Fancy a Latin America VPS server? Larry from Costa Rica based Abdicar emailed me a few days ago with their special promotion for the LowEndBox readers. Use the code lowendbox to get 15% off recurring discount from their unmanaged OpenVZ VPS hosting plans. “crVPS I” would be $6.80/month after the discount. Direct signup link. Do not that you would be paying prorata for the first month, and you can pay by credit card or PayPal.
- 256MB guaranteed/384MB burstable memory
- 40GB storage
- 1Mbps unmetered bandwidth
- 1x vCPU core
- OpenVZ/SolusVM
They