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
Hey @PulsedMedia
Is your "Storage Box" like a jailed VPS with a lot of storage? It seems pretty nice. Can I only buy one without seedboxes and use it externally?
Next Business Day + 5 == Next Business Day + 5 Business Days.
It's a shared account like seedboxes, limited jailing since it's shared server.
You can buy just the storage individually. It has fully featured software (for now) but very very limited performance capacity. We are forced to start limiting software soon tho due to rampant abuse.
Gotcha, that's a tough sell compared to hosting at home and amazon prime-ing hardware replacements
There's right now pressure for prices to increase; Dynamic pricing seems to be working really nicely. Some fine-tuning on the values will always be needed. Who wants charts to see history?
I do atleast!
Question is, full blown grafana or just simple charts showing hourly for past 30 to 90d
I'm waiting for a bug to order at 15€.
I think you'll have to wait for a reaaaally long time for that.
Bugs fixed now.
Tho once we add a lot of nodes at once, it could potentially happen for something.
Current nodes on offer update;
@PulsedMedia Any updates on this slow download? The ticket also not been responded. The password still default, feel free to login and check, i cannot use it since very laggy
We cannot change the speed of 3rd parties, this on your own test shows you are getting full speed.
Those test servers are always busy, and this testing is invalid for any debugging purposes and will not translate into real world usage in any manner.
See results from others;
How could this be improved upon?
Not in any manner.
You are on the same network, getting same access speed, and might even be connected to the same switch.
If you want better speed test results, contact the speedtest server providers and ask them to ugprade.
okay blaming speedtest provider won't get anywhere,
i can do many other test like download test to looking glass, still speed is same, i tried with my other server i got full speeds. FYI: there are always many other factors why speed is limited, for example these ips were previously Blackholed so they were re-routed, etc, I've had similar problems in the past. I just want you or you team check abou this.
Download speed to Germany, Hetzner
https://prnt.sc/rvam4ftPGkL6
Download speed to Finland, Hetzner
https://prnt.sc/jrvPw8qwyo0E
Download speed to France, Vultr
https://prnt.sc/vnl6QP5Nx0Ap
Only one that matters is this;
If you can get to one full speed, it means from our network you can get to anywhere full speed, you are not limited by our side. Don't expect us to fix Novoserve or Hetzner networks (or whatever is in between, such as Ficix)
So rest is irrelevant. We can't do anything about it, cannot build dedicated fiber link to every possible place, and upgrade 3rd party networks and servers.
At the very minimum, for any even shred of validity in your testing, you need to do MTR 1000+ packets both ways, at the same time; At the very least, just to start with.
Let us know if there is an actual problem. Right now, there is none, you got full 1Gbps access to your server, as shown by yourself.
so that mean, in my case, the connection only truly 1G port if to NL only and the rest not even reach 100Mbps or that is the best network it can provided
IPerf3 test to GRA and RBX OVH
`root@le2-3-156:~# iperf3 -c rbx.proof.ovh.net -4 -R -p 5205
Connecting to host rbx.proof.ovh.net, port 5205
Reverse mode, remote host rbx.proof.ovh.net is sending
[ 5] local 185.148.3.156 port 53822 connected to 141.95.207.211 port 5205
[ ID] Interval Transfer Bitrate
[ 5] 0.00-1.00 sec 392 KBytes 3.21 Mbits/sec
[ 5] 1.00-2.00 sec 378 KBytes 3.09 Mbits/sec
[ 5] 2.00-3.00 sec 395 KBytes 3.23 Mbits/sec
[ 5] 3.00-4.00 sec 370 KBytes 3.03 Mbits/sec
[ 5] 4.00-5.00 sec 403 KBytes 3.30 Mbits/sec
[ 5] 5.00-6.00 sec 402 KBytes 3.29 Mbits/sec
[ 5] 6.00-7.00 sec 451 KBytes 3.69 Mbits/sec
[ 5] 7.00-8.00 sec 390 KBytes 3.20 Mbits/sec
[ 5] 8.00-9.00 sec 440 KBytes 3.60 Mbits/sec
[ 5] 9.00-10.00 sec 423 KBytes 3.46 Mbits/sec
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.04 sec 4.11 MBytes 3.43 Mbits/sec 75 sender
[ 5] 0.00-10.00 sec 3.95 MBytes 3.31 Mbits/sec receiver
iperf Done.
