I made this thread so that people can say what they want about VirMach's service. That includes the good, the bad, and the ugly. You want to say the same thing ten times, knock your socks off. I don't know how much good it will do, but go right ahead.

Where I am drawing the line, for both sides, is personal attacks. The people who think VirMach is the worst host in the world can voice that, but let us try to reasonable in how we express that. If you attack a subject that is fine. If you attack on a personal level I am getting less comfortable with that as time goes on. This also goes for the people that think VirMach is ok as well. Let's stop attacking the people that don't, no matter how wrong you may think they are. We are all members of the same community let's figure out a way to act civil to each other even if we don't agree on this particular subject.

If you make an personal attack on another member that in my opinion crosses the don't be a dick line, your comment will be removed.

The current state of things VirMach

The below is not to be taken as official VirMach policy and is just provided as a way to assist members that do not want to try and find the related information in a long thread. I will try to keep this up to date. If you find obvious errors, please DM me.
For official VirMach policy please see the VirMach website.

PayPay Charge Backs

Under no circumstances should you initiate a charge back, or dispute a payment, through PayPal unless you understand that this action will cause your account with VirMach, and all the services in it, to be suspended permanently. You will also no longer be able to purchase any new services with VirMach in the future. IMO Just don't do it.

Flash sale provisioning

If you have purchased a VirMach flash sale VPS, it has not been provisioned, and still shows "Pending" in your billing panel screen. You should wait for your VPS to be provisioned. Opening a ticket regarding a pending VPS and/or posting your ticket number in this thread will probably lead to your VPS purchase being canceled and the purchase amount being refunded to account credit.

UPDATE

@VirMach said on Jan 25th, 2023:
Thank you everyone who was patient and sorry for the last bit of these flash specials taking so long to activate. I finally got some functional code to be able to activate these efficiently so that's why they were done quickly today. It's not that we were ignoring it, or had any bad intentions, it was just several failed iterations of solutions being worked on in the background.

Anyway, all* flash specials are provisioned now. There's still vserverid issue for most of the recent activated ones, which just means you won't be able to access the full SolusVM control panel yet, everything else should work. And some activated today are still waiting on due date to be adjusted. So please do not make tickets for these two issues.

*A little exception to this, the following have not been provisioned:

• 12 quantity for DALZ010 offer, 4GB RAM / 4 Core / 130GB Disk
• 1 12 quantity for NYCB041 offer (node is having problems still)

However, at this point in time, if you already haven't, feel free to contact us if:

1. There's any missing orders, make sure to provide all details such as invoice ID and transaction ID, so we can refund you
2. Yours is paid and still pending or in a weird halfway state like cancelled but paid, etc.
3. Incorrect specifications provisioned (these are super rare, maybe 1 or 2 out of each 1,000 so please make 100% sure first.)

For #3 if we get dozens of tickets claiming this, they'll all just be closed because it means people are reporting it improperly in which case you'll have to use LES support instead.

Regarding $0.25 3 Month VPS

VirMach said on Jan 15th: The $0.25 orders, there's still hundreds left. So far we got one $0.25 chargeback and maybe a hundred tickets from people asking for updates. Already covered but we didn't expect nearly 1,000 orders all in Tokyo with like half of them multiple accounts. I mean I guess I could say we did expect that as a possibility and it happened, so these are just being mass refunded. It's taking longer because I'm actually double flagging down a lot of people for their behavior here, like signing up for multiple accounts specifically to order a bunch of the $0.25 specials is obviously not okay and creating 3-4 tickets screaming about it is also something we're going to make sure sticks with your account in the form of a note.

Migrations

Currently paid migrations, with and without data, are not available between VirMach locations using the migration button in the billing panel. Although there has been no official statement by VirMach regarding this, I expect that after recent sales have been provisioned, and/or new nodes are brought on line this option may return for some locations. VirMach has stated that migrations to Tokyo will not be available until after more nodes arrive at the DC and are setup. There has not been any date given for when this should be expected.

rDNS aka PTR DNS records

Currently rDNS is not available in any location. There has been no date given by VirMach as to when rDNS would be available in the future. Consequentially members should not expect rDNS to be available from VirMach.

TOS Guide Lines

@VirMach said on January 23rd, 2023:
Easiest answer is if you have 1% or the node's dedicated disk amount, you get 1% of all I/O operations over like a day. That's not what it is, but the easiest rule of thumb to communicate. So 20GB on 2TB drive = 1%

2TB NVMe drive Gen3/Gen4, look up what it's capable of for the type of operations you do.

So if it can do like 500,000 IOPS for your operation type/size, then that's 5,000 IOPS in the example of having 1% share of the disk, there are 86,400 seconds in a day so 432,000,000 operations a day for that example of that specific type of operation. Anything more, we may decide to be lenient to some level where possible automatically.

Most cases if you get that message you're doing a lot more.

And of course over a shorter period of time, let's say 10 minutes out of 24 hours, there are 144x 10 minute chunks in 24 hours so let's say you do 10x more than that, it's fine as long as it's not more than 14x 10 minute chunks that do 10x more since any more and you'd average out more for the day. However, if you do 2 x 10 minute consecutive chunks where you're doing like 50x more that could still be problematic since at that point you're most likely also breaking the load and CPU fair use policies. Basically if any of that is confusing just refer to my first sentence and do your own math/conceptualization of it.

FAQ About VirMach Issues

Q. Is there a workaround for 'Invalid vserverid or username' when trying to access the SolusVM panel ?
A. Currently there is no workaround for this issue. It must be resolved by VirMach. Previous workarounds no longer work due to changes made by VirMach.

Q. 2FA is not working and I keep needing to use the backup code to get in to the billing panel. Why is your 2FA broken ?
A. Unsubscribe from 2FA and then logout. Log back in to the billing panel and setup 2FA again. Make sure to make a note of your new backup code.

Official Virmach links & comments that may be of interest.

Ryzen Server Status Page
Terms and Conditions, AUP, Service Agreement, Fee Schesdule, and Privacy Policy.
Looking Glass & Locations
FAQ
Refund policy and procedure
Billing Assistance
Common issues
Third party issues

Members comments that may be of interest

Mason said: VirMach is a stubborn, masochist

I received an abuse ticket (Ticket ID: 203275), and the situation was beyond bizarre. The ticket was about an IP address (142.147.98.21) that wasn't even associated with my server! In fact, it was a server IP that had already been canceled, and I had never hosted any site on it. The ticket came with a message from Carter Kirsty Ann from Merseyside Police, which raised huge red flags. And as if that wasn’t bad enough, on 29/03/2026, just hours after I opened the ticket, I received a notification that my service had been terminated by the police department’s demand.

But here’s the kicker: They didn’t give me any heads-up, no time to fix anything, no notice of the termination. Just bam, it was gone, and I was left scrambling. I’ve paid for a full year, and now my server is completely shut down without any explanation or recourse.

Now, let’s talk about their Acceptable Use Policy (AUP), which is clearly a joke. According to heymman.com’s AUP, each customer is responsible for the activities of their users and must ensure that their users/representatives abide by the policy. It goes on to say that if any violations occur, heymman.com has the right to take action without notice to stop the offending customer. I understand the importance of following AUPs, but heymman.com didn’t even follow their own terms here. They acted without providing notice, and without any evidence of violation from my side, my service was abruptly terminated.

And don’t think the mess stops with heymman.com. Their parent company, UnReal Servers LLC, is just as bad. It’s clear they don’t care about their customers or providing any sort of legitimate service. This is beyond unacceptable. A legitimate business would never treat its customers this way.

I’ve tried reaching out to both companies, but it’s like talking to a brick wall. I can’t recommend this service to anyone. Stay far away from them unless you want to risk losing your server without warning or explanation. Avoid heymman.com and UnReal Servers LLC—they are a disaster waiting to happen.

HostCram review — FiberState (over two years in)

It’s been over two years since we signed a colocation agreement with FiberState (we started in November 2023), and I can say with confidence that this is one of the best vendor relationships HostCram has had. We run only custom servers, ship racks of gear, and push our providers hard — FiberState has handled all of it with competence, patience, and real care.

Shipping, intake, and hardware handling

We shipped a lot of hardware to the data center and never had anything missing or damaged. The intake process is clean and professional; they treat gear like it matters because it does. When we needed to RMA parts, FiberState made it easy to ship components out from the DC via UPS to the manufacturer. On the rare occasions we didn’t have spares, they offered replacement hardware at cost, which kept services running and saved us from long outages.

Network help and technical expertise

Their network team deserves special mention. We brought in a Juniper QFX, and I didn’t know how to get it online or build VLANs the right way. FiberState walked us through the setup, helped configure VLANs, and assisted with our custom BGP sessions. That kind of hands-on network expertise — not just hand-waving, but actual configuration and testing — is rare and extremely valuable for a shop running bespoke infrastructure.

Remote hands, emergency support, and people

Support is where FiberState really stands out. Remote hands are excellent: small, quick tasks are often handled without charge, and longer jobs are billed at $100/hr, which is fair for the level of skill and responsiveness we get. When a server went offline, we had emergency troubleshooting and escalation available immediately. There’s one person I always call out: Dan — whenever I needed something done, I could count on him to get it done, no drama.
We’ve been tough customers — custom servers mean oddball requests and a lot of pressure on support — and the team stayed professional and generous throughout. That professionalism under stress is a huge differentiator.

Culture, value, and why we’re staying

Beyond the technical wins, FiberState’s team has been generous and relationship-focused. They consistently choose a long-term partnership over nickel-and-diming small issues. That attitude shows up in free quick remote hands, practical help with RMAs, and stepping in with hardware at cost when needed.
After more than two years of real-world use, HostCram has decided to stay with FiberState for the long haul. They’ve earned our trust through reliability, technical competence, and plain old good service.

Final verdict

If you need a colocation partner who treats your custom gear seriously, provides real network expertise, responds quickly in emergencies, and backs it up with practical, no-nonsense support, FiberState is an excellent choice. Highly recommended by HostCram.

Note: FiberState affiliate links are in use. Non-affiliate link.

Some polically incorrect puns I couldn't resist making:

• Enough free space - won't get quickly occupied like Poland.
• Safe and stable - doesn't fall as quickly as France!

Anyone using Filen.io (my affiliate link that should give +10 GB for free)?

