New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
I Have 3 VPS with them, all in Germany.
Here are the yabs for the 2 linux.
`# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Yet-Another-Bench-Script
v2022-05-06
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Sun 08 May 2022 09:02:32 AM CEST
Basic System Information:
Uptime : 85 days, 13 hours, 36 minutes
Processor : AMD EPYC 7282 16-Core Processor
CPU cores : 4 @ 2794.748 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 7.8 GiB
Swap : 2.0 GiB
Disk : 97.9 GiB
Distro : Ubuntu 20.04.3 LTS
Kernel : 5.4.0-62-generic
fio Disk Speed Tests (Mixed R/W 50/50):
iperf3 Network Speed Tests (IPv4):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 198 Mbits/sec | 199 Mbits/sec
Online.net | Paris, FR (10G) | 192 Mbits/sec | 199 Mbits/sec
Hybula | The Netherlands (40G) | 199 Mbits/sec | 198 Mbits/sec
Clouvider | NYC, NY, US (10G) | 140 Mbits/sec | 194 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 174 Mbits/sec | 188 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 180 Mbits/sec | 186 Mbits/sec
Geekbench 5 Benchmark Test:
Test | Value
|
Single Core | 752
Multi Core | 2547
Full Test | https://browser.geekbench.com/v5/cpu/14781526`
and for the other one:
`# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Yet-Another-Bench-Script
v2022-05-06
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Sun 08 May 2022 09:01:49 AM CEST
Basic System Information:
Uptime : 100 days, 14 hours, 15 minutes
Processor : AMD EPYC 7282 16-Core Processor
CPU cores : 4 @ 2794.748 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 7.8 GiB
Swap : 2.0 GiB
Disk : 97.9 GiB
Distro : Ubuntu 20.04.3 LTS
Kernel : 5.4.0-91-generic
fio Disk Speed Tests (Mixed R/W 50/50):
iperf3 Network Speed Tests (IPv4):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 198 Mbits/sec | 199 Mbits/sec
Online.net | Paris, FR (10G) | 190 Mbits/sec | 199 Mbits/sec
Hybula | The Netherlands (40G) | 199 Mbits/sec | 200 Mbits/sec
Clouvider | NYC, NY, US (10G) | 146 Mbits/sec | 194 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 106 Mbits/sec | 1.41 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 176 Mbits/sec | 186 Mbits/sec
Geekbench 5 Benchmark Test:
Test | Value
|
Single Core | 729
Multi Core | 2615
Full Test | https://browser.geekbench.com/v5/cpu/14781522`
I have never had to deal with the support so can't comment on this.
Hetrixtool uptime reported is 99.9841% and 99.7047% but to be honest I'm clearly responsible for stopping the later (Dev) server manually a few times.
I'm happy with them but don't host any mission-critical app on them either, even the Mailing App could cope with a few downtimes.
The only thing I really hate is their invoices... what a mess...
Cheers
Eh. Made a decent seedbox. 4/5
i voted 1 because they never activate my order and always cancel it.
They offer great service at unbeatable prices on the VPS market.
Try looking in your firewall logs and you may notice search hits for the b'stards that port scan and attempt hacks, from Contabo.
All the ones I see are from chinese residential IP's and alibaba cloud
lfd: (PERMBLOCK) 154.12.239.122 (US/United States/vmi868424.contaboserver.net) has had more than 4 temp blocks in the last 86400 secs - Sat Jun 4 01:09:37 2022
Random grab from one recent Ryzen VPS.
I guess it depends on the IP range. I would recommend to switch the ssh port and install a honeypot on port 22.
Whoosh! The point was, Contabo host malicious lusers. (..as do many other providers.)
I was both curious and bored for a minute and I always like to keep an eye on UK locations so I gave their UK VPS 8.99 tier a try. Dogshit. That is all.
All their other USA locations are singlehomed to Cogent (even on IPv6, so goodbye HE customers!) - St. Louis is the one that has a relatively decent network connectivity.
I'm not sure they even have Ryzen nodes, AFAIK it's all Epycs.
But yes, I wouldn't be surprised if they had tons of abusers. As I already said multiple times, their hardware is really fine (incl. top-class SSDs and NVMes) and their network is OK too (but not great), but their node and VM monitoring seems to be poor, so many (most?) nodes look "overloaded"/"high steal" due to abusers.
