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
Why do everyone want to offload a ks-le, it is a great box for that price and can replace quite a few vps's ?
I currently have a ks7-ca, and cant decide if i should upgrade or retain. ks-le is like 50% costlier though.
(btw speedtest isn't the most reliable tool for assessing available bw)
Yabs gives the same results so I guess it's fine.
I'm interested if you drop it!
You're on the list.
Just out of curiosity, what BHS Dc is your server located?
Speed is on BHS 3. Storage is on BHS6.
Add me too in the list for 2 x 4 TB box.
You're added to the list.
I am in the same boat here. I finally got my ID verified a day ago, now waiting patiently for provisioning.
To get my ID verified I had to create a second ticket wherein someone raised the priority of the ID request ticket otherwise I feel like it would be stuck in limbo forever.
I have no active tickets so I just have to hope they will provision when able. It sure is disconcerting to see them sell and instantly provision servers for other people while I have paid and continue waiting two weeks and counting.
Keep in mind they can "fix" the bandwidth by properly clamping it to 100 Mbit at any time, whereas they certainly won't go into people's servers to rip out the 4TB HDDs.
Do we have evidence that such a downgrade ever happened?
(... and considering the ease with which OVH replaces faulty hdds, what if one of your 4TB goes boom? you may end up with 2TB. Bye bye sweet softRAID.)
Most likely you will, so hope that it doesn't.
But what this has to do with it? At least the enterprise Hitachi on offer don't fail with "ease".
Even then though, mdadm RAID doesn't have to use full drives, so you could have 2x2TB in RAID, and the tail of the larger drive as separate.
Yep. Hope that they don't downsize bandwidth, too. Lots of hoping to be done.
What has the reliability of Hitachi Ent HDD to do with the well-known (and much appreciated) OVH policy of replacing HDDs in a snap (even non-faulty ones actually, if you see what I mean)... and thus the non-null likelihood of ending up with some 2TB ones at the end of the day? OTOH, I can find no evidence of OVH ever downgrading "excessive" bandwidth. Hence my question. (And no answer yet. Anyone?)
Sure (I do it routinely, although with HW raid). Or even don't use raid at all. Or just use 10GB of the 2x4TB. I'm sure some people do. So what? In contrast with your evocation of a possible (but never observed before afaik) downgrade in bandwidth, I just provided an instance among other of a non-unrealistic drawback that might plausibly result from a plausible downgrade in hardware (similarly never observed either afaik, they tend to upgrade stuff).
The "cost" to get someone to go looking for your server just to swap out your 4 TB drive is probably higher than the price difference between two very old 4 TB and 2 TB drives.
Disk failure, how funny.
Does anyone remember the free upgrades from i3 to i5?
Mainboard, CPU, Memory, if anything of this dies, I am getting a downgrade.
These disks tend to turn out very solid, and even if, one fails, you still have the other.
I've actually seen mobos die at Kimsufi (due to some electrical fan-hitting sh*t iirc). That would be a bitch
I'm still not sure what you mean. They don't replace the HDD abruptly on their own, unless you ask (or consent). And will not in this case either, just to reclaim the mistakenly given 4TB and provision you with the specced 2. That's just out of the question, and not a "hope" matter. You supposedly already have your data there, how do you imagine they go ahead and pull the HDD from under you and cause you data loss.
Kimsufi France network is faster and much more stable than Kimsufi Canada.
If anyone is interested I can transfer Kimsufi Canada for $20. The monthly renewal is $18 per account.
For transfer, the buyer must have an account with Kimsufi US.
If it's 500Mbps or 1Gbps upload I'll take it lol.
Of course they don't. You're severely misrepresenting my point, that's akin to a strawman.