It's nice to see the option of lifetimes, but not going 100% on selling lifetimes (so it just might survive for more than a few more years - hopefully).

I saw some older LES threads, but no recent experience shared.

Has anyone bought any Filen.io lifetime deals, or used monthly subscriptions?
How is the speed, stability?

I’m sharing my experience with HostDize so others can make an informed decision.

My VPS services were suspended without any prior notice, warning, or request for clarification. There was no communication about:

any policy violation,
any complaint on my IP,
or any activity that allegedly triggered the suspension.

When I contacted support, I was immediately demanded full KYC documents (passport, photograph, address proof, etc.) before they were willing to explain why my services were suspended.

In fact, I had recently purchased a 1-year annual VPS from the Blackfriday deals this year!

I was never asked to share my personal documents with the host while I was ordering and making the purchases. In the interest of transparency, I've been trying to understand from HostDZire.

what triggered the suspension,
whether there was a complaint or abuse report,
and the basis for suddenly demanding personal documents now, after arbitrarily suspending my services without any reason whatsover.

These questions were not answered.

HostDize has publicly shared a 99% blurred image succumbing to pressure (no clue if it's real or AI generated; Link: https://i.postimg.cc/c4Nsz2zt/image.png ) stating that they'd received that 'complaint' from the law enforcement agencies.

No where can you make out anything, thanks to the blur. And now HostDzire demands my private information, under the garb of a blurred image. They don't need my KYC to serve me the complaint copy, if it's legit.

• What stops them from sharing the so-called complaint to me on my registered email address?
• Why are my other services suspended?
• India's been prey to many fake cyber-fraud notices. And in the era of AI, it isn't a big deal.

Again, no answers.

This creates a concerning situation:

You pay for a service.
Service gets suspended before anything else
No notice or explanation is given
Customers are pressured to submit sensitive personal documents
Transparency is withheld until compliance

At no point was proportionality observed. Even basic, non-sensitive information (like “there was an abuse report” or “policy X was triggered”) was refused.

I understand KYC requirements where legally mandated - but suspending services first and refusing to explain anything unless personal documents are handed over is not acceptable.

Prospective customers should be aware that:

Your services can be suspended without warning
You may not be told why
You may be required to submit sensitive KYC documents just to get an explanation

I’m posting this as a factual account so others can make an informed decision.

Thank you!

Horaco - ZX-SWTGW2C8F - likely being sold under other brands too

https://www.aliexpress.com/item/1005006765378093.html

Good

• Inexpensive - paid 95 USD / 72 GBP
• Switching it to English is idiot proof
• Doesn't feel cheap - normal metal consumer level switch case, decent looking power plug with UK pins (no janky adapter)
• Fanless
• Wide compatibility. Thus far tried 2.5G RJ45 transceiver, 10g DAC and SFP+ 10g fiber
• Reports temps/watt on adapters where available
• Web interface feels functional, though somewhat spartan. Haven't seen anything broken yet
• Has SNMP - new to me but looks interesting

Bad

• Mine shipped in 1G/10g mode so thought its broken cause my 2.5G RJ45 transceivers couldn't reach web interface, meaning you need 10G to enable 2.5G
• Saw forum mentions of there being a hidden factory account - no idea if true. Doesn't seem to have any online features so I just skipped setting gateway and blocked it at FW for good measure.

Ambivalent

• No dedicated management port
• No rack ears/holes
• UI is pretty spartan - not much explanations / guidance
• Manual included seems to be for 2.5G/10G model
• Struggled to change management IP - likely user error
• Mixed messaging on what exactly 1G mode it supports - some places refer to 1000M others to 1.25G

LET $7 KVM Super Sale NY
  30 GB SSD Storage
  2 GB RAM
  1 vCPU Core
  2 TB Bandwidth
  1 Gbp/s connection
  1 IPv4 Address
  KVM Virtualization
  Promotional Price: $7.00/Year

The purchase was quick, no upselling, setup almost immediate once the payment went through. Overall, flawless shopping experience.

The server itself came with both IPv4 and IPv6. At that low price, it is certainly something... The uptime, for the nearly 1 month I had it for, was great. The server did become unreachable once during that time, but reaching out to support fixed it quickly. Great work @DediRock

Now on to the bad...

I was planning to use this as one of my image dumps for the imagehost I have, however the server's fluctuating network speed leaves a lot to be desired. I am based in Singapore, with a 2.5Gbps internet, and for me, just a 1 MB image takes 3 seconds to load from this server. Do note that this speed keeps fluctuating up and down, so first image may load immediately, but thæn the next one will be slow and not loading...

Ofcourse your results may vary depending on your location, so here are some speed test files hosted on the server for you to try yourself:

http://192.210.217.184/100MB.bin
http://192.210.217.184/1GB.bin

(they may or may not have been copied from hetzner's speed test site )

Other thæn that, when running ping test (again, from Singapore), the server drops packets every so often.

Pinging 192.210.217.184 with 32 bytes of data:
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=174ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Request timed out.
Reply from 192.210.217.184: bytes=32 time=201ms TTL=51
Reply from 192.210.217.184: bytes=32 time=174ms TTL=51
Request timed out.
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51
Reply from 192.210.217.184: bytes=32 time=173ms TTL=51

Ping statistics for 192.210.217.184:
    Packets: Sent = 15, Received = 13, Lost = 2 (13% loss),
Approximate round trip times in milli-seconds:
    Minimum = 173ms, Maximum = 201ms, Average = 175ms

Finally, some YABS:

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-04-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Thu Oct 30 12:35:14 AM UTC 2025

Basic System Information:
---------------------------------
Uptime     : 10 days, 21 hours, 23 minutes
Processor  : Intel(R) Xeon(R) CPU E5-2660 v2 @ 2.20GHz
CPU cores  : 1 @ 2199.998 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 1.9 GiB
Swap       : 2.0 GiB
Disk       : 29.4 GiB
Distro     : Ubuntu 24.04.3 LTS
Kernel     : 6.8.0-85-generic
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : ColoCrossing
ASN        : AS36352 HostPapa
Host       : ColoCrossing
Location   : Buffalo, New York (NY)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 26.63 MB/s    (6.6k) | 323.88 MB/s   (5.0k)
Write      | 26.65 MB/s    (6.6k) | 325.58 MB/s   (5.0k)
Total      | 53.29 MB/s   (13.3k) | 649.47 MB/s  (10.1k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 473.98 MB/s    (925) | 566.76 MB/s    (553)
Write      | 499.17 MB/s    (974) | 604.50 MB/s    (590)
Total      | 973.15 MB/s   (1.8k) | 1.17 GB/s     (1.1k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 206 Mbits/sec   | 52.9 Mbits/sec  | 141 ms
Eranium         | Amsterdam, NL (100G)      | 328 Mbits/sec   | 343 Mbits/sec   | 139 ms
Uztelecom       | Tashkent, UZ (10G)        | busy            | 148 Mbits/sec   | --
Leaseweb        | Singapore, SG (10G)       | 354 Mbits/sec   | 338 Mbits/sec   | 177 ms
Clouvider       | Los Angeles, CA, US (10G) | busy            | 413 Mbits/sec   | 0.432 ms
Leaseweb        | NYC, NY, US (10G)         | 314 Mbits/sec   | 298 Mbits/sec   | 1128 ms
Edgoo           | Sao Paulo, BR (1G)        | 2.08 Mbits/sec  | busy            | 177 ms

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
-----           | -----                     | ----            | ----            | ----
Clouvider       | London, UK (10G)          | 12.1 Mbits/sec  | 194 Mbits/sec   | 142 ms
Eranium         | Amsterdam, NL (100G)      | 301 Mbits/sec   | 455 Mbits/sec   | 140 ms
Uztelecom       | Tashkent, UZ (10G)        | 335 Mbits/sec   | 152 Mbits/sec   | 232 ms
Leaseweb        | Singapore, SG (10G)       | 4.15 Mbits/sec  | 273 Mbits/sec   | 196 ms
Clouvider       | Los Angeles, CA, US (10G) | 716 Mbits/sec   | 506 Mbits/sec   | 0.661 ms
Leaseweb        | NYC, NY, US (10G)         | 259 Mbits/sec   | 314 Mbits/sec   | 72.7 ms
Edgoo           | Sao Paulo, BR (1G)        | 22.2 Mbits/sec  | 148 Mbits/sec   | 173 ms

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 310
Multi Core      | 294
Full Test       | https://browser.geekbench.com/v6/cpu/14741419

YABS completed in 45 min 24 sec

Tier.net is pleased to introduce a dedicated thread for customers (or potential customers) and Tier.net community members. This thread will serve as a central place for sharing official Tier.net updates, announcing new services and improvements, receiving requests and feedback, and engaging in open discussion.

Our goal is to provide a transparent, organized, and easily accessible channel to keep you up to date with our activities, regardless of promotional or sales communications. We encourage you to use this thread not only for submitting questions and suggestions but also for general conversation with other Tier.net community members.

We strive to respond to your questions and concerns promptly and professionally, and we look forward to creating a productive and collaborative environment here.

Don't worry – you can also share your experiences with our company or ask technical questions – we are happy to help. If you believe that some of our offerings could be improved or are uncompetitive, please contact us. We strive to improve based on customer feedback.

My Experience
I’ve been using my VM primarily for BGP tunneling to my home network, allowing me to utilize my /24 IPv4 and /48 IPv6 subnets seamlessly. Even during their infrastructure migration, the downtime was minimal—just a brief interruption, which is impressive given the scale of such operations.

The support team at No Ack Hosting has always been quick, professional, and knowledgeable, which is rare to find. Whether it’s answering technical questions or resolving minor hiccups, they’ve been consistently reliable.

Service Highlights
Currently, I’m running the KVM VPS 1G BGP, which has been perfect for my needs. Its specifications include:

CPU: 1Core
RAM: 1GB (1024 MB)
Disk Space: 15GB SSD
IPv4: 1 Address
IPv6: /64 included (expandable to /48)
Network Interface: 100Mbps (Upgradeable if needed)

Finding a good BGP-enabled VM in Northern Europe—especially in Stockholm—is a challenge, but No Ack Hosting delivers where others fail. Their reliability, combined with the ability to customize BGP support, makes them an excellent choice for networking enthusiasts like me.