That said, while I wouldn't put a high load, very high disk IO or latency sensitive application on their VPS the way some people paint Contabo / @contabo_m VPSs is just ridiculous. Those systems are easily good and fast enough for normal use cases and certainly not worse, and indeed even better, than many, many VPS offered here at LET. Look at their "oh so poor 200 Mb/s only" connectivity; first, that's a very normal and common limit, second you actually get 200 Mb/s (which you don't with many other 200 Mb/s providers), and third 95+% of Contabo's customer do not even use and need half of that. A propos "Cogent only": Their clearly fattest path is 1299/Telia and AFAIK there is no single homed Contabo location, not a single one.
Looking at it with a reasonable expectation those VPS clearly are among the better ones for normal users and one gets a lot of bang per buck.
And btw, one doesn't grow to become a global Tb/s provider with 6 digit customer - and server! - numbers if one doesn't offer attractive products and services.
So, all in all, is Contabo a top-performance provider? Nope - but it's a provider offering lots of bang for one's bucks plus some cream on top like e.g. 32 TB traffic volume or really nice VPS at very reasonable prices in SGP.
Contabo never changes.
Just choose netcup/phpfriends for Germany and Hetzner for US. Enjoy 1-10Gbps good quality network for pennies and no big problems with abusers/overselling.
Chose Contabo only for things noncritical things /as second failsafe server / if you need SGP with 32TB traffic.
https://bgp.tools/prefix/154.53.48.0/21#connectivity
He means it's his ryzen vps getting attacks from contabo's network
Pardon me, but that's a bunch of anti-Contabo BS. phpfriends I didn't test/benchmark, so I won't say anything about them, netcup is nice, I like them and have a VDS with them myself but I fail to see why you say they're "better", and Hetzner, I did test and frankly, I was quite disappointed. My cheap Contabo VPS easily wins over the 2 Skylake cores, 4 GB Hetzner VPS I tested (roughly same price as the small Contabo VPS). Single core performance and spread similar, multi-core very clear win for Contabo, AES and RSA similar, 4k/4t disk performance ca. 5 times that of the Hetzner VPS and IOPS over 4 times higher. Network results were largely in the same ballpark but Hetzner seems to not have limited the bandwidth which gave them a significant advantage to a few (as in '3 or 4') european targets (like e.g. UK, LON).
Get it already, every provider in that price range has to make compromises somewhere. Some simply give you crappy stone-age E5-v2 cores (both Hetzner and Contabo dont, they both use decent modern processors), many give you crappy disks (Contabo doesn't and neither does Hetzner I guess but my DE Contabo VM happens to be NVMe while Hetzner (I presume) was only SSD but by no means a crappy one), some have crappy connectivity, many have BW limits, with some service is poor, whatever it may be but all of them have to and make compromises.
Do I say that because I'm a Contabo fanboy? Nope, I do see their problem and I do wish that they soon get their monitoring up to par and reign in on abusers; I like everyone else hate to basically play lottery when getting a VPS ("will I end up on an abuser ridden node or on a lucky one?").
So, unlike many I do not consider Contabo an excellent provider - but neither is it crappy as some here like to insinuate.
Plus - and that's a big plus: I've seen very few providers who were so transparent and provided even info that most providers would keep closed with teeth and claws.
From your source:
.
OK, maybe I got him wrong, but anyway that was more of a side note.
Edit: And if one really needs outstanding connectivity there's @Hybula (but understandably and justifiably rarely in the Contabo or Hetzner price range). Next to Hybula pretty much all cheap providers (and many expensive ones too) look crappy in terms of connectivity.
So, if someone wants to think I'm some providers fanboy then that'd be Hybula and NexusBytes.
I use Contabo for my YunoHost self-hosted Apps. never had any problem .. can't complain about what I get for cheap.
>
for the specific subnet, the upstream is only cogent given the link
OK, do you happen to know (with reasonable certainty) how Contabo uses that subnet?
And still, I guess they don't aquire (not the cheapest) bandwidth, e.g. from 1299 or L3, just for the fun of it. Maybe the subnet you're talking about isn't representative.
I know they use that subnet because I had a VPS on it
Have fun arguing with this guy, where he’d go around in circles claiming that there’s a probability that statement X is not true, even though you just presented evidence to the contrary. See also: the HS involucration thread, and the “benchmark war” (not my words) thread.
No, I thanked him.
No, he didn't present evidence, he just made a statement that may or may not be true.
No problem, but typical for you to make up things and/or to "enhance" facts.
1/1 for me, pretty bad.
contabo was for me one of the worst expierience ever. The disks were usually slow as floppy drives. Not worth the money.