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
EDIS Iceland has been getting worse and worse network wize, I finally gave up on it and cancelled all my VPS's there.
I've always been really happy with EDIS, and I did get the email about staff being on sick leave, but I've sent a few emails to their support@edis mailbox and have had no replies, not even in my spam folder. One was a query, no answer. The other was an issue with my VPS, which was fixed shortly but still didn't get a reply as to what the issue was or anything. This was a few weeks after the sickness email.
Oh, and my VPS which had 130+ days uptime has suddenly gone to 3-4 days uptime
Iceland location seems to have gotten better for me recently, a few weeks ago there was terrible packet loss and it was unusable as a simple HTTP proxy but now it's okay.
I don't use IPv6 though.
But I run other applications in the other virtual machines I have with them and recently the host node is always overloaded, along with network problems, and I can't really run it anymore. A month or so ago for example Russia, Germany, and Austria location were great, but now the CPU usage is always covered by other VM, sometimes load goes to 50, when I contact support they say 50 is temporary no problem basically.
I really liked their Russia location, no one I can switch to who seems as legit as EDIS.
I have been having some issues as in uptime or just general less than usual performance from EDIS. Really not like them at all.
Hi guys, this is Gerhard form EDIS!
I'm sorry to say that, but we were not doing well in terms of staff.
We had lots of employees being sick in Jan/Feb and now there seems to be a second wave.
Lots of new hires onboard already and we're slowly able to keep-up with the pace of tickets rolling in again, but it is hard. Yes we removed the cheapest products from the website and we stopped doing dedicated servers for now. This has nothing to do with us wanting to earn more: too many people purchased EDIS products for cheap DDOS attacks and this caused additional headache on top of about 10 people missing in our support-team in Jan/Feb and causing around 15K Euro just for the overtime we had to pay for in Feb ;-)
In case we missed one or another ticket, please submit it again.
We're willing to answer every single ticket.
Apologies in advance for mistakes that happen these days. New staff is in place, but they first need to learn everything from scratch (apocalyptic days are over and they are getting better every day!)
In case we delivered a product too late, we made amendments to the subscription in every client's favor. If you want a refund, please place a cancellation and submit your inquiry to [email protected]
Yes, we're facing hard times, but we're alive - more than ever before and make things better now than they've ever been before.
If you are looking for a job, willing to work hard and you have nothing against nightshifts, then you are our candidate. Just send your CV to [email protected] (freelancers from non-EU timezones welcome).
What's going to happen next:
News servers waiting in all locations can finally be rolled out to improve stability
(btw: Iceland: Advania / THOR are having issues in the last weeks - this affects our services, but is related to the datacenter)
Abusers are going to be kicked out
Lots of other things happening as well here, but this is not the right time and place to share all of this.
Sorry to so many of you whom we could not entirely satisfy.
Thanks to all of you guys who believe in us and stand behind us!
This is the beginning of EDIS 2.0!
Kind regards,
Gerhard @EDIS ;-)
... just let me repeat this once again:
looking for a job? willing to work hard? freenlancer?
Just send your CV to [email protected]
>
Hope everything works out, apart from those one or two tickets EDIS have always been great and reliable.
@hostnoob Muchas Gracias!
this is very good, wanna try it, i'm from non eu after all......
@ExPl0ReR gute besserung to all your staff of course, but was there some sort of epidemic in the office?
Thank you! good questions how it all started. First 2 important staff-members went sick (some kind of burnout syndrome) and the remaining people started getting sick from stress and/or flu/fever.
@ExPl0ReR - thanks for the update.
Chain reaction or Murphy's law, depends
absolutely: first HP delivered servers too late, then Christmas, New Year, people got sick, left, remaining staff panicked and now we're trying to get back on track ... (and we're almost back to the point ;-)
I hope it's the staffing issues that explain the 50% non-response on support tickets (out of about six emails, I think, in the last two - three weeks). It's very unfortunate when flu season hits companies. Major kudos to them for sending out an email to all customers informing us about it.
However, I no longer use EDIS for anything mission critical. Simply cannot trust them enough, if their support can be knocked out this easily.
Still consider them an excellent host for their choice of locations, though. Hopefully things improve soon.
@ExPl0ReR if so many of your workers burn out maybe you should revisit your company policy
Also am i crazy or did a pretty long post disappear?