I’ve also noticed they now offer a smaller VPS option, which is great for users with lighter workloads. The specs for this smaller VPS are:

CPU: 1 Core
RAM: 256 MB
Disk Space: 10 GB SSD
IPv4: 1 Address
IPv6: 1 Address
Network Interface: 100Mbps

It’s a fantastic entry point for anyone looking to get started with a VPS at an affordable price.

Final Thoughts
No Ack Hosting has been around since 2018 (or earlier?), though I only became a customer in 2023. In this time, they’ve proven to be a reliable, professional, and high-quality hosting provider. If you’re in need of a VM, Dedicated Server, or Co-Location (from 1U to a full cabinet) in Stockholm, I highly recommend reaching out to them.

This review comes from genuine satisfaction with their services and the peace of mind they’ve given me. It’s hard to find a trustworthy provider in forums like LowEndTalk, where “summer hosters” often dominate, but No Ack Hosting is a refreshing exception. If you’re searching for reliability, excellent support, and great uptime, they’re the provider to choose.

Don't be intimidated by their website; their primary focus is on infrastructure and servers.

----------------------------------
----------------------------------

-YABS FROM MY BGP VM-
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Yet-Another-Bench-Script
v2025-01-01
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##

Tue 28 Jan 2025 07:20:04 PM CET
Basic System Information:

Uptime : 4 days, 3 hours, 17 minutes
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 1 @ 2799.998 MHz
AES-NI : ❌ Disabled
VM-x/AMD-V : ❌ Disabled
RAM : 976.6 MiB
Swap : 975.0 MiB
Disk : 13.7 GiB
Distro : Debian GNU/Linux 11 (bullseye)
Kernel : 5.10.0-21-amd64
VM Type : KVM
IPv4/IPv6 : ✔ Online / ❌ Offline
IPv4 Network Information:

ISP : No ACK Group Holding AB
ASN : AS30893 No ACK Group Holding AB
Host : Bgpcustnoack
Location : Stockholm, Stockholm County (AB)
Country : Sweden
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
Block Size  4k (IOPS)   64k (IOPS)
Read    23.72 MB/s (5.9k)   283.15 MB/s (4.4k)
Write   23.74 MB/s (5.9k)   284.64 MB/s (4.4k)
Total   47.47 MB/s (11.8k)  567.79 MB/s (8.8k)

Block Size  512k (IOPS)     1m (IOPS)
------  --- ----    ---- ----
Read    767.71 MB/s (1.4k)  765.00 MB/s (747)
Write   808.50 MB/s (1.5k)  815.95 MB/s (796)
Total   1.57 GB/s (3.0k)    1.58 GB/s (1.5k)
iperf3 Network Speed Tests (IPv4):
Provider    Location (Link)     Send Speed  Recv Speed  Ping
Clouvider   London, UK (10G)    4.20 Gbits/sec  4.34 Gbits/sec  --
Eranium     Amsterdam, NL (100G)    3.00 Gbits/sec  4.61 Gbits/sec  --
Uztelecom   Tashkent, UZ (10G)  1.24 Gbits/sec  373 Mbits/sec   --
Leaseweb    Singapore, SG (10G)     976 Mbits/sec   954 Mbits/sec   --
Clouvider   Los Angeles, CA, US (10G)   1.02 Gbits/sec  1.08 Gbits/sec  --
Leaseweb    NYC, NY, US (10G)   1.66 Gbits/sec  1.91 Gbits/sec  --
Edgoo   Sao Paulo, BR (1G)  893 Mbits/sec   763 Mbits/sec   --

• Frankfurt Intel Xeon: 99.9156%
• Paris AMD Ryzen: 99.7938%
• Paris Intel Xeon: 99.8536%

They don't have a looking glass AFAIK, but I'm hosting one in Frankfurt and Paris (mainly because I was bored when I coded it).
If I forgot to mention an important aspect, feel free to ask.

TL;DR: IMHO they're a great small provider, and I would absolutely recommend them.

Priced at $120 or £114 if you're in the UK its maybe a little outside of the low end hardware category that the Pi fits in although for twice the price of a 4GB model you're getting twice the memory you're also getting the same CPU performance so is this worth it?

Yes and no, I purchased the device as I need to do some CI/CD work for memory insensitive applications on aarm64 CPUs and it will serve its purpose well for that but how does it work as an addition to a homelab?

Most applications will run on the 4GB or 8GB model however if you want to run Virtualization the 16GB model actually comes in useful.
I am running Proxmox-port which gives me the ability to run KVM VMs and LXC containers - in my use case this is preferred as I can quickly swap OS for application testing without having to reinstall the device - but its just as simple as swapping an SD card I hear you say? This brings me onto the hardware additions:

This Pi 5 is running the official Active cooler: https://thepihut.com/products/active-cooler-for-raspberry-pi-5
A 128GB Patriot P320 128GB NVME SSD: https://www.amazon.co.uk/Patriot-P320-128GB-Internal-SSD/dp/B0D4RD18YV?th=1
using a GeeekPi N05 M.2 NVME hat: https://www.amazon.co.uk/GeeekPi-N04-Raspberry-Peripheral-Included/dp/B0CTLX5MW5

The NVME is running on a Single PCI-3 3.0 lane and is vastly quicker than an SD card and theoretically a whole load more reliable.

This brings the total cost to around £150 and I still don't have a case. For this price there are defiantly better X86 alternatives for homelab use. NUCs, Thinclients and even 2nd hand PCs although the power consumption here is in theory a total of 10w under load so what it lacks in performance it makes up for in power consumption.

Benchmarks:

I used YABS for these, skipping the Iperf tests.

Pi 5 - No Virtualization

Pi 5 - Ubuntu Server 22.04 VM

Virtual Machines

I am running 3 VMs currently, all installed from official ISOs

Debian 12
Ubuntu Server 22.04
Windows 11 ARM

Yes Windows 11 will run on a Pi4 in KVM:

The Linux performance is really good, better than a lot of budget KVM VPS'. I believe this could easily run 5 or 6 small linux VMs with absolutely no issue. CPU performance isn't amazing but its adequate for most needs.

Overall its a good addition to the homelab if you need to test on ARM64 or if you have strict power limitations. This will be added to my proxmox cluster and used for any testing I need to do on ARM64 environments. If you don't need ARM based CPUs I'd probably give this a miss.

@imok and I each got Netcup ARM G11 VPSes for Black Friday! Yaaay!

@imok's VPS is bigger than mine!

I finally was able to order mine as a result of seeing and following the link in @Blembim's post on December 20. Thanks again @Blembim!

CPU

Both @imok's and my new VPSes run on ARM CPUs. They also are Netcup's "11th generation" VPSes. So, @imok and I got Netcup "ARM G11 Netcup VPSes."

@imok has 10 cores, but I only have 6.

Nested virtualization (VM-x/AMD-V) to set up more fast KVM VPSes inside the main VPS is not enabled.

RAM

@imok has 15.6 GiB RAM, but I only have 7.7.

Disk

Our Netcup Holiday bargains have double the disk size above the non-Holiday, standard offers.

@imok has 1007.9 GiB SSD (not NVMe) disk, and I only have 503.9.

Network

2.5 Gbit/s; temporarily throttled to 200 Mbits/sec if previous 24 hrs average traffic exceeds 2 TB

1 x IPv4/32; 1 x IPv6/64

Location

Both @imok's VPS and mine are located in Manassas, VA, USA.

Pricing

My VPS costs €5.26 per month. No VAT for me because I ordered from the Mexico location where I currently am visiting.

There was an annual pricing option, which provided, in effect, two months free, but I prefer monthly pricing.

I used a $5 Netcup coupon ($5.00 = €4.20) for the first month. So my first month is costing only €1.06 = $1.11.

@imok posted that his VPS costs "$9/m."

Operating Systems

Netcup offers the expected Linux and FreeBSD images. One thing I really like is that the Netcup Server Control Panel ("SCP") makes it possible to upload custom images and ISOs via sftp. No support ticket and no waiting for support is needed to upload and use a custom image or ISO.

Account Setup and Ordering

Here is a list of links to my previous posts about the ordering process. The overall process seemed lengthy and painful. Product availability seemed intermittent. I had to try repeatedly. I had difficulty navigating the order and payment processes where I had to co-ordinate between email and web pages on my phone and on my Chromebook. The terms of service and all the notices seem like long and complicated walls of text.

Yabs

A yabs from my VPS is posted in the Pit.

@imok also posted his VPS's yabs.

Performance Decreases

I haven't analyzed the data carefully, but it looks like my VPS' disk I/O and Geekbench 6 performance have been descreasing, maybe as the Node filled up or as new users added their workloads -- or maybe both.

root@manassas:~# uptime
 18:48:36 up 12 days, 20:23,  2 users,  load average: 0.03, 0.19, 0.27
root@manassas:~# 

20241222:

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1056                          
Multi Core      | 4726                          
Full Test       | https://browser.geekbench.com/v6/cpu/9565111

20250104:

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 994                           
Multi Core      | 4351                          
Full Test       | https://browser.geekbench.com/v6/cpu/9766934

Steal

I have been keeping an eye on top to see what it says about steal. Mostly the steal says 0.1 or 0.2. Rarely, once or twice, I have seen a slightly higher steal.

Discussion

So far, after upgrading to Debian sid, I only have used my new Netcup ARM VPS for storing a few backups. It's performance has seemed great. Certainly, if I hadn't been rerunning yabs from time to time, I wouldn't have noticed the performance decreases. The Netcup ARM VPS gives the impression of being very well connected. It's fast, and snappy.

Eventually I will try a few processor-intensive tasks, such as compiling. Also, it would be fun to install NetBSD and Plan 9, which should be easier to do because I don't have to contact support to upload ISOs.

I think these Netcup ARM VPSes in Manassas USA seem great! Especially when we consider the combination of the USA location; the reliability expected due to Netcup's reputation; the generous CPU, RAM, and disk provisioning; and the low pricing. My VPS costs only approximately $5/mo. That's less than $7! What more could a Low Ender ask!? Plus, for more cores, RAM, and disk space, anyone can get the bigger, slightly more expensive machine, like @imok did.

Of course there are trade offs. Ryzen cores are faster than ARM. NVMe disks are faster than SSD. 10 Gbps is faster than 1 Gbps. But, so far, my Netcup VPS has been reliable. Everything at Netcup seems very well organized, and, once one naviagtes through the account setup and ordering processes, hassle free.