One more (last) time : in the event (not unlikely, even Hitachis can fail) that a HDD fails (or is showing worrying signs and you proactively force OVH to replace it, wink wink)... you are far from certain of obtaining a 4TB replacement to keep enjoying that "special" KS (and e.g. recreate your 4TB raid) when the server model explicitly states 2x2TB. So, at the end of the day, it is not extraordinarily implausible that one would end up with 4TB+2TB. Or 4TB+3TB (if they ran out of 2TBs - not so likely either, they still have lots of them). See what I mean? Drawback of having "won" the Kimsufi lottery.
In the meantime, does anyone have evidence of them ever downgrading "bonus" bandwidth? (voluntarily, or after maintenance, whatever, I don't care). Can't find anything. Did it ever happen, or not?
I never had a disk replacement on Kimsufi yet, despite services since 2016.
Mainboard died a few times, which the monitoring notices and gets replaced automatically fine.
If your disk dies, your system goes to read only, which does not trigger anything but at this point you want it replaced right.
100 Mbps upload
1000 Mbps download
If someone want to sell their KS-LE ( FR or CA ) I will take it.
Thank you !
I have one in France on a French account. I can let it go as I don't need it (thanks to my SYS-2-SAT-32).
I still don't know how to format YABS results properly !!!
## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Yet-Another-Bench-Script
v2021-10-09
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Thu Dec 2 08:15:02 UTC 2021
Basic System Information:
Processor : Intel(R) Xeon(R) CPU E5-1620 v2 @ 3.70GHz
CPU cores : 8 @ 3900.000 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 31.2 GiB
Swap : 1024.0 MiB
Disk : 1.8 TiB
fio Disk Speed Tests (Mixed R/W 50/50):
Block Size 4k (IOPS) 64k (IOPS)
Read 790.00 KB/s (197) 10.97 MB/s (171)
Write 827.00 KB/s (206) 11.46 MB/s (179)
Total 1.61 MB/s (403) 22.43 MB/s (350)
Block Size 512k (IOPS) 1m (IOPS)
------ --- ---- ---- ----
Read 42.84 MB/s (83) 54.58 MB/s (53)
Write 45.19 MB/s (88) 58.52 MB/s (57)
Total 88.04 MB/s (171) 113.10 MB/s (110)
iperf3 Network Speed Tests (IPv4):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 95.6 Mbits/sec | 899 Mbits/sec
Online.net | Paris, FR (10G) | 95.7 Mbits/sec | 877 Mbits/sec
WorldStream | The Netherlands (10G) | 95.4 Mbits/sec | 926 Mbits/sec
WebHorizon | Singapore (1G) | 15.7 Mbits/sec | 159 Mbits/sec
Clouvider | NYC, NY, US (10G) | 33.4 Mbits/sec | 417 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 30.3 Mbits/sec | 291 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 22.4 Mbits/sec | 180 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | 16.1 Mbits/sec | 161 Mbits/sec
iperf3 Network Speed Tests (IPv6):
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 94.4 Mbits/sec | 909 Mbits/sec
Online.net | Paris, FR (10G) | 94.4 Mbits/sec | 912 Mbits/sec
WorldStream | The Netherlands (10G) | busy | busy
WebHorizon | Singapore (1G) | 12.6 Mbits/sec | 124 Mbits/sec
Clouvider | NYC, NY, US (10G) | 39.1 Mbits/sec | 260 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 21.9 Mbits/sec | 193 Mbits/sec
Geekbench 5 Benchmark Test:
Test | Value
|
Single Core | 914
Multi Core | 3734
Full Test | https://browser.geekbench.com/v5/cpu/11370387
I've picked two KS-LE as well in different regions as I wasn't sure which one I was going to use I'm letting one of them go and keeping the other. FR/IE region if anyone is interested.
What are these different benchmark results on dedis?
I have seen everything from 3.5k to 3.7k even 3.8k.
Same CPU, I can replicate this on booth of my machines, CONSISTENTLY.
Same Kernel, same OS, less benchmark.
Anyone care to explain?
I did not look at the CPU temp's yet.
The difference between Canada and France locations in terms of network speed is day and night as well.
A congested network has nothing to do with cpu benchmark results.