root@le2-3-156:~# iperf3 -c gra.proof.ovh.net -4 -R -p 5205
Connecting to host gra.proof.ovh.net, port 5205
Reverse mode, remote host gra.proof.ovh.net is sending
[ 5] local 185.148.3.156 port 33596 connected to 141.94.30.145 port 5205
[ ID] Interval Transfer Bitrate
[ 5] 0.00-1.00 sec 362 KBytes 2.96 Mbits/sec
[ 5] 1.00-2.00 sec 338 KBytes 2.77 Mbits/sec
[ 5] 2.00-3.00 sec 383 KBytes 3.14 Mbits/sec
[ 5] 3.00-4.00 sec 441 KBytes 3.61 Mbits/sec
[ 5] 4.00-5.00 sec 382 KBytes 3.13 Mbits/sec
[ 5] 5.00-6.00 sec 348 KBytes 2.85 Mbits/sec
[ 5] 6.00-7.00 sec 444 KBytes 3.64 Mbits/sec
[ 5] 7.00-8.00 sec 406 KBytes 3.32 Mbits/sec
[ 5] 8.00-9.00 sec 372 KBytes 3.05 Mbits/sec
[ 5] 9.00-10.00 sec 379 KBytes 3.10 Mbits/sec
[ ID] Interval Transfer Bitrate Retr
[ 5] 0.00-10.04 sec 3.95 MBytes 3.30 Mbits/sec 67 sender
[ 5] 0.00-10.00 sec 3.76 MBytes 3.16 Mbits/sec receiver
iperf Done.
root@le2-3-156:~#
`
@vba Ping the gateway, if it doesn't show any packet loss, and you got from one place full 1Gbps/1Gbps you can conclude it's not on our side the issue.
If you want us to pursue further, do the MTR 1000+ packets at the same time both ways.
We will not spend time ghost hunting bullshit for hours. You have to prove there is an actual issue, beyond shadow of doubt.
Found your ticket; even your own testing shows there's no issue there too on the screenshot as well. If one target gets full speed === It's not on our network.
this speed is not satisfying, hopefully I can cancel the server and get a refund
I see, so the proof of iperf3 is no longer valid
i forget why people use iperf3 now
https://www.speedtest.net/apps/cli
(just DO NOT use the python version, that's known borked)
https://network-speed.xyz/
OVH has had issues for more than a decade, especially their speedtest servers.
You have not pointed any actual issue, only 3rd parties being congested.
Also, tested myself; no issues with OVH.
Server 1 (different switch);
Server 2 - This is on same switch, or the neighbouring switch than yours (both have exact same uplink config);
None of the switches involved with MD has congestion, peak at the busiest switch is ~40%. Bursted peak.
Like shown by your own testing; You are not being congested at our end. You got one of the speedtest servers ~950Mbps both ways; Which means that throughput is fine at our end.
You cannot force any banwidth or datarates forcefully; Just because you have XYZ speed does not mean that PointB has that to offer.
Can you truly ask for more than this?
Do the damn gateway ping test If there's an issue isolated to your node, you have to do the tests. We cannot do them, we don't have crystal balls. Start with pinging the damn gateway.
Again, what i reported is my server, please login to my server do that test and see if you can reach that speed. something wrong with my server, that is why said many times, dont compare to other server, that is fresh server no application installed. probably cable issue or other, i dont know. if hardware or cable issue, those are things out of our hands are users
my ip is 185.148.3.xx6
and here what you just asked,
OK, I lost interest in using your services, after seeing the way you treat customers in solving problems. just take the money, I don't care anymore
if anyone here want to try this server, feel free to dm i will send the server detail, asking provider to login to check by himself is too much
This is not a proper support channel neither; and it's kinda nasty trying to force priority concierge support like this.
You are getting good speed.
Rest of the upload speeds (usually the limiting factor) are all good.
Download speeds are suss from many targets tho; But rest shows you are getting good speeds.
You need to determine if there is packet loss, or that it's not your configs which made this happen. These servers are self-managed. You know that means? You have to figure basics out yourself. You can't ask us to spend hours on wild goose chases, you have to provide the rock solid evidence.