The combination of the US location, generous resource allocation, and low price makes me wonder -- despite the trade offs -- whether these Netcup Gen 11 ARM machines are the best Low End VPSes. The fact that @imok and @Blembim also got them makes me feel good about my purchase.

Is anyone aware of a better or cheaper USA VPS deal --

• from a well known provider and

• with equivalent or better CPU, RAM, disk, and network?

@imok @Blembim How are your new Netcup ARM VPSes doing?

Did anyone else get one of these for Black Friday? How is it doing?

Thanks and best wishes!

Frequent Downtime

September 19, 2024: The VPS went down, and I had to raise a ticket.
September 26, 2024 (11:49 AM): The VPS was down again. Linveo responded at 12:01 PM, claiming, "The internet is not down for these servers. Google's traffic is not reaching our ASN right now." A completely unsatisfactory explanation!
December 4, 2024 (6:22 AM): Another outage, and they admitted, "Our Dallas datacenter is currently offline. The DC is working on finding the issue and resolving it."
Unresolved Networking Issues

December 15, 2024 (11:10 AM): The VPS went down yet again, and this time, I discovered that the IP was unreachable from my home internet. I raised the issue with my ISP, who confirmed it was a problem with Linveo's network.
Linveo failed to address the problem and closed my ticket (#21763) without providing a solution. Their response: "The status has been changed to Closed. Dec 15, 2024, 2:42 PM by Adam Bednar."
Terrible Customer Support
Linveo's support team is unresponsive and dismissive. They seem more interested in closing tickets than solving customer problems.

Inconsistent Service
For a VPS provider, network reliability is crucial. Unfortunately, Linveo's network is plagued with issues, and they lack the competence to resolve them effectively.

Conclusion:
Linveo has proven to be unreliable, unprofessional, and unworthy of trust. Their frequent downtimes and lack of proper support make them one of the worst VPS providers I have ever dealt with. I would strongly recommend avoiding Linveo if you value your time, money, and sanity.

Rating: ⭐☆☆☆☆ (1/5)

I got a vps deal from virtwire back in 2015. This was $5 per year for a 512MB RAM OVZ/IP4+IP6. Virtwire closed its doors around 2016 and Hostsailor, their provider at NL, offered to take care of their NL clients. I was one of them so I ended up being a client of hostsailor.

I was using this VPS only as a sandbox for testing things, but I constantly monitored it all these years among my other machines: Zero downtime and respectable performance. But this was about to change.

About a couple of months ago, the machine became sluggish, close to unusable. Well to be honest at this point, I was feeling embarrassed to create a ticket ranting about the performance for a $5 service that was bought almost 10 years ago and they were loosing from it. So I was like, this machine has served its purpose all these years, now let's call it a day.

To my surprise HostSailor reached out by themselves, so I got a long email saying that they constantly monitor performance and have noticed performance degradation of this old node and yada yada so they decided to move clients to a new more powerful CPU machine with faster disks etc

And so it happened. IIRC it was a new EPYC node and this old vps became anew again.

However after a few days I was experiencing crashes. The machine was crashing and did not boot through the panel. At this point I created my first ticket about this. Long story short this on and off happened a few times. So I got a second email apologizing about the downtime, saying that the issues were due to the OVZ migration did not go well due to the age, so they decided to shutdown these openvz machines and replace them with KVM machines with doubling RAM and giving 3 days to migrate data.

Well I got the new 1G KVM and it is excellent, steal is unnoticeable, high IO, very good cpu performance.

TL;DR I have to say that it's rare to find such professionalism, i.e. excellent business culture and service, no matter what you pay for it. And this should be a good example for low-end philosophy.

I figure i have to give you guys for more context since me (as the direct buyer) partially responsible for this.
To make one thing clear: I'm not selling account. I already registered using my org email to begin with.

this is how all of this began, i think it's from the 900GB storage vps, but i need KVM and 1 IPv4. so i got a supposedly custom invoice:

notice anything wrong?. yep, a provider asking for f&f is a giant red flag and I should've stopped there. retard me still did the payment though (it ended more than $20).

Anyway, this is the review:

root@debian:~# curl -sL yabs.sh | bash
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2024-01-01                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Fri 01 Mar 2024 07:26:03 PM PST

Basic System Information:
---------------------------------
Uptime     : 15 days, 12 hours, 32 minutes
Processor  : QEMU Virtual CPU version 2.5+
CPU cores  : 2 @ 2599.998 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 1.9 GiB
Swap       : 975.0 MiB
Disk       : 883.9 GiB
Distro     : Debian GNU/Linux 11 (bullseye)
Kernel     : 5.10.0-28-amd64
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : Cinzia Tocci trading as C1V
ASN        : AS212271 Cinzia Tocci trading as C1V
Host       : C1V di Cinzia Tocci
Location   : Pomezia, Lazio (62)
Country    : Italy

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 27.75 MB/s    (6.9k) | 94.29 MB/s    (1.4k)
Write      | 27.78 MB/s    (6.9k) | 94.79 MB/s    (1.4k)
Total      | 55.53 MB/s   (13.8k) | 189.09 MB/s   (2.9k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 1.70 GB/s     (3.3k) | 176.56 MB/s    (172)
Write      | 1.79 GB/s     (3.5k) | 188.31 MB/s    (183)
Total      | 3.50 GB/s     (6.8k) | 364.87 MB/s    (355)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | busy            | 384 Mbits/sec   | 33.5 ms        
Scaleway        | Paris, FR (10G)           | busy            | 367 Mbits/sec   | 32.3 ms        
NovoServe       | North Holland, NL (40G)   | busy            | 377 Mbits/sec   | 32.6 ms        
Uztelecom       | Tashkent, UZ (10G)        | 127 Mbits/sec   | busy            | 116 ms         
Clouvider       | NYC, NY, US (10G)         | busy            | 239 Mbits/sec   | 102 ms         
Clouvider       | Dallas, TX, US (10G)      | busy            | 304 Mbits/sec   | 135 ms         
Clouvider       | Los Angeles, CA, US (10G) | busy            | 225 Mbits/sec   | 156 ms         

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 234                           
Multi Core      | 358                           
Full Test       | https://browser.geekbench.com/v6/cpu/5138003

YABS completed in 58 min 8 sec

I personally disappointed too. I was getting this vps in hope my friend can utilize this comfortably. but nope, c1vhosting just decide not to give a shit about performance. my kvm hdd vps in ramnode never have cpu steal even once.

After doing some digging about Luca Marini, also found out how bad he is at handling customer service. I.. can't find the right word for this, this person is just have no honor. really need to learn how you talk to other people as they're human(interpersonal skill), you're in the low market dude, your holier-than-thou corporate bullshit speak is laughable. I can't even fathom how I could eat or feed my family with money that I got from other people even though you could argue it's still "legal".

It sure is legal, the transaction is accepted, but the way you ignore about performance issue, or the way you're too dense(pretend to be retarded?) to act fixing vpn account when it's already shown the id, really shows how low you are.

I will never touch this provider ever again, and in the future when I have to buy new vps from italian provider, I'll make sure they have no Luca Marini name in there.

1. Network Downtime:
   The VPS went down due to a network issue for an entire day right after setup. That should have been a red flag, but I gave them the benefit of the doubt.

2. Disk Failure and Data Loss:
   Today, the VPS went down again, this time due to a “disk failure” caused by a supposed power interruption. I lost ALL my important data. C1vHosting’s response? They simply powered the VPS back on and claimed everything was fine, but my data was gone forever.

3. Unhelpful Support and Excuses:
   When I opened a ticket, they admitted the issue but casually mentioned the VPS might have different CPU and RAM resources than originally ordered. To "restore" my setup, they suggested I initiate a reinstallation through the client area. Guess what? When I tried, I got this error:

Error Occurred
An error occurred. Please contact the administrator and provide the details:
Message: Login to Proxmox host '10.0.4.3' port '8006' failed

When I asked their team to restore my data, their response was laughable:

“Hello,
I can see the VPS is up: 2 days 02:35:58”

Seriously? The VPS wasn’t even operational, and they insisted it was fine.

4. Unprofessional and Clueless Support:
   This team is completely unprofessional and seems to operate in some delusional state where they assume everything works without checking. It’s like they’re on drugs or something.

   Conclusion:
   Avoid C1vHosting at all costs. Their service is unreliable, their support team is incompetent, and they show zero accountability. I lost valuable data because of their negligence, and instead of offering a proper resolution, they gave me excuses and false assurances.

Trust me, spend your money elsewhere—this provider is not worth the frustration!

After a few months of use, it is stable and strong, and the machine is affordable

Pros:
$1.99/month
2x AMD Ryzen 9 7950X (4.5 GHz++)
2048MB DDR5 RAM
20GB NVMe (PCIe 4.0)
5,000GB Bandwidth
10Gbit Port
1x IPv4 Address & /64 IPv6
DDOS Watch™
5x FREE Backup Slots
Automated rDNS
VirtFusion Control Panel
Spokane, WA (North West USA)

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2024-01-01                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Wed 17 Jan 2024 10:22:30 AM CST

Basic System Information:
---------------------------------
Uptime     : 0 days, 20 hours, 20 minutes
Processor  : AMD Ryzen 9 7950X 16-Core Processor
CPU cores  : 2 @ 4499.980 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 1.9 GiB
Swap       : 1.0 GiB
Disk       : 20.0 GiB
Distro     : Debian GNU/Linux 11 (bullseye)
Kernel     : 5.10.0-27-amd64
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : Layer Marketing Services L.L.C
ASN        : AS216382 LAYER MARKETING SERVICES L.L.C
Host       : Layer
Location   : Spokane, Washington (WA)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda3):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 553.04 MB/s (138.2k) | 2.27 GB/s    (35.5k)
Write      | 554.50 MB/s (138.6k) | 2.28 GB/s    (35.7k)
Total      | 1.10 GB/s   (276.8k) | 4.56 GB/s    (71.3k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 2.02 GB/s     (3.9k) | 2.10 GB/s     (2.0k)
Write      | 2.13 GB/s     (4.1k) | 2.24 GB/s     (2.1k)
Total      | 4.16 GB/s     (8.1k) | 4.35 GB/s     (4.2k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 1.20 Gbits/sec  | 1.38 Gbits/sec  | 128 ms         
Scaleway        | Paris, FR (10G)           | busy            | busy            | 141 ms         
NovoServe       | North Holland, NL (40G)   | 1.16 Gbits/sec  | 1.33 Gbits/sec  | --             
Uztelecom       | Tashkent, UZ (10G)        | 555 Mbits/sec   | 700 Mbits/sec   | 1540 ms        
Clouvider       | NYC, NY, US (10G)         | 1.61 Gbits/sec  | 2.45 Gbits/sec  | 69.4 ms        
Clouvider       | Dallas, TX, US (10G)      | 1.56 Gbits/sec  | 2.45 Gbits/sec  | 73.3 ms        
Clouvider       | Los Angeles, CA, US (10G) | 1.62 Gbits/sec  | 2.88 Gbits/sec  | 90.1 ms        

