understand this. It means similarly oversold but we're going to obviously try our best to make it better. As in, more idlers!
Looks like another issue, if you used the button and it broke make sure to mention that in ticket
Yes. I kind of want to do Singapore and Minneapolis.
UK? Somewhere, anywhere - London, Manchester, Aberdeen (that'd be a major coup!)
In stasis until the shitposting stops/abates.
Than=compare;then=sequence:brought=bring;bought=buy:staffs=pile of sticks:informations/infos=no plural.
It wisnae me! A big boy done it and ran away. || NVMe2G for life! until death (the end is nigh).
Here I'll list of all the locations I've thought about, but it's definitely not going to go as fast as we want and we definitely have to end up skipping some of these locations until things get back on track.
North America:
APAC:
Europe/Middle East:
South of the Equator:
Maybe: (But too close to an existing location.)
Maybe: (But meh.)
Maybe: (It'd just be cool to have but probably not a good idea, an extension of my Hawaii dreams.)
Maybe: (Haven't really thought of it but maybe this is the ultimate mediocre for everyone location.)
I miss my UrPad Iceland server. Man that was a good server.
The Yeti has left the building.
We have South Pole Colocation.
You get 100% off if you reserve a whole rack.
We accept Karma donations for the last flan. 🍮 affbrr
If there's any recommendations for serious datcenters there it could be a reality assuming the whole lower power costs thing they have going there if it hasn't been "ruined" by an influx of crypto stuff (aka retaliatory regulations.) Since we could just ship off super beefy stuff there.
This is just me making a wild guess though but I'm assuming Iceland would be one of those places where the citizens have access to 1Gbps fiber for like $20-50 a month but 1Gbps at a datacenter might cost $5,000 a month.
Sorry but we're going to have to go with a more serious south pole hosting such as a US astronomy research center.
Preparing the 882 page whitepaper for them as we speak on why we need colocation there.
Sorry no idea man. I just remember the server was in Iceland. Never really looked into coloing anything there. I would imagine as you said though.
The Yeti has left the building.
You'd please a lot of folks, with an "off-shore" location i.e. Hawaii, isle of Man, Guernsey, Cook Islands, Malta, Fuerteventura.
You may even get a lot of new recruits, for hands-on approach!
In stasis until the shitposting stops/abates.
Than=compare;then=sequence:brought=bring;bought=buy:staffs=pile of sticks:informations/infos=no plural.
It wisnae me! A big boy done it and ran away. || NVMe2G for life! until death (the end is nigh).
I'm kind of liking this idea, probably until I see the prices. It's too late I have to commit now, I'm the proud owner of VPShore.com and islev.ps (not to be confused with v.ps.) Actually just kidding I didn't register islev.ps even though iSlev sounds kind of cool, we don't have the kind of budget for $59 a year domains but I did also grab isle.host to add to my C-tier list of domains we'll never use. Then later we'll buy the PREMIUM (apparently) domain hostisle.com for $2500 because it sounds/looks like hostile and register some shady company name like Hostile Host Pty for our ASN because it'd match all the abuse reports we have to end up dealing with and make it easier for company to blanket blocklist.
Spain --> Canary Islands instead.
France --> New Caledonia instead.
UK --> Isle of Man & Gibraltar instead (I can't pronounce Guernsey or maybe don't like how it comes out so we'll skip that.)
Okay like only 90% of that was a joke, we'll see how far the other 10% gets us.
are the TYO maintenance still taking place or have they been completed already?
I bench YABS 24/7/365 unless it's a leap year.
Maybe Jersey is more appropriate, given how frrreezing it is here, this week!
Guernsey sounds very like girning - a Scottish word for a teething, whining bairn (baby).
In stasis until the shitposting stops/abates.
Than=compare;then=sequence:brought=bring;bought=buy:staffs=pile of sticks:informations/infos=no plural.
It wisnae me! A big boy done it and ran away. || NVMe2G for life! until death (the end is nigh).
Fined by me with the CPU oversell and bandwidth cap if the price was fair enough.
But oversell mem and dik is a new level even for VirmArch
No more 2 years commitment for an experimental oversell plan. BRING BACK ANNUAL BILL
Only wonna pay biannually for storage plan and GCV EPYC 6666
Hope my remain days at VirmArch does not end up like the early-stage CloudCone and shitty PacificRack.