Edit: I actually mixed the two Edis threads up
@ExPl0ReR Just want to point out that I very much appreciate your transparency. This is a good example of how to handle big internal problems while staying transparent
Most other providers would have gone bankrupt in this situation.
As an anecdote in the other thread, referenced as statement from my side: http://lowendtalk.com/discussion/comment/493917/#Comment_493917]
Is this also for 5 digits like what William is earning? lol
But looks like your employees get burn a lot. A very stressful company might not be worth a lot of money.
5 digits in what... per year yea.
Fixed
It depends. What stresses some people, motivates others. In the end it is all about finding the right mix of people and personalities and give to each the burden they can carry and even be happy about.
For example, I do not get upset when people call me names abuse me in any way, I will not be willing to help them or give them the benefit of the doubt, but I will not feel bad in any way. The same thing makes other people snap and behave unprofessionally, I am merely amused.
The same happens here however, here I try to prove a point that abuse does not work that bans will not contribute to the discussion and the exchange of ideas, that biased admins and users are usually not right and resorting to bans and abuse is not gonna help them in any way.
Now they sent me an email claiming that I was sending broadcast storm or some stupid thing like that, even though it is just gameserver that runs over UDP...
Where is your VPS located? Sweden?
Austria, they have problem with their networking setup and they think it's problem with my VM. Probably I send packet to 255.255.255.255 and their setup wrong and it gets routed to monitor or something.
And it is vServer, not KVM, but the support person thinks it is KVM and spoofing Mac address or some BS even though it is vServer and doing anything like that is not even possible since there's not even any actual guest kernel.
I attached a screenshot to this mail that clearly shows you did broadcast storm the network, how is not important (and, by extension - Your issue, not ours.).
This is also not any problem with the network setup - Any large VPS provider uses a shared netmask (Amazon, Rackspace, DO, Vultr, KT Cloud etc etc), and on your product it does not matter at all as IP stealing is impossible virtualization wise (neither on others by usage of ebtables). It also saves a lot of IPs and there are dedicated providers doing the same (notable OVH).
The attacked KVM was also my PRIVATE monitoring KVM, nothing to do with EDIS (not even EDIS IP space) - 90% of the traffic was directed there (i'm still investigating how this is possible at all, but UDP can do funky stuff in the same L2 it seems), 10% going to others (which, by some multiplication, was still enough to cause issues). After shutting down your VM a lot of network load just disappeared.
I also NEVER stated it was spoofing MACs.
I am also not saying it was intentional (neither in the mail i sent, i don't look into customer VPS (privacy) so no idea what process causes it) - Gameservers are often misused for DDoS and alike (and if it is UDP there is next to nothing to prevent it), regardless of this it caused issues and was suspended and you were informed within minutes (it was not even suspended yet on the node) of the suspension with a clear reason why and a screenshot of the traffic from another host + a description what it did.
Sorry, but i see this as entirely correctly done, when the problem is analyzed it will be reactivated and unless it does it again (after some timeframe to investigate it yourself and stop it) it will be again suspended and further measures might be taken.
It sends like 100 UDP packets per second, your network cannot handle that? And why is your network setup to route 255.255.255.255 to all the VM's?
I can't believe it I am sending packet to 255.255.255.255 and his network system is messing up and routing it to his KVM for some reason instead of just looping it back to my server, and now he thinks I'm attacking his KVM!
Yes, to 5000+ IPs at the same time. Again - Broadcast.
That is a fact? I had a normal DDoS yesterday and it's quite common for others to attack my infra - In this case a VRS inside this network was used for it, either intentionally or simply as UDP reflector because it was scanned by the attacker (like any open, non-limited, DNS server).
This entire case is likely not more than anyone externally abusing whatever gameserver you run for DDoS, formerly for others, now by chance for me, in the future probably for others and by doing stupid things (like, redirecting UDP to 255.255.255.255, i won't go into detail now but please just read some L2 networking standards) caused problems - It likely "belongs" to some booter that feeds it with spoofed traffic from Ecatel or alike.
Again, i never stated it was you - The service is just abused, simple as that.
Actually...
The 6112 is from my program, the default is to send to 255.255.255.255, if you have so many IP that any packet sent to broadcast address causes problem for your network then you should just disable any broadcast packets?
I moved the program already since your server cannot handle it but VM is still down.
Actually what? You can't send raw packet with the permissions inside the vServer container.
Edit: and if problem is with DDoS yesterday, what does that have to do with my VM sending broadcast packets? It is only routing packets from another program running on same VM on specific port.