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 1.19 Gbits/sec  | 1.22 Gbits/sec  | 128 ms         
Scaleway        | Paris, FR (10G)           | busy            | busy            | 132 ms         
NovoServe       | North Holland, NL (40G)   | 1.01 Gbits/sec  | 1.16 Gbits/sec  | 152 ms         
Uztelecom       | Tashkent, UZ (10G)        | 503 Mbits/sec   | 679 Mbits/sec   | 250 ms         
Clouvider       | NYC, NY, US (10G)         | 1.53 Gbits/sec  | 2.62 Gbits/sec  | 69.4 ms        
Clouvider       | Dallas, TX, US (10G)      | 1.55 Gbits/sec  | 2.39 Gbits/sec  | 70.0 ms        
Clouvider       | Los Angeles, CA, US (10G) | 1.46 Gbits/sec  | 1.99 Gbits/sec  | 90.0 ms        

Geekbench 6 test failed. Run manually to determine cause.

Sun Oct 13 08:52:36 AM CST 2024

Basic System Information:

Uptime : 0 days, 17 hours, 56 minutes
Processor : AMD EPYC 9554P 64-Core Processor
CPU cores : 2 @ 3099.998 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 1.9 GiB
Swap : 3.6 GiB
Disk : 48.9 GiB
Distro : Debian GNU/Linux 12 (bookworm)
Kernel : 6.1.0-26-cloud-amd64
VM Type : KVM
IPv4/IPv6 : ✔ Online / ✔ Online

IPv4 Network Information:

ISP : NextGenWebs, S.L.
ASN : AS41608 NextGenWebs, S.L.
Host : NextGenWebs, S.L
Location : Amsterdam, North Holland (NH)
Country : The Netherlands
Running fio random mixed R+W disk test with 4k
Running fio random mixed R+W disk test with 64kRunning fio random mixed R+W disk test with 512Running fio random mixed R+W disk test with 1m

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):

Block Size | 4k (IOPS) | 64k
(IOPS)
------ | --- ---- | ----
----
Read | 265.92 MB/s (66.4k) | 3.17 GB/s
(49.5k)
Write | 266.62 MB/s (66.6k) | 3.19 GB/s
(49.8k)
Total | 532.54 MB/s (133.1k) | 6.36 GB/s
(99.4k)
| |
Block Size | 512k (IOPS) | 1m
(IOPS)
------ | --- ---- | ----
----
Read | 3.07 GB/s (6.0k) | 3.66 GB/s
(3.5k)
Write | 3.24 GB/s (6.3k) | 3.90 GB/s
(3.8k)
Total | 6.31 GB/s (12.3k) | 7.57 GB/s
(7.3k)

Running GB6 benchmark test... *cue elevator musGeekbench 6 Benchmark Test:

Test | Value
|
Single Core | 1654
Multi Core | 3014

But it gets worse. On Saturday, August 24th, 2024, at 5:08 AM, they canceled my service without any prior notice or email notification. Not only was I left in the dark, but they also didn't issue a refund. When I opened a ticket about the sudden cancellation (Ticket ID: #MRB-399189), I received no satisfactory explanation.

You can see the cancellation screenshots here: https://prnt.sc/mms8wAoO1NTu.

Support : https://prnt.sc/gi3XYJtb1OSZ

Support : https://prnt.sc/zR88Du_XBDlE

I do NOT recommend Dedi Rock. They are an ultimate scam, and their support is beyond terrible. Save yourself the headache and go elsewhere!

I'm looking to purchase a few cPanel shared hosting accounts and would appreciate your recommendations on the best providers available. My primary requirement is hosting that offers decent resources to ensure reliable performance.

I would be grateful for any suggestions or advice you can provide on the top options.

Thank you for your assistance!

Best regards,
Maria

I'm considering signing up for 2-3 dedicated servers with Reliablesite.net and would appreciate your insights on their service. Specifically, I'm curious about which location would be the best choice.

I've also noticed that many providers offer uptime monitoring, but I couldn't find any information on Reliablesite.net regarding their server uptime or any page that tracks data center outages.

Could anyone provide guidance or share their experiences?

Thank you in advance for your help.

Best regards,
Maria

Control panel: Crunchbits has my favorite client area and user control panel/Dashboard. It's uncluttered, fast, easy to use and understand, and always works smoothly. A nice feature is that when you've started a task it shows you how much time is remaining until completion. When you use the backup feature (one free backup is included with deal) you also get an email telling you when it has finished.

Support: Only had to contact support once because I mistakenly generated an invoice for a new service, asked for it to be canceled and it was done within a few hours.

Operating Systems: Using Debian 11 and haven't had to resinstall or had any issues, don't recall if I used a template or what to install but it was trouble free. I see a lot of new OS options have been added to the control panel too.

Network: One IPV4. IPv6 included and very easy to add additional addresses in the control panel.

Overview: For me Crunchbits is the most reliable and professional option here in the LowEnd VPS space. As I understand it while they specialize in high end GPU servers they will provide basically the same level of service and support for users who get one of their black friday or other inexpensive options. That's exactly what you want from a Low End provider and suspect it's very difficult to deliver over a long period of time and not go out of business. Another thing in the plus column is that they are quite active in the Low End forums and go out of their way to come up with entertaining posts, giveaways, and Black Friday deals. Free hardware giveaways, win a VPS contests, occasional loss leader deals, humorous and insightful posts, and always maintaining professional discourse. Highly recommended.

Yep, the "2nd Amendment" title was just click bait.

Obligatory Masonic YABS:

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2024-06-09                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Sun 30 Jun 2024 11:48:46 AM EDT

Basic System Information:
---------------------------------
Uptime     : 38 days, 11 hours, 43 minutes
Processor  : Intel Core Processor (Broadwell)
CPU cores  : 2 @ 2194.916 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 2.8 GiB
Swap       : 0.0 KiB
Disk       : 25.0 GiB
Distro     : Debian GNU/Linux 11 (bullseye)
Kernel     : 5.10.0-29-amd64
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : Redoubt Networks
ASN        : AS400304 Redoubt Networks
Host       : Redoubt Networks
Location   : Liberty Lake, Washington (WA)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda3):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 162.74 MB/s  (40.6k) | 999.83 MB/s  (15.6k)
Write      | 163.17 MB/s  (40.7k) | 1.00 GB/s    (15.7k)
Total      | 325.92 MB/s  (81.4k) | 2.00 GB/s    (31.3k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 1.37 GB/s     (2.6k) | 1.44 GB/s     (1.4k)
Write      | 1.44 GB/s     (2.8k) | 1.53 GB/s     (1.5k)
Total      | 2.82 GB/s     (5.5k) | 2.97 GB/s     (2.9k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed                                   | Ping
-----           | -----                     | ----            | ----                                         | ----
Clouvider       | London, UK (10G)          | 846 Mbits/sec   | 406 Mbits/sec                                | 129 ms
Eranium         | Amsterdam, NL (100G)      | 967 Mbits/sec   | 649 Mbits/sec                                | --
Uztelecom       | Tashkent, UZ (10G)        | 378 Mbits/sec   | 371 Mbits/sec                                | 260 ms
Leaseweb        | Singapore, SG (10G)       | 705 Mbits/sec   | 425 Mbits/sec                                | 250 ms
Clouvider       | Los Angeles, CA, US (10G) | 1.05 Gbits/sec  | busy                                         | 40.5 ms
Leaseweb        | NYC, NY, US (10G)         | 977 Mbits/sec   | busy                                         | 93.8 ms
Edgoo           | Sao Paulo, BR (1G)        | 196 Mbits/sec   | 420 Mbits/sec                                | 173 ms

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed                                   | Ping
-----           | -----                     | ----            | ----                                         | ----
Clouvider       | London, UK (10G)          | 981 Mbits/sec   | 603 Mbits/sec                                | 128 ms
Eranium         | Amsterdam, NL (100G)      | busy            | busy                                         | --
Uztelecom       | Tashkent, UZ (10G)        | busy            | 274 Mbits/sec   | --
Leaseweb        | Singapore, SG (10G)       | 644 Mbits/sec   | 474 Mbits/sec   | 249 ms
Clouvider       | Los Angeles, CA, US (10G) | 1.04 Gbits/sec  | 798 Mbits/sec   | 40.5 ms
Leaseweb        | NYC, NY, US (10G)         | 952 Mbits/sec   | 594 Mbits/sec   | 93.9 ms
Edgoo           | Sao Paulo, BR (1G)        | 561 Mbits/sec   | 305 Mbits/sec   | 170 ms

i thought to myself if there is a poll of past and current customer, it would be easier to understand.
im one of those lazy people who like too look at polls/rating, without reading the whole review, before trying new stuff after all. i.e. i only look at watch it or skip it section (may be scan it a bit) of Decider post before watching.

So, without further ado, join the poll.

Side note, every review should make a poll of some kind. it helps to see if your case was isolated or a regular pattern. Plus im kinda lazy

[DET01] Datacenter Move
Scheduled for June 08, 2024 at 11:00 PM – June 09, 2024 at 7:00 AM
We will be moving our equipment from our current building to a datacenter operated by the same company across the street. Please expect about 4 hours of downtime during this migration as we migrate our equipment, including all customer servers and hardware powering our website. We appreciate your patience and understanding. We will follow up with more details and information.