The promised land for Flying Chinaman, also not near Seattle at all
bring back customizationOntario Dildo Inspector
@VirMach said: I don't know how I'd feel about ever having to fly to Australia. Maybe New Zealand.
How do you feel about 17 or 18 hour non-stop flights? Plus connections if you're not on the west coast. There's also 19 or 20 hour flights from Dallas.
+1 for Dubai, Singapore.
+1000 for Kochi. I'm from Kochi.
The Ultimate Speedtest Script | Get Instant Alerts on new LES/LET deals | Cheap VPS Deals | VirMach Flash Sales Notifier
FREE KVM VPS - FreeVPS.org | FREE LXC VPS - MicroLXC
+1 for São Paulo, let's drink a coffee, It’s on me!
It's even worse on the backend. Whoever made it (not me) did very poorly. But that's how it'll be... it's a feature! You're not supposed to put the slider that far, if you do then it's just a failsafe so the slider doesn't break off.
Okay this part I can admit I don't care enough to fix, I'm anti-anti-blinding lights.
@VirMach But my account can't send you a ticket, as I mentioned before when I send a ticket the button is not available, I made sure I checked the button where I agree to pay, so now I can't send a ticket on your site to let you know the problem and I can't use the vps, I hope you can help me to solve this problem, it's been 4 months since this problem.The id of this vps is 634681
And on your site, I use Troubleshoot with this result
I also don't see any notification that CHIZ002 is offline on your Server Status page.
I don't mean to get political but I guess I will, just a tad. I wouldn't feel very welcome based on the last couple years. I know that's not the best thing to base it off but even getting detained or whatever you want to call it, for a few days, would be an absolute no-go for me right now after you add in the flight times. And then next time something similar happens, trying to figure out how to plan travel would already be a nightmare but doubly so it seems with that region since they obviously are an island nation and want to be cautious. As far as islands go though I think we'd have better luck with all the others since I assume they rely more on tourism (might as well add in that as far as the semantics of islands go, everything would technically be an island then but it has that vibe so I apologize if I'm using the wrong term.)
On top of that, we would also have to deal with the high power costs, and high bandwidth costs, and high labor costs, except it seems like there'd be no upside since no one really outside of Australia benefits (and again, that'd be like saying "no one outside of the US benefits from US hosting" except at least Tokyo comes with the better connectivity to pretty much everywhere in that region.)
I will say though I wasn't even thinking about the flight being that long, for some reason in my head it was the same as going to Tokyo even though that's wrong and even then I already had that mindset.
TLDR if I get a signed and backed by gold pass from the prime minister of Australia that I can come whenever, maybe an ounce of gold every day it's delayed due to government policies, I'd be more keen.
I have some bad news for you if you actually use Contabo and you believe they don't oversell.
This isn't as true as you think on so many levels but I'll spare you the long read. I will say something though, outside of any pretty much "guaranteed loss" low quantity plans, any KVM Ryzen host that has up until today sold any 1GB / 1 IPv4 / plan for less than $17 per year is either (1) overselling memory to some level, (2) lying to you with they claim otherwise, or (3) going to genuinely go out of business. Remember I said outside of "guarantees loss" low quantity, as in a provider can obviously do this if they just make a lot of money from full-price plans and have extra space (except if they have extra space it's most likely from selling regular plans, and the node is full, except 128GB assigned RAM equals like 40GB actually used memory.)
Even someone who doesn't know what they're doing can easily oversell RAM on KVM.
And what you said about guest OS and cache, it's just not how it works but it's complicated. We'd have to do a very long lesson that no one will read.
We have 100% oversold memory when we switch from 256GB dual E5 to 128GB Ryzen. There's was other way to do, but we carefully measured everything out server by server before moving. Those servers are pretty much perfect usage levels. Where we did screw up was for Tokyo Ryzen, due to how the semi-automated system activated orders, there's one or two nodes that ended up with too many high RAM plans and even then our highest, out of ALL servers, in terms of memory usage, is around 80GB out of 128GB actual RAM used, with around 8GB free no matter how you calculate it, even with cache usage, etc. And that node has been stable for half a year, 100% uptime, and closely monitored.
Trust me this isn't the long version.
