Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Advertise on LowEndTalk.com
Waveride network problems (DDoS)
New on LowEndTalk? Please read our 'Community Rules' by clicking on it in the right menu!

Waveride network problems (DDoS)

qheadqhead Member
edited July 2013 in Providers

Anybody else having frequent (almost daily) problems with Waveride boxes?

My VPS just went offline few minutes ago (again) and notices from Pingdom are coming daily.

Comments

  • When I had a Waveride box, had the same problems.

  • qheadqhead Member

    I just checked my Pingdom report and yesterday there was 6(!!) ~5min drops.

    Yeah, I think I'm done with Waveride. I mean I don't expect 99.99% uptime but this is too much.

  • awsonawson Member
    edited July 2013

    I got this email from EDIS yesterday, might be related:

    Ladies and Gentlemen,

    during the commissioning of additional 10G fiber optic lines on Wednesday 07/10/2013 between 05:00am and 06:00am CEST, you may encounter short network interruptions of max. 15 minutes. All services located in our Austrian locations will be affected. Additional bandwidth is required in order to better protect our infrastructure against attacks.

    Thank you for your understanding!

    Best Regards

    Yours EDIS Team

  • might be related:

    more likely it's another DDoS attack....the last Austria DDoS outage was Saturday

  • GunterGunter Member
    edited July 2013

    Mine is unreachable again, sad considering I got this yesterday night.

    It's been far longer than 15 minutes and the upgrades are scheduled in 6 hours, not now.

  • patpat Member

    I would expect this on such cheap accounts.

  • One offline waveride vps here also. Waveride support said that host node has problems with ddos.

    I hope that the network improvements tomorrow help with the situation.

  • jmginerjmginer Member, Provider
    edited July 2013

    2013 is the year of DDoS explosion, carries must provide IPS by default and prevent IP spoofing!!

    Voxility DDoS protected BGP starting from 250 EUR/month. Contact us.
    VPS in Spain ☛ 5.99€/month ★ We accept Bitcoins! ★ DMCA ignore ★
  • GunterGunter Member

    I've have good uptime for the past 4-5 hours or so, I imagine the problem has been fixed until 5AM CEST when the new 10Gbit lines get installed.

    @jmginer said:
    2013 is the year of DDoS explosion, carries must provide IPS by default and prevent IP spoofing!!

    It's disappointing, that any script kiddie has access to $4/hr DDOSing services and botnet software.

  • flyfly Member
    edited July 2013

    i mean, you really shouldn't be surprised about DDoS outages given the stance on "abuse" from a very public EDIS rep

  • WilliamWilliam Member, Provider
    edited July 2013

    oh @fly, don't even get me started on you - yes, i know exactly who you are, stop annoying me.

    So for conclusion, yes, mainly issues with reflection attacks - This should be much faster solved now after the upgrades (there were some more upgrades not mentioned in the maintenance at the same time) and prevent most from causing issues at all.

    Theres another 10G link coming in at the end of August and probably another one shortly after along with a few network changes and more migrations of hardware/customers (not VPS) from Graz to Vienna.

    We also got rid off most HE BW and now almost exclusively use Retn which should give better BW quality as well.

    Waveride will also liklely be placed in it's own BW mix to prevent issues for EDIS customers (which is the main reason behind it, the Waveride customers are by far not as important)

  • @William: So, if Waveride customers are by far as not important, why should we buy a VPS from you? Now all new customers know they will have 'by far as no support' when a DDoS is running. And that's my own experience, too.

    It's so confusing. WaveRide is a product from EDIS, right? But what should customers on the WaveRide service do, when a DDoS is attacking the network? Even when it's unmanaged, still.. What should they do? Wave around with a wizard stick?

  • AnthonySmithAnthonySmith Top Provider

    And ^^^this^^^ ladies and gentlemen is the problem with the Ram Race, DireVPS got slammed for being completely honest, they were honest from the get go about support and priority levels to avoid feelings like this.

    I don't blame the customers I blame badly set expectations.

    @DennisdeWit yes that is exactly the case, it is a super cheap oversold economy brand, that is why it is not under the EDIS banner.

    Thanked by 1seriesn
  • I fully agree with @AnthonySmith. I've had too high expectations on WaveRide. You should not use it for full production purposes.

  • @William said:

    Waveride will also liklely be placed in it's own BW mix to prevent issues for EDIS customers (which is the main reason behind it, the Waveride customers are by far not as important)

    Please note that this is not the opinion of the company, but merely represents the personal subjective opinion of the above person.

    one vps is never enough ;-) http://www.edis.at

  • @William said:
    the Waveride customers are by far not as important

    Thanked by 2DalComp asf
  • WilliamWilliam Member, Provider
    edited July 2013

    oh yea rip me apart about some wording in a not even native language, as usual - LET really degraded after the hacks and sale to CC.