Latest $17/yr Summer Sale:
lowendtalk.com/discussion/195058/hostodo-gen2-epyc-nvme-vps-15-off-free-nvme-cpu-bw-upgrades-1g-17-yr-2g-ram-52g-nvme-34-yr

DET01 Detroit, MI
GEN2BETA1GN
1GB RAM
16GB NVMe
1 vCPU Cores
3000GB Data Transfer
$10.00/year

Yet-Another-Bench-Script
v2024-04-22
Sat Jun  1 06:47:57 UTC 2024

Basic System Information:
---------------------------------
Uptime     : 27 days, 14 hours, 37 minutes
Processor  : Intel(R) Xeon(R) CPU E5-2690 v4 @ 2.60GHz
CPU cores  : 1 @ 2599.996 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 977.3 MiB
Swap       : 0.0 KiB
Disk       : 15.7 GiB
Distro     : Debian GNU/Linux 12 (bookworm)
Kernel     : 6.1.0-18-cloud-amd64
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : Hostodo
ASN        : AS399804 Hostodo
Host       : Hostodo
Location   : Sterling, Virginia (VA)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/sda1):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 104.91 MB/s  (26.2k) | 1.04 GB/s    (16.3k)
Write      | 105.19 MB/s  (26.2k) | 1.05 GB/s    (16.4k)
Total      | 210.11 MB/s  (52.5k) | 2.10 GB/s    (32.8k)
           |                      |
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ----
Read       | 1.22 GB/s     (2.3k) | 1.24 GB/s     (1.2k)
Write      | 1.29 GB/s     (2.5k) | 1.32 GB/s     (1.2k)
Total      | 2.52 GB/s     (4.9k) | 2.57 GB/s     (2.5k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed              | Ping
-----           | -----                     | ----            | ----                    | ----
Clouvider       | London, UK (10G)          | 523 Mbits/sec   | 785 Mbits/sec           | 90.4 ms
Eranium         | Amsterdam, NL (100G)      | 807 Mbits/sec   | 690 Mbits/sec           | --
Telia           | Helsinki, FI (10G)        | busy            | busy                    | 120 ms
Uztelecom       | Tashkent, UZ (10G)        | 764 Mbits/sec   | 351 Mbits/sec           | 185 ms
Leaseweb        | Singapore, SG (10G)       | 7.50 Mbits/sec  | 585 Mbits/sec           | 217 ms
Clouvider       | Los Angeles, CA, US (10G) | 842 Mbits/sec   | 776 Mbits/sec   | --    
Leaseweb        | NYC, NY, US (10G)         | 867 Mbits/sec   | 823 Mbits/sec   | --    
Edgoo           | Sao Paulo, BR (1G)        | 740 Mbits/sec   | 673 Mbits/sec   | 130 ms

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping  
-----           | -----                     | ----            | ----            | ----  
Clouvider       | London, UK (10G)          | 805 Mbits/sec   | 264 Mbits/sec   | 90.3 ms
Eranium         | Amsterdam, NL (100G)      | busy            | busy            | --    
Uztelecom       | Tashkent, UZ (10G)        | 585 Mbits/sec   | 392 Mbits/sec   | 185 ms
Leaseweb        | Singapore, SG (10G)       | busy            | 667 Mbits/sec   | 219 ms
Clouvider       | Los Angeles, CA, US (10G) | busy            | 355 Mbits/sec   | --    
Leaseweb        | NYC, NY, US (10G)         | 869 Mbits/sec   | 873 Mbits/sec   | --    
Edgoo           | Sao Paulo, BR (1G)        | busy            | 565 Mbits/sec   | 130 ms

Geekbench test failed and low memory was detected. Add at least 1GB of SWAP or use GB4 instead (higher compatibility with low memory systems).
YABS completed in 15 min 11 sec

Traceroute to China Telecom:
1 74.80.183.89 (74.80.183.89) 0.218 ms 0.193 ms 0.178 ms
2 v165.agr3.sfld2-mi.incx.net (23.26.250.101) 0.366 ms 0.352 ms 0.338 ms
3 te0-0-0-18.rcr21.dtw04.atlas.cogentco.com (38.122.60.22) 1.637 ms 1.569 ms 1.813 ms
4 be2123.rcr51.tol01.atlas.cogentco.com (154.54.82.137) 2.620 ms 2.470 ms 2.774 ms
5 be3744.ccr21.cle04.atlas.cogentco.com (154.54.0.145) 4.590 ms 4.777 ms be3745.ccr22.cle04.atlas.cogentco.com (154.54.30.129) 4.756 ms
6 be2717.ccr41.ord01.atlas.cogentco.com (154.54.6.221) 11.113 ms be2718.ccr42.ord01.atlas.cogentco.com (154.54.7.129) 11.192 ms be2717.ccr41.ord01.atlas.cogentco.com (154.54.6.221) 11.170 ms
7 be2831.ccr21.mci01.atlas.cogentco.com (154.54.42.165) 22.911 ms be2832.ccr22.mci01.atlas.cogentco.com (154.54.44.169) 22.924 ms 22.902 ms
8 be3035.ccr21.den01.atlas.cogentco.com (154.54.5.89) 34.392 ms be3036.ccr22.den01.atlas.cogentco.com (154.54.31.89) 34.411 ms be3035.ccr21.den01.atlas.cogentco.com (154.54.5.89) 34.186 ms
9 be3037.ccr21.slc01.atlas.cogentco.com (154.54.41.145) 44.664 ms 44.685 ms be3038.ccr32.slc01.atlas.cogentco.com (154.54.42.97) 44.671 ms
10 be3109.ccr21.sfo01.atlas.cogentco.com (154.54.44.137) 58.669 ms 58.703 ms be3110.ccr22.sfo01.atlas.cogentco.com (154.54.44.141) 58.758 ms
11 be3670.ccr41.sjc03.atlas.cogentco.com (154.54.43.14) 60.263 ms be3669.ccr41.sjc03.atlas.cogentco.com (154.54.43.10) 60.243 ms be3670.ccr41.sjc03.atlas.cogentco.com (154.54.43.14) 59.925 ms
12 * * *
13 202.97.6.1 (202.97.6.1) 210.801 ms 210.814 ms 211.303 ms
14 * 202.97.12.209 (202.97.12.209) 200.735 ms 202.97.12.185 (202.97.12.185) 202.326 ms
15 202.97.57.146 (202.97.57.146) 204.114 ms 202.97.85.50 (202.97.85.50) 199.017 ms 202.97.83.134 (202.97.83.134) 216.811 ms
16 202.97.102.30 (202.97.102.30) 214.449 ms 202.97.100.254 (202.97.100.254) 207.743 ms *
17 * 220.191.199.2 (220.191.199.2) 222.044 ms 220.191.199.134 (220.191.199.134) 201.896 ms
18 * * *
19 * * *
-- Traceroute timed out --

Traceroute to China Unicom:
1 74.80.183.89 (74.80.183.89) 0.224 ms 0.186 ms 0.172 ms
2 v165.agr3.sfld2-mi.incx.net (23.26.250.101) 0.323 ms 0.311 ms 0.296 ms
3 te0-0-0-18.rcr21.dtw04.atlas.cogentco.com (38.122.60.22) 1.125 ms 1.112 ms 1.097 ms
4 * * *
5 be3744.ccr21.cle04.atlas.cogentco.com (154.54.0.145) 4.887 ms 4.887 ms 4.980 ms
6 be2717.ccr41.ord01.atlas.cogentco.com (154.54.6.221) 27.738 ms 26.503 ms be2718.ccr42.ord01.atlas.cogentco.com (154.54.7.129) 11.029 ms
7 be2832.ccr22.mci01.atlas.cogentco.com (154.54.44.169) 22.708 ms 23.357 ms 23.741 ms
8 be3035.ccr21.den01.atlas.cogentco.com (154.54.5.89) 46.151 ms be3036.ccr22.den01.atlas.cogentco.com (154.54.31.89) 52.022 ms be3035.ccr21.den01.atlas.cogentco.com (154.54.5.89) 34.113 ms
9 be3038.ccr32.slc01.atlas.cogentco.com (154.54.42.97) 44.370 ms 44.397 ms 44.465 ms
10 be3109.ccr21.sfo01.atlas.cogentco.com (154.54.44.137) 72.450 ms be3110.ccr22.sfo01.atlas.cogentco.com (154.54.44.141) 58.793 ms 70.913 ms
11 be3669.ccr41.sjc03.atlas.cogentco.com (154.54.43.10) 59.992 ms 59.923 ms 59.961 ms
12 38.88.225.98 (38.88.225.98) 93.474 ms 38.88.225.106 (38.88.225.106) 90.843 ms 38.88.225.98 (38.88.225.98) 88.696 ms
13 219.158.117.1 (219.158.117.1) 273.654 ms 273.607 ms 272.034 ms
14 219.158.9.221 (219.158.9.221) 289.952 ms 303.372 ms 302.658 ms
15 219.158.9.241 (219.158.9.241) 252.053 ms 251.610 ms 266.107 ms
16 202.96.12.38 (202.96.12.38) 250.060 ms 61.149.203.210 (61.149.203.210) 273.455 ms 124.65.194.10 (124.65.194.10) 273.145 ms
17 61.51.169.170 (61.51.169.170) 255.929 ms 61.51.169.138 (61.51.169.138) 284.843 ms 61.51.169.166 (61.51.169.166) 246.031 ms
18 61.49.170.2 (61.49.170.2) 256.123 ms * 251.804 ms
19 * * *
20 * * *
-- Traceroute timed out --