Obviously if you get full speed to any target, that means there's no issue on the node itself, or the network itself. You wouldn't get good speeds to any target if there was an hardware issue.
What your results have common is long distance drops download speeds; This leads me to believe you have changed kernel settings, possibly MTU window in attempt to get higher speeds. Which means they could be all out of whack.
MTR and/or ping testing is needed, to determine if there is packet loss. No testing? Don't expect us to do it on your behalf.
ok fuck this shit.
Thanks for making this so fucking ridiculously easy; as to having to guess even whatever might be your server because in your entitlement you cannot be arsed to even tell that, just accuse and demand special privileges, priority concierge support in non-support channel.
Paid the toll of 2 burnouts, peered into the crystal ball and found what the fuck is your node even because you don't bother to even tell that. Crystal ball is again well hidden not to accidentally peer into it and having to pay for extra burnouts.
Found neighbor node is free for testing, connected to the same switch, with exactly the same hardware config.
Yea all low latency connections are fast both ways, but high latency not for downstream. Only comes with high latency; But not all connections/routes. That might be an actual switch failure. But can't do anything about it during a holiday ~midnight, not going to risk having 1000s users down until morning to satisfy your entitlement.
But F* for making this ridiculously difficult.
There are 4 potential reasons;
All of which needs physical presence to check and be safe in case something goes rotten. TOR switch has been rebooted, no change. Config gone through line by line.
My money tho is on the router ASIC chip messing up with FIB. Those issues don't leave errors behind them, and can manifest in the most uncanny ways. Just has the feel that those packets drop into CPU out of the ASIC on that linecard.
TOR is regular 48 port switch with ultra deep buffers, so issue is limited to them and i will absolutely not risk the whole rack (or couple) going down during a holiday.
Links are barely used on that switch
After holidays i'll pick a few of these switches from warehouse, and go replace this one.
Was scheduled to install one new one of these on that particular rack regardless during January.
Sorry i have no magic wand which would magically replace a switch, network cables etc. remotely, or magic wand which could reset that particular ASIC chip // linecard, but none of the others safely. (in theory i can flip that linecard on and off remotely ... but it's too risky)
I got it for the quick sync
!!! UPGRADED MY SERVER Order 8486182088
`# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Yet-Another-Bench-Script
v2023-11-30
https://github.com/masonr/yet-another-bench-script
## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ##
Thu 28 Dec 2023 04:21:54 PM EET
Basic System Information:
Uptime : 0 days, 0 hours, 38 minutes
Processor : Intel(R) Core(TM) i5-7500T CPU @ 2.70GHz
CPU cores : 4 @ 1200.059 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM : 31.2 GiB
Swap : 0.0 KiB
Disk : 3.6 TiB
Distro : Debian GNU/Linux 11 (bullseye)
Kernel : 5.10.0-26-amd64
VM Type : NONE
IPv4/IPv6 : ✔ Online / ❌ Offline
IPv4 Network Information:
ISP : Magna Capax Finland Oy
ASN : AS203003 Magna Capax Finland Oy
Host : Pulsed-Media
Location : Helsinki, Uusimaa (18)
Country : Finland
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/nvme0n1p2):
iperf3 Network Speed Tests (IPv4):
Provider | Location (Link) | Send Speed | Recv Speed | Ping
----- | ----- | ---- | ---- | ----
Clouvider | London, UK (10G) | 836 Mbits/sec | 926 Mbits/sec | 35.3
ms
Scaleway | Paris, FR (10G) | busy | busy | 34.7
ms
NovoServe | North Holland, NL (40G) | 822 Mbits/sec | 929 Mbits/sec | 30.4
ms
Uztelecom | Tashkent, UZ (10G) | 789 Mbits/sec | 885 Mbits/sec | 102 ms
Clouvider | NYC, NY, US (10G) | 759 Mbits/sec | 885 Mbits/sec | 99.8
ms
Clouvider | Dallas, TX, US (10G) | 739 Mbits/sec | 860 Mbits/sec | 137 ms
Clouvider | Los Angeles, CA, US (10G) | 664 Mbits/sec | 832 Mbits/sec | 167 ms
Geekbench 6 Benchmark Test:
Test | Value
|
Single Core | 1318
Multi Core | 3893
Full Test | https://browser.geekbench.com/v6/cpu/4171825
YABS completed in 10 min 47 sec
`
@PulsedMedia new order just arrived, thanks
Conjured has been a +1 month for a server of Jason5545's!
so .... we wrote a think @vba might want to take a looksie at;
Comprehensive Network Diagnostics!