    This was merely meant BW wise, not customer wise and that is pretty obvious in this context.

  • @DennisdeWit said:
    I fully agree with AnthonySmith. I've had too high expectations on WaveRide. You should not use it for full production purposes.

    Well, it depends on how you manage it, of course ;-)

    But for business purposes, I would definitely go with normal EDIS packages and not for Waveride. For a personal dev machine or something, Waveride is awesome. I think people shouldn't see these high RAM offers as something intended for high-availability, high-performance services but rather as something "on-the-side".

    I recommend Prometeus, the best provider ever!

  • WilliamWilliam Member, Provider

    That sums it up pretty good, thanks Liam :)

  • SpiritSpirit Disabled

    @William said:
    oh yea rip me apart about some wording in a not even native language, as usual - LET really degraded after the hacks and sale to CC.

    You said something stupid, obviously without thinking, got undesired responses and now that's LET fault? Low kick @William

  • WilliamWilliam Member, Provider
    edited July 2013

    Just read the last few pages and compare them to a few months back and you see yourself how it degraded, you can't deny that.

    And no, it was not stupid (or i would have edited, which i actually could still do), it's simply not an entire sentence and misuse of a word to describe it the way it should be (as @Liam did) - The meaning of the network and not the service itself should be pretty clear for anyone in a NETWORK related thread but it seems some just don't get it.

    As the OPs problem/question is solved i request this thread to be closed @Liam

  • How does Edis manage to attract all the DDoS? I remember that when i was with them some time ago my node's network was down/unusable every other week while the hosts i'm with right now (ProviderService and Prometeus) never had any problems of that sort. Is this due to different technologies or client bases?

  • SpiritSpirit Disabled

    @William I see only degradation in your statements in this thread. This simply isn't the way to defend own "mistake".
    You said something stupid. Got back some heat. Explained to clarify. Done. But don't bring into thread something completely irrelevant. Whatever happened in this thread it had nothing to do with hack, CC or even LET itself... and only one you can put blame on is yourself.
    I will repeat it again. The way how you involved LET into your own lets call it "mistake" is low. It's irrelevant. And it had nothing to do with this thread. Take responsability for things you say or do and don't blame others.

  • rm_rm_ Member
    edited July 2013

    to prevent issues for EDIS customers (which is the main reason behind it, the Waveride customers are by far not as important)

    I wonder WTF is everyone's problem with this statement. Or do you not like being told the truth, honestly and bluntly. If a company has two customers, where the first one pays 6 times as much, and on top of that receives 4 times less product than the second one, is anyone really surprised that the 1st customer will be more important to the company than the 2nd one?

    5 EUR/mo Waveride minecraft kid gets DDOSed, and someone's company business website on 30 EUR/mo EDIS KVM server goes down. What needs to be done is to separate these so than #1 doesn't affect #2, and that's exactly what they now plan on doing.

  • AnthonySmithAnthonySmith Top Provider
    edited July 2013

    @rm_ I agree completely the fault lies entirely with self expectations, I was simply pointing out that it would be impossible to get the wrong expectations from someone like direvps while waveride covers the website in enterprise grade statements so its not a big surprise that expectations are that they get the same in terms of service priority.

    I am not having a go at William at all, I think it sucks that people would open a thread about this sort of thing but when it comes to setting expectations edis need to take some of the blame here too, this is just a chance for improvement.

  • @AnthonySmith said:

    I am not having a go at William at all, I think it sucks that people would open a thread about this sort of thing but when it comes to setting expectations edis need to take some of the blame here too, this is just a chance for improvement.

    Absolutely! We strive to permanently improve our services based on feedback and reviews.
    There is also room to improve the way of how we communicate to prevent future misunderstandings. Honestly, I was shocked about Williams posting, seeing it taken out of the context and having it quoted on our twitter feed. Now I understand what he meant, but I agree that the wording was a bit "unlucky". Sorry for that guys!

    Cheers,
    Gerhard @EDIS

    one vps is never enough ;-) http://www.edis.at

  • AnthonySmithAnthonySmith Top Provider

    heh no one will pay much attention to this anyway Gerhard, the wording was not that bad just the context was a little out of place, Williams written English is probably better than mine and I am from England :)

    What people need to remember is that Edis have a massive customer base, the fact that this is the first complaint I have seen here regarding Edis should talk volumes about the overall quality of service, even the best of us have a "blip" even softlayer.

