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.
seflow [split from multi-year necro]
This discussion was created from comments split from: seflow review - stay far away: Scammers and liars.
Comments
Me to similar problem with seflow.it they are not honest services look at:
http://www.webhostingtalk.com/showthread.php?t=1778129&p=10173301&posted=1#post10173301
I have a issue with seflow.it.
I was use Hetzer dedicated servers. But i try seflow.it dedicated servers. It's not good services.
I have a:
Intel I7 8700 4,2Ghz Hexa-Core
480GB SSD
4000GB sATA
250Mbps Unmetered bantwith.
Problem is network!
Its not 250mbps. I am live in Turkey but i have 6mbps upload/download speed from my server to home(i have 20mbps internet).
I am making speed test for europe/uk/usa and many countries but average 10mbps speed! Its not 250mbps. I am open ticked from seflow.it but they are telling lies and many excuses. And they can send my money back.
United statest tests:
https://www.speedtest.net/result/8532110422
https://www.speedtest.net/result/8532113090
United Kingdom test:
https://www.speedtest.net/result/8532115152
And another Turkey test
https://ttnet.speedtestcustom.com/re...7-97caa74609e5
https://turknet2.speedtestcustom.com...1-43a3b566822d
Last words not buy anythink from SEFLOW.it
We know. Sorry to hear you did not.
Sweet mother of Taco loving Jesus!
Thanks, i buy with paypal.Try my chance with paypal for justice. ı write it, because i want to google index it. All people see it, no one buy again from seflow s.n.c
Just accept that, ask them to remove all your customer data as per GDPR and call it a day.
I am not siding with Seflow nor do I care about them for any reason, however, often times these types of issues are more so caused by a lack of understanding of how networks work.
Just because your connection has the possibility of reaching 20Mbit, this doesn't mean every location you download from will be able to max your connection. This is HIGHLY dependent on the routes between you and that location. In fact, you can even find cases where you have 20ms ping to an IP but throughput to the network is absolutely abysmal. Just because you have good latency doesn't mean the route between you can actually support your full throughput and the limitation can be on either side. In fact, more often than not, this type of slowness is caused by that route using a high-end Tier 1 network and your provider only having limited capacity to that high-end network causing you to be quality of serviced down to a slower speed. Again, in this case you can have low latency but also low throughput.
There are SOME things network providers can do to try and work around this in some cases, such as try to change network blend used to reach your network, but even then, that doesn't mean the link between you isn't at capacity or that this route will be static. Routing is dynamic, so even if they fixed it for now, it could revert later during a normal BGP route update.
Overall it just seems you were trying to be super cheap and instead of purchasing a service that works well to your connection you just purchased something cheap and hoped it would be able to fill your connection. In the end you should just cancel services going forward and let them know the service doesn't work for you with their current mix and move on. I don't think you have any good ground to make a Paypal claim though, they can easily prove they are providing to you what you purchased, even if it doesn't fill YOUR whole connection. The only thing this will probably result in is you being banned from their services and if your really lucky, maybe getting your money back, even though you went about it dishonestly.
my 2 cents.
Cheers!
Test with a proper tool like iperf instead of using the absolute abomination that speedtest-cli is.
Never had an issue with Seflow (or maybe not in the past 1.5 years at least)
We need a networking equivalent of PEBKAC.
Dear @mk148a i'm sorry you had issue with us.
I called Matteo to understand who you are and i reconstructed the full story.
When you opened TAC for your issue our support ask you some additional tests from locations that we know had enough bandwidth. Speedtest is not a good way to troubleshoot because we not know the remote locations. Following our instruction is very important to understand where the issue is, but you refuse to do them because "locations is too near to italy" reasons. Our intentions was not an easy way to justify the speed, but to understand where the issue is.
After some debates on WHT you did the tests and we confirmed bad speed confirming that the issue is not on peering side, but something on the switch or server side. Matteo asked you login access to do some additional test (or reboot on rescue mode if you want keep your data private), but you decide to copy on public forum all TAC conversation and Matteo personal details, also you opened a paypal dispute forcing us to suspend the servers (that became unpaid).
Meanwhile we did some additional tests and found that you installed virtualization on the server and did test from virtual machine. We unplugged your server and attached one of our notebook on your switch port and we was able to push at full speed. This mean that the issue is on cable (difficult) or on server (hardware or software or vm configuration i not know).
I'm sorry you had issue with us, but is very important for our engegneers that customers collaborate with us to find the root cause.
At the moment i can confirm that something is not working fine and we focused the issue on server or cable, but due to lack customer collaboration we're unable to troubleshoot better.
In my opinion there is a misconfiguration on virtualized enviroment, but is only my opinion as customer refused more investigations.
Kind Regards
@SeFlow_Ele so you couldn’t determine this by simply running the tests you requested in the rescue OS following the Customer’s termination ?
Never trust Don Matteo.
No, it didn't matter anymore as the contract was closed. We replaced cable (just to be sure), erased data, reimaged server with new o.s. did some download tests and now the server is in use by new happy customer.
So you said you confirmed the issue, but you are not bothered to actually investigate it as the new Customer didn’t complain (yet?)? Is that correct summarisation?
Please read carefully my last reply:
"... We replaced cable (just to be sure) ...... did some download tests ..."
99% was customer side configuration issue, but we not investigate anymore because our focus was give to new customer a full working server and we reached our target.
Have nice day
I did read, and you said you have confirmed the issue exists, then you replaced the cable as a precaution (while advising of none indications as to why would you do that?) while running tests from your laptop, with which you have concluded the fault to be cable or server, you failed to run any more diagnostics, according to your own post, and returned the server to the pool.
With the information presented by yourself I’m inclined to side with your Customer.
Please keep in mind that download tests was made first from notebook while OP still customer. Once he canceled the server we did new ones from the server (not the notebook) with new fresh o.s. on it.
Cable was replaced to avoid any possible issue with new customer.
Thank you for your opinion. I will share it with our egnegneers.
Kind Regards
There is a point being made here. If a stock virtualization system has problems then the provider has to run some tests with the VM if the customer has a reasonable expectation that it should work.
That additional step is one of the things that separate so-so providers from good providers.
I recently had a problem with a provider where performace would crawl below 1.9 MB/s, sometimes even below 1MB/s and I wanted to check if it was because the some of the latest kernels, Debian 10, System Rescue CD 6 were bringing problems which were not visible before.
Just waiting to run these checks when I was ready resulted in a useless tech guy who said I was being insulting because he was totally clueless and the more clued up regular more knowledgeable tech support guy I was dealing with returned to support "totally clueless guy" because "totally clueless guy" was probably the owner or a senior guy.
End result they "banned" me from their services because I was being "insulting". "totally clueless guy" doesn't know that calling a customer "boring" because the customer has to deal with something else is an insult.
Note to service providers - when a customer has a problem and you both need to run tests work out a convenient time when you can both work together