Probably needs more specific step-by-step still, but lots of information there, and why it's difficult to diagnose stuff.
Most Importantly; How to test and report.
So go ahead; https://wiki.pulsedmedia.com/index.php/Network_Diagnostics
Current dynamic pricing pressure of available nodes (yea i reckon we need better output for sharing!)
Here is the updated list of offers (ought to be same sorting as above)
Few more nodes added, updated list;
Just received your email about the network issue on some nodes, and just checked mine although my CPU is 7th gen, looks speechless but in a good way, out of my expectation.
` Location Latency Loss DL Speed UP Speed Server
ISP: Magna Capax Finland Oy
Nearest 0.67 ms 0.0% 934.01 Mbps 852.38 Mbps Valoo - Helsinki
Kochi, IN 188.76 ms 0.0% 943.27 Mbps 454.28 Mbps Asianet Broadband - Cochin
Bangalore, IN 154.51 ms 0.0% 945.42 Mbps 515.27 Mbps Bharti Airtel Ltd - Bangalore
Chennai, IN 161.00 ms N/A 949.76 Mbps 508.61 Mbps Jio - Chennai
Mumbai, IN 149.10 ms 0.0% 944.21 Mbps 535.74 Mbps i3D.net - Mumbai
Delhi, IN 204.85 ms 0.0% 944.55 Mbps 402.26 Mbps Tata Play Fiber - New Delhi
Seattle, US 168.16 ms N/A 944.55 Mbps 477.03 Mbps Comcast - Seattle, WA
Los Angeles, US 159.19 ms 0.0% 941.70 Mbps 509.83 Mbps ReliableSite Hosting - Los Angeles, CA
Dallas, US 137.11 ms 0.0% 697.13 Mbps 575.58 Mbps Hivelocity - Dallas, TX
Miami, US 136.43 ms 0.0% 943.99 Mbps 598.19 Mbps AT&T - Miami, FL
New York, US 109.09 ms 0.0% 942.01 Mbps 731.34 Mbps GSL Networks - New York, NY
Toronto, CA 112.21 ms 0.0% 948.39 Mbps 728.08 Mbps Rogers - Toronto, ON
Mexico City, MX 175.73 ms N/A 946.10 Mbps 457.61 Mbps INFINITUM - Mexico City
London, UK 35.72 ms 0.0% 941.58 Mbps 827.79 Mbps VeloxServ Communications - London
Amsterdam, NL 37.81 ms 0.0% 941.54 Mbps 830.49 Mbps 31173 Services AB - Amsterdam
Paris, FR 39.92 ms N/A 941.65 Mbps 820.68 Mbps Axione - Paris
Frankfurt, DE 30.74 ms 0.0% 918.53 Mbps 825.09 Mbps 23M GmbH - Frankfurt am Main
Warsaw, PL 17.39 ms 0.0% 941.33 Mbps 833.48 Mbps Play - Warszawa
Bucharest, RO 53.46 ms 0.0% 940.30 Mbps 810.18 Mbps Vodafone Romania Fixed Bucharest - Bucharest
Moscow, RU 43.75 ms 0.0% 940.82 Mbps 832.16 Mbps MegaFon - Moscow
Jeddah, SA 94.34 ms 0.0% 939.93 Mbps 788.63 Mbps Saudi Telecom Company
Dubai, AE 154.46 ms 0.0% 944.93 Mbps 533.65 Mbps du - Dubai
Fujairah, AE 140.10 ms 0.0% 938.85 Mbps 559.03 Mbps ETISALAT-UAE - Fujairah
Istanbul, TR 60.70 ms 0.0% 951.83 Mbps 813.13 Mbps Turkcell - Istanbul
Tehran, IR 84.72 ms 0.0% 916.58 Mbps 764.35 Mbps Asiatech - Tehran
Tokyo, JP 281.21 ms N/A 937.53 Mbps 272.17 Mbps fdcservers.net - Tokyo
Shanghai, CU-CN 308.92 ms 0.0% 943.06 Mbps 248.63 Mbps China Unicom 5G - Shanghai
Nanjing, CT-CN 271.16 ms 1.7% 35.46 Mbps 57.36 Mbps China Telecom JiangSu 5G - Nanjing
Hong Kong, CN 266.65 ms N/A 941.92 Mbps 280.53 Mbps STC - Hong Kong
Singapore, SG 175.57 ms 0.0% 940.91 Mbps 467.93 Mbps i3D.net - Singapore