  • @awson said:
    I got this email from EDIS yesterday, might be related:

    Yours EDIS Team

    They didn't have Fibre Optic lines? Or is that just something to fill the space and just say they got a 10Gbit line?

    RaptorNode.com - DDoS Protected|VPS|Cloud|Dedicated Servers|Collocation - Los Angeles, CA
    SamuraiBit Security
  • EvoEvo Member

    the Waveride customers are by far not as important

    I understand that Waveride comes without any management and support, and I completely understand that Waveride is the "budget", low cost brand (the service is actually great for the price), but I find this comment a little bit offensive.

    I could restructure it to "Paying my invoice for Waveride by far is not as important".

    Thanked by 1ErawanArifNugroho

    http://NeatLogos.com - Easy, Affordable, Fast - Logo and Web Design

  • MaouniqueMaounique Member
    edited July 2013

    We have our share of attacks too, network is not separated here and has the potential to affect big customers, therefore we have to watch it all the time and have spare, non-commit capacities for these cases. When amplification attacks with more than 20 gb hit, those take over some of that until we sort it out (usually a few minutes).

    However, individual nodes are dead in the water in all this time, since 1 gbps NIC will not take 20+ gbps in no way shape or form.

    Uncle was very concerned about this and put some tools in place for me to watch from the very beginning, we now hove 24/7 coverage, more or less, when a human is available to look over it.

    Thanked by 1ErawanArifNugroho

    Extremist conservative user, I wish to preserve human and civil rights, free speech, freedom of the press and worship, rule of law, democracy, peace and prosperity, social mobility, etc. Now you can draw your guns.

  • Keep in mind it could also be a local DDoS.

    VPSCorner is a Offshore, Secure, Cheap VPS Provider. Posting as Luke, not all opinions are that of VPSCorner.

  • WilliamWilliam Member, Provider
    edited July 2013

    When amplification attacks with more than 20 gb hit, those take over some of that until we sort it out (usually a few minutes).

    Oh yes, DNS amplification is a very annoying problem for todays hosting providers - TCP attacks nearly ceased to exist or are easily done by local null (largest i've seen was around 6G TCP) but DNS reflection (and now more common, Teamspeak and GoldSrc servers) are very heavy (largest to date i've seen was large enough to cause massive packetloss on HEs FRA-PRA-VIE 100G/40G transport circuit and for most HE customers in Austria and CZ, possibly also for further HE locations in Europe)

    Williams written English is probably better than mine and I am from England :)

    Haha, depends on subject, i have my issues ironically more with basic things (', an/a, sentence building, it just makes not much sense when you use mainly German all day) but on technical stuff it's pretty good i guess (German also uses many English words unlike for example French). Whenever possible i prefer to use English, but i am simply working in the wrong country to use it as main language ;o

    They didn't have Fibre Optic lines? Or is that just something to fill the space and just say they got a 10Gbit line?

    Let me provide some more infos...

    We had 10G tech in use before this as well.

    The core is in Vienna and we utilize a longhaul DWDM (L2, rented) link to a main site in Graz where 2 locations in Graz (soon 3) are connected and maintain 2 locations in Vienna over CWDM (Vienna, own equipment). This was merely an in-place upgrade of an existing upstream (Retn.net, our main one) that also consisted of some background work (routing wise) to improve the network stability and resilience against DDoS attacks and traffic spikes as well as some small peering changes/additions on VIX. While doing this most of the HE routing has been removed as well, first for IPv4 and soon for IPv6.

    So far it has been proven successful a few times already with local nullroutes for DDoS targets, at no serious disturbance of the network (1-3% packetloss while the attack was spiking heavily, no downtime).

    We do not run darkfiber currently as it makes no economical sense, a C/DWDM wave can be easily used for 10G (40G common as well, Infinera equipment mainly) and even a lot of these are by far cheaper than a darkfiber link (GRZ-VIE is repeated x4, which means power/ac/space costs and there is not much demand here for transport, so we'd have to mainly use it alone or resell it to a Slovenian ISP for backhaul usage).

    Our datacenter in Vienna runs a redundant darkfiber ring between the DC and InterXion (which is the main carrier hotel in Austria) where we use dark CWDM waves to connect to our private patchpanel in InterXion and from there to our transport and upstream carriers (partly) over some equipment colo'd there.

    There's a posting here on LET somewhere from the move to Vienna where i explained some parts a bit more detailed if you want to search, or just ask. (Theres also wiki.edis.at that needs some Content, but i am not very good (ha, the irony again) at translating that to German so probably not writing it down for it.)

Sign In or Register to comment.