Traceroute to China Mobile:
1 74.80.183.89 (74.80.183.89) 0.186 ms 0.142 ms 0.117 ms
2 v165.agr3.sfld2-mi.incx.net (23.26.250.101) 0.331 ms 0.301 ms 0.265 ms
3 te0-0-0-18.rcr21.dtw04.atlas.cogentco.com (38.122.60.22) 1.148 ms 1.382 ms 1.239 ms
4 be2123.rcr51.tol01.atlas.cogentco.com (154.54.82.137) 2.439 ms 2.581 ms 2.643 ms
5 be3745.ccr22.cle04.atlas.cogentco.com (154.54.30.129) 4.651 ms be3744.ccr21.cle04.atlas.cogentco.com (154.54.0.145) 4.790 ms be3745.ccr22.cle04.atlas.cogentco.com (154.54.30.129) 4.571 ms
6 be2718.ccr42.ord01.atlas.cogentco.com (154.54.7.129) 11.247 ms be2717.ccr41.ord01.atlas.cogentco.com (154.54.6.221) 11.133 ms be2718.ccr42.ord01.atlas.cogentco.com (154.54.7.129) 11.142 ms
7 be2831.ccr21.mci01.atlas.cogentco.com (154.54.42.165) 23.083 ms 22.789 ms be2832.ccr22.mci01.atlas.cogentco.com (154.54.44.169) 22.814 ms
8 be2433.ccr32.dfw01.atlas.cogentco.com (154.54.3.213) 32.856 ms be2432.ccr31.dfw01.atlas.cogentco.com (154.54.3.133) 72.837 ms be3036.ccr22.den01.atlas.cogentco.com (154.54.31.89) 34.216 ms
9 154.54.165.26 (154.54.165.26) 45.562 ms 154.54.165.30 (154.54.165.30) 44.636 ms 154.54.165.26 (154.54.165.26) 45.694 ms
10 be3872.ccr32.phx01.atlas.cogentco.com (154.54.26.53) 54.012 ms 53.975 ms 53.947 ms
11 be2932.ccr42.lax01.atlas.cogentco.com (154.54.45.162) 65.984 ms 64.240 ms 64.103 ms
12 be3359.ccr41.lax05.atlas.cogentco.com (154.54.3.70) 67.155 ms 67.130 ms be3243.ccr41.lax05.atlas.cogentco.com (154.54.27.118) 64.095 ms
13 38.104.85.162 (38.104.85.162) 118.121 ms 117.999 ms *
14 223.120.6.217 (223.120.6.217) 118.856 ms 115.822 ms 116.813 ms
15 * * *
16 * 221.183.89.178 (221.183.89.178) 318.529 ms 316.457 ms
17 221.183.89.69 (221.183.89.69) 346.795 ms 316.437 ms 315.222 ms
18 221.183.89.10 (221.183.89.10) 324.710 ms 221.183.89.46 (221.183.89.46) 536.826 ms 221.183.89.50 (221.183.89.50) 310.820 ms
19 221.183.40.174 (221.183.40.174) 336.841 ms 221.183.40.94 (221.183.40.94) 345.314 ms *
20 221.183.71.246 (221.183.71.246) 340.508 ms * 397.027 ms
21 221.183.139.234 (221.183.139.234) 589.631 ms * 221.183.139.238 (221.183.139.238) 338.675 ms
22 * * *
23 * * *
-- Traceroute timed out --

I decided to give away 4, which later was increased to 8, Crunchy VPSes made with bashvm. Bashvm is an open source set of bash shell scripts from @babywhale that simplifies KVM VPS creation via libvirt, along with IPv6 and firewall management for IPv4 NAT.

The 8 Crunchy VPSes were sold quickly for a one time payment of $1.01 per VPS covering the entire period from March 11 until November 8. That's practically free!

You can still see the offer thread here: MetalVPS at Crunchbits: E3-1245, 2 vCores, 4 GB DDR4 ECC, 1000GB SSD RAID 10, 1 Gbps, $1.01 once!

All 8 Crunchy VPSes have been running continuously, in addition to a couple of small VPSes that @subenhon created manually via the command line. I have two test VPSes (currently shut down) and have occasionally run tinyproxy. Recently I started keeping some backup files on Crunchy as well.

Ubuntu and bashvm have handled the few Node updates seamlessly. The bashvm VPSes are set to restart automatically upon reboot.

I have logged in to Crunchy multiple times daily, checking load averages, running vnstat, and sometimes using tools like top or htop.

A couple of days ago, it struck me that Crunchy has been 100% available! There hasn't been even a single issue of which I am aware, nor have any of our Crunchy neighbors reported any problems. I've been particularly pleased with the lack of network issues. A new yabs is posted below.

Based on three months use by ten of us (8 VPS users, @subenhon, and myself), having zero problems is quite impressive! Congratulations to @crunchbits and @babywhale! Thanks again to @crunchbits for wonderful Crunchy!

We will announce here all our updates about our company (iHostART.Com)

Our official contacts
Official website: https://ihostart.com
Mail contact: office@ihostart.com
Discord: @ihostart
Telegram: @ihostart
Skype: live:.cid.e09478c64eb647ce
Whatsapp: +40799818190

Our official Discord server https://discord.gg/mG3Pv76rmB JOIN HERE , most probality in future make more giveaways

Here we are not so active, but to prevent fake accounts
Linkedin: https://www.linkedin.com/in/calin-maniu-2203b3239/
Reddit: https://www.reddit.com/user/ihostart

https://drive.google.com/file/d/1z4Wqg_4S_Xgxfi4IsqtuetMYsAiZ5xm9/view?usp=sharing
https://drive.google.com/file/d/1AexrPn6c3HPdPj8vzKkJ9JizKrcpovrE/view?usp=sharing
https://drive.google.com/file/d/1uablKtzvHs9UE7ABF7C8Joujj_cH96m5/view?usp=sharing
https://imgur.com/HFP1Iyf

I plan to try Oracle free tier. I hear you can only pick your region for the servers once. Any recommendations on which region is best?

I see in the US, Ashburn, Chicago and Phoenix have extra resources compared to others.
https://docs.oracle.com/en-us/iaas/Content/General/Concepts/regions.htm

Best defined by best availability, less likely to be reclaimed by oracle, reliable and no issues. Feel free to share anything else one should look out for. Plan to use this to test stuff. Not production.

Also if you can only create two block volumes for storage. Does that mean you can only create two VPS’? Curious if one could divide resources and have 2 AMD and 4 AMP at once.

Thanks

• Trustpilot's trustworthiness issues related to how their "Content Integrity Team" handles stuff.
• Charging an arm and a leg to let you display your Trustpilot reviews on your company's website​.

TL/DR As far as I can tell, it's better less bad to stick to Google reviews.

A long-winded post on the topic, with linked resources (blog and Linkedin posts basically):

Trustpilot trustworthines issues

It's not an easy job to build and run a platform like that. And I'm sure it costs money.
On the other hand, the way Trustpilot does things does not seem very good.

It would be nice if they were better and more reasonable than Google.
The way it is now, they seem to be slownly jockeying into a de-facto (hosting) company reputation/review centre, without even trying to assure that the reviews are legit (which, if they did it, just might justify their outrageously high prices).

What's your take?

Backround
I needed another server for a specific use case, and I reached out to NetDynamics about reselling me a Hetzner dedicated server. We talked a little back and forward about pricing, and ended up agreeing on a price who made sense to me (since I would have to pay VAT at Hetzner), and leaving NetDynamics some headroom for profit. Michael Vlastos took his time with me, I got lightning fast responses through email / ticket, and he communicated in a good and professional way.

The dedicated server I wanted showed in stock at Hetzner with instant delivery, but they wasn't able to deliver it. After a few days, we ended up ordering a different server instead of waiting forever, and we agreed on a price on that yet again. This server was delivered instantly, all good.

After some more days, the original server was delivered. I emailed Michael Vlastos again to work out some solution to swap over to that. We agreed on a price for this (since they do not do refunds on dedicated servers), and I ended up paying an extra invoice for a fee for all of his time spent here. The server was cancelled, and the original planned server was delivered. All good!

After putting my workload on the shiny new server, it turns out after a week or so that the server can't handle what I want it to do. I yet again emailed Michael Vlastos, and we agreed on a fair partial refund. Both the servers we are talking about here can be refunded in full from Hetzner due to the limited timeframe here. Still I purposed a refund amount, leaving him with total 64 euro "profit" for his time and effort spent on me trying to help me out.

Im aware that this is not a long term relation review, but since this is about a resold Hetzner server, I don't see any point in writing about benchmarks, network and performance, as this is irrelevant to this situation.
The key here is : Michael Vlastos is good at customer service. Lightning fast responses (take into consideration that he is human, and also needs sleep). His responses are fast, accurate and professional. He was 100% willing to work out solutions working out for both of us, and asked me several times what I think Is fair pricing, fair refund and so on. Good guy! That said, he is human, running a business, and wants to make $ for his time spent. This is reflected in our "working" on pricing. Though, he was always fair and straight forward, easy to communicate with.

So to sum up, I was happy with Michael Vlastos when it comes to communication and customer service. I only have 1 thing I want to point out, who is a BIG NONO. That was the fact that he straight out asked for a review posted on public forums after the final refund was done. I don't think you should ask a customer to post a review on public forums, but he did. However, I honestly don't have anything bad to say about this guy. We had 1 miss communication about the final refund, but as soon as I pointed it out, he went above and beyond sending me funds from his personal PayPal account instead, due to the fact that I used a burner card number who cannot be refunded to as my initial payment method. Right after this, is the point where he asked for a public review.

So, if you are looking at his services, you should be able to feel safe, and get the help you need. Michael Vlastos is a good guy, who responds fast, accurate and straight forward without any bullshit. Can 100% recommend NetDynamics24 if you are looking to use his services.

First, you get one of their "VM.Standard.A1.Flex" instances:

• Ampere A1 (arm64) processors (that's Neoverse-N1 cores)
• 24gb RAM
• Up to 200gb of block storage
• 1 ipv4

The bad:

• Oracle Cloud UI is subpar, even compared to their competitors' which aren't great themselves (AWS, Azure...)
• Instances may not be available for creation in your prefered region depending on when you're trying it. I choose a nearby region and it's worked fine for me.
• Depending on your luck, registering can be a hassle due to strong anti fraud checks which also impact legitimate users: UI is slow and unreliable, and the process may fail without a clear reason.
• You need to give a credit card number, although they won't charge you.
• We're talking about an ARM instance, this can be a problem if you want to run x86/amd64 specific software, though that's becoming less of an issue over time.

The good:

• Unusually powerful cpu, incredible ram amount, sizeable storage.
• Instance/network seem fairly reliable, no major problem since I've been using it.
• I've run heavy stuff such as Elasticsearch and Kubernetes on it, with success.

You can look at some benchmarks just after, but to sum up, I'd say that if you can pass their registration process and there are instances available, it's kind of a no-brainer.