Jakarta, ID 193.74 ms 0.0% 942.78 Mbps 351.22 Mbps PT. Telekomunikasi Indonesia - Jakarta
Avg DL Speed : 904.02 Mbps
Avg UL Speed : 589.12 Mbps
Total DL Data : 40.47 GB
Total UL Data : 22.85 GB
Total Data : 63.32 GB
Duration : 14 min 59 sec
System Time : 31/12/2023 - 08:39:22 EET
Total Script Runs : 32776
Result : https://result.network-speed.xyz/r/1704003830_095PPQ_GLOBAL.txt`
!!! UPGRADED MY SERVER Order 8209445107
I got it for idling (for now).
Unfortunately, my node is having network performance issue.
I hope it will be fixed soon.
Oh even 6 gen? I thought only 8 gen is affected.
+1 Month added
Do the network-speed to get more results and total average.
also lspci output
apt install lm-sensors and sensors output too
There's a hunch this might be with realtek chips and we also need to validate if those chips may be overheating.
This will take time to solve potentially. Network testing is always excruciatingly difficult when it goes beyond simple obvious things (signal degradation, cable failures, optic failures etc.); Even when not going outside your own network.
Corrected link; https://result.network-speed.xyz/r/1704003830_095PPQ_GLOBAL.txt
(code is just empty space on start of line here on LET that last character got added to the url)
can you grab a lspci, sensors output as well?
Here
```bash
root@le2-3-110:/server# lspci
00:00.0 Host bridge: Intel Corporation Xeon E3-1200 v6/7th Gen Core Processor Host Bridge/DRAM Registers (rev 05)
00:02.0 VGA compatible controller: Intel Corporation HD Graphics 630 (rev 04)
00:14.0 USB controller: Intel Corporation 200 Series/Z370 Chipset Family USB 3.0 xHCI Controller
00:14.2 Signal processing controller: Intel Corporation 200 Series PCH Thermal Subsystem
00:17.0 SATA controller: Intel Corporation 200 Series PCH SATA controller [AHCI mode]
00:1d.0 PCI bridge: Intel Corporation 200 Series PCH PCI Express Root Port #9 (rev f0)
00:1f.0 ISA bridge: Intel Corporation 200 Series PCH LPC Controller (Q270)
00:1f.2 Memory controller: Intel Corporation 200 Series/Z370 Chipset Family Power Management Controller
00:1f.4 SMBus: Intel Corporation 200 Series/Z370 Chipset Family SMBus Controller
00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection (5) I219-LM
01:00.0 Non-Volatile memory controller: Micron/Crucial Technology P2 NVMe PCIe SSD (rev 01)
root@le2-3-110:/server# sensors
coretemp-isa-0000
Adapter: ISA adapter
Package id 0: +59.0C (high = +74.0C, crit = +80.0C)
Core 0: +59.0C (high = +74.0C, crit = +80.0C)
Core 1: +58.0C (high = +74.0C, crit = +80.0C)
Core 2: +57.0C (high = +74.0C, crit = +80.0C)
Core 3: +57.0C (high = +74.0C, crit = +80.0C)
nvme-pci-0100
Adapter: PCI adapter
Composite: +58.9C (low = -0.1C, high = +84.8C)
(crit = +94.8C)
Sensor 1: +58.9C (low = -273.1C, high = +65261.8C)
Sensor 2: +62.9C (low = -273.1C, high = +65261.8C)
Sensor 8: +58.9C (low = -273.1C, high = +65261.8C)
I got an Intel one.