Where did you get the 50% disk oversell from? We definitely wouldn't oversell disk by that much, that's where I would agree with you that it's dangerous. One of the reasons we did KVM Lite the way we did instead of thin LVM was to avoid it altogether. So I guess think of it this way instead, what we could do is do 15GB dedicated, or where you start at 10GB and can extend up to 20GB. Something like that.
And like it says on the page what's advertised for disk is 100% guaranteed. So if you run out of disk in that scenario, and the server runs out of disk, which would be rare, we'd move you to one with disk and give you the 20GB.
I think you underestimate how many people don't use their disk. When we do backups and anything related to grabbing the actual disk sizes, it ranges from maybe 30-65% actual usage, average.
But anyway, I don't think we'll be doing thin LVM unless we go with a panel that does a really great job and even then I don't think we'd actually be overselling it.
Okay just because you said this if we do Epyc Tokyo, it's going to be 3 year for the 2GB. That's a good idea. $45/3Yr. Then 4GB $40/2YR maybe, and if you want 1YR and don't want to be part of the experiment, 8GB will be 1Yr still.
Or we could do 2GB for $15/YR plus $15 setup fee instead of $30/2YR if you want. Let me know. We can do whatever you want, but it will add up to $30. Even 6 months for $7.50/6Months + $22.50 setup.
I almost went there maybe a year or two ago, it's hard to tell how time has transpired recently. But I didn't probably because of my stupid jobs kicks trashcan and frowns while crossing arms.
Any recommendations for DC/network blend?
You're facing at least two issues right now simultaneously. Then a third issue is that you have multiple accounts, and a fourth issue is that you're flagged for any of the following, one or more: (1) trying to force migration to Tokyo but it's very close to the cutoff time for definitely so maybe not, (2) creating tickets about tickets and making migration request and another one about your initial issue that did get fixed at some time, and ISO request, and migration request all together while another request was still being fulfilled, and then doing it again after you specifically got a message saying not to do that. And then ticket requesting refund for failed migration while opening more migration requests. And then replying to a ticket about your IP changing that says not to use this ticket for other issues, and you use it for other issues. Oh and you had two tickets for the first issue I think, and you didn't follow the guidelines. Then you replied to two tickets like that where you also ignored the guidelines being re-iterated. Yeah looks like it's definitely for #2. Oh looks like you have another ticket open you replied to same thing that's completely also unrelated.
As for your current two main issues not including the multi-account or ticket creation issue (which isn't an issue, you're just not allowed to use priority department anymore)
Anyway I just went ahead and moved you to the staging area, you need to use redeploy button now.
If you make any more tickets about this or talk about it here, the best I can do for you at this point is just a pro-rated refund and cancellation maybe.
The price for getting your issue resolved here in this manner instead of just making one single proper ticket instead of like what 6 replies in random tickets or wrong departments or multiple tickets merged/closed is that your 2nd account that wasn't properly flagged is now also flagged.
where is the link please
I’m a simple man I see gifs, I press thanks
Okay @Flying_Chinaman here are the final changes to address your biennial and oversell concerns. Above is "before" and below is "after" and we can give you credit for the changes if you desire, maybe Flying_Chinaman special. I know I joked about it in my last post but this is more realistic (and still not good in my opinion but it gives you what you wanted.)
For all plans, 2 year optional, but setup fee waived. 2 year price is 2x annual. Guaranteed no oversell on the RAM since we just cut it in half. CPU also reduced so we can basically guarantee same AUP. Disk stays almost the same and also guaranteed, not oversold but just to be careful reduced a bit. Bandwidth amounts lowered to guaranteed amounts at full port speed.
We might add customization option since you want us to bring that back, prices for some customization on top of above prices:
+1GB dedicated RAM, +$6/YR
+1 Core, +$10/YR
+1TB bandwidth, +$10/YR
Customization prices are just estimates, depends on how we integrate it.
Knew my disk IO / IOPS will be capped. but don't know their memory & disk oversell.
Main problems for me is their raised renew price, increasingly unreliability(CPU steal surge, network congestion and so on).
Whaaagh, I meant simply annual bill without repercussion. So I could jump off the bus if grabbed GCV EPYC 6666 or things go wrong
Whaaaaaaaaaaaaaaaagh screw ya, that's the illusion of choices, how typical of ya!
Even MJJ with peanut sized brain will pick first one.
Fine... I shall be your lab rat.
Ontario Dildo Inspector