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2023-04-23                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

ARM compatibility is considered *experimental*

Basic System Information:
---------------------------------
Uptime     : 740 days, 15 hours, 23 minutes
Processor  : Neoverse-N1
CPU cores  : 4 @ ??? MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM        : 22.6 GiB
Swap       : 8.0 GiB
Disk       : 100.0 GiB
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : Oracle Corporation
ASN        : AS31898 Oracle Corporation
Host       : Oracle Cloud Infrastructure (uk-london-1)
Location   : London, England (ENG)
Country    : United Kingdom

fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 13.45 MB/s    (3.3k) | 27.16 MB/s     (424)
Write      | 13.47 MB/s    (3.3k) | 27.97 MB/s     (437)
Total      | 26.93 MB/s    (6.7k) | 55.13 MB/s     (861)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 24.80 MB/s      (48) | 24.45 MB/s      (23)
Write      | 26.92 MB/s      (52) | 27.28 MB/s      (26)
Total      | 51.73 MB/s     (100) | 51.73 MB/s      (49)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 3.99 Gbits/sec  | 1.22 Gbits/sec  | 2.09 ms        
Scaleway        | Paris, FR (10G)           | 4.00 Gbits/sec  | 371 Mbits/sec   | 8.80 ms        
NovoServe       | North Holland, NL (40G)   | 4.01 Gbits/sec  | 2.55 Gbits/sec  | 7.46 ms        
Uztelecom       | Tashkent, UZ (10G)        | 717 Mbits/sec   | 49.8 Mbits/sec  | 102 ms         
Clouvider       | NYC, NY, US (10G)         | 937 Mbits/sec   | 82.4 Mbits/sec  | 70.2 ms        
Clouvider       | Dallas, TX, US (10G)      | 662 Mbits/sec   | 89.1 Mbits/sec  | 107 ms         
Clouvider       | Los Angeles, CA, US (10G) | 491 Mbits/sec   | 68.4 Mbits/sec  | 132 ms 

-------------------- A Bench.sh Script By Teddysun -------------------
 Version            : v2023-06-10
 Usage              : wget -qO- bench.sh | bash
----------------------------------------------------------------------
 CPU Model          : CPU model not detected
 CPU Cores          : 4
 AES-NI             : Enabled
 VM-x/AMD-V         : Disabled
 Total Disk         : 100.0 GB (22.8 GB Used)
 Total Mem          : 22.6 GB (4.5 GB Used)
 Total Swap         : 8.0 GB (275.6 MB Used)
 System uptime      : 740 days, 15 hour 34 min
 Load average       : 0.06, 0.06, 0.04
 Arch               : aarch64 (64 Bit)
 TCP CC             : cubic
 Virtualization     : KVM
 IPv4/IPv6          : Online / Offline
 Organization       : AS31898 Oracle Corporation
 Location           : London / GB
 Region             : England
----------------------------------------------------------------------
 I/O Speed(1st run) : 55.7 MB/s
 I/O Speed(2nd run) : 50.5 MB/s
 I/O Speed(3rd run) : 52.2 MB/s
 I/O Speed(average) : 52.8 MB/s
----------------------------------------------------------------------
 Node Name        Upload Speed      Download Speed      Latency     
 Speedtest.net    1543.83 Mbps      947.64 Mbps         0.81 ms     
 Los Angeles, US  301.05 Mbps       22.04 Mbps          129.99 ms   
 Dallas, US       289.66 Mbps       538.64 Mbps         102.12 ms   
 Montreal, CA     176.57 Mbps       140.22 Mbps         75.06 ms    
 Paris, FR        1465.03 Mbps      159.86 Mbps         9.59 ms     
 Amsterdam, NL    1338.97 Mbps      189.83 Mbps         6.98 ms     
 Shanghai, CN     113.45 Mbps       535.47 Mbps         216.68 ms   
 Nanjing, CN      170.08 Mbps       513.50 Mbps         259.69 ms   
 Hongkong, CN     177.91 Mbps       1735.52 Mbps        246.73 ms   
 Singapore, SG    286.40 Mbps       8.75 Mbps           161.03 ms   
 Tokyo, JP        133.62 Mbps       16.18 Mbps          230.15 ms   
----------------------------------------------------------------------
 Finished in        : 6 min 15 sec
----------------------------------------------------------------------

If you'd like any other benchmarks, feel free to ask.

Here is my full ProWritingAid free & Premium review (a work in progress - I'll keep it up to date as stuff gets worse improves).

It's not (too) bad. Fixes any basic, simple grammar and spelling mistakes - and sucks with complex sentences. Better than Grammarly when dealing with commas and my writing style in general ( doesn't come close to @AlwaysSkint or a non-AI editor, of course).

Also ( @Ympker might appreciate this), it comes with a lifetime plan - at a 40% discount (every 6 months or so there is some 40 - 50% discount as far as I know). Here's the link to the page that leads to the discount link:

https://nanowrimo.org/offers

Premium NVMe VPS Servers - [PRE-ORDER - 20% OFF] Singapore | PRO-1 | 25GB | £4.00 Monthly

YABS

# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2023-04-23                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Mon 03 Jul 2023 09:23:15 AM +08

Basic System Information:
---------------------------------
Uptime     : 111 days, 14 hours, 57 minutes
Processor  : AMD EPYC-Rome Processor
CPU cores  : 1 @ 3792.874 MHz
AES-NI     : ? Enabled
VM-x/AMD-V : ? Disabled
RAM        : 1.9 GiB
Swap       : 1024.0 MiB
Disk       : 24.6 GiB
Distro     : Debian GNU/Linux 11 (bullseye)
Kernel     : 5.10.0-21-amd64
VM Type    : KVM
IPv4/IPv6  : ? Online / ? Online


IPv6 Network Information:
---------------------------------
ISP        : Host Universal Pty Ltd
ASN        : AS136557 Host Universal Pty Ltd
Host       : Dot-Tech LLC
Location   : Shinagawa, Tokyo (13)
Country    : Japan

fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 290.96 MB/s  (72.7k) | 2.05 GB/s    (32.1k)
Write      | 291.73 MB/s  (72.9k) | 2.06 GB/s    (32.3k)
Total      | 582.70 MB/s (145.6k) | 4.12 GB/s    (64.4k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 2.17 GB/s     (4.2k) | 1.88 GB/s     (1.8k)
Write      | 2.29 GB/s     (4.4k) | 2.01 GB/s     (1.9k)
Total      | 4.46 GB/s     (8.7k) | 3.90 GB/s     (3.8k)

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | busy            | busy            | 224 ms         
Scaleway        | Paris, FR (10G)           | busy            | busy            | 164 ms         
NovoServe       | North Holland, NL (40G)   | busy            | busy            | 236 ms         
Uztelecom       | Tashkent, UZ (10G)        | busy            | busy            | 200 ms         
Clouvider       | NYC, NY, US (10G)         | busy            | busy            | 233 ms         
Clouvider       | Dallas, TX, US (10G)      | 543 Mbits/sec   | 766 Mbits/sec   | 197 ms         
Clouvider       | Los Angeles, CA, US (10G) | 500 Mbits/sec   | 797 Mbits/sec   | 176 ms         

iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
-----           | -----                     | ----            | ----            | ----           
Clouvider       | London, UK (10G)          | 546 Mbits/sec   | 640 Mbits/sec   | 224 ms         
Scaleway        | Paris, FR (10G)           | 764 Mbits/sec   | 771 Mbits/sec   | 147 ms         
NovoServe       | North Holland, NL (40G)   | 622 Mbits/sec   | 648 Mbits/sec   | 236 ms         
Uztelecom       | Tashkent, UZ (10G)        | 772 Mbits/sec   | 752 Mbits/sec   | 200 ms         
Clouvider       | NYC, NY, US (10G)         | 535 Mbits/sec   | 38.6 Mbits/sec  | 233 ms         
Clouvider       | Dallas, TX, US (10G)      | 690 Mbits/sec   | 707 Mbits/sec   | 197 ms         
Clouvider       | Los Angeles, CA, US (10G) | 772 Mbits/sec   | 803 Mbits/sec   | 176 ms         

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1543                          
Multi Core      | 1549                          
Full Test       | https://browser.geekbench.com/v6/cpu/1797013

YABS completed in 15 min 44 sec

what the most impressive is that hetrixtools says it has 100% uptime.

I rarely require support as I am usually capable of handling things on my own (except for that one time when I asked for reverse DNS, but now it's available in the panel). From what I've observed, their Discord is quite active, making it easy to ask questions and receive answers there.

5/5 recommended, would use again.

My man onepoundemail got dirt cheap email hosting service, at that time I was looking for alternative from amazonSES for sending really not important email (mostly grafana alert stuff) so this one fit the bill.

I'll walk you trough on my experience using this service, first off i got yeeted with error when placing order so.. I got one more invoice for free.

alas, my order is complete, payment with paypal went smoothly. after my order is active I remember i have to put 2FA in my account. surprisingly enough I'm unable to do that since the member edit page doesn't work properly.

surely they will notice the page error from server side if there's 500 code right?. so I moved on into my email package, the log in menu doesn't throw me directly into directadmin using single click login, and it also works funny in modern web browser since they don't put https by default.

I'm also not sure where my domain or subdomain supposed to be pointed at which MX record, luckily it's available in their knowledge base.

after I'm able to log in, everything works as expected. even since day one the mail went trough flawlessly.

the uptime also great, hetrixtools says it has 99% uptime

To put in summary:
1. pros: it's not 7$
2. cons: it's not 7$

onepoundemail has a lot to work with. IMO they can put improvement in these area:

• check 2fa setup page (or.. well, all of your pages) before selling the service
• invoices doesn't seems to get automatically cancelled after n days, i don't mind it as it doesn't auto charge my paypal but some other people might do (just qol things)
• web interface safari.mxrouting.net doesn't seems to use the same ssl certificate with safari.mxrouting.net:993 for email. this can be seen one month ago and today, so it's kind of questionable for me, why would you micromanage such stuff?. If you're looking forward to grow big you have to drop these kind of tedious task
• MX record information should be provided as clear as possible. I'm hoping it's placed in my client area dashboard to see my server uses which MX record, and then you put a link at there to direct user into knowledge base if they need more detailed information
• single click login is great if it's possible, if your url doesn't point to port 443 I think it's wise just use https by default

If you ask me "Would you recommend onepoundemail service to others?", I have two answers:

• Yes, provided i know they're familiar with directadmin and have at least some basic technical skill
• No, if they're just average pleb that doesn't know what a http error code is

Onepoundemail is off to a great start, as all of their essential features function flawlessly and they offer competitive pricing. If they continue to enhance their service, I would definitely consider using them again for my future email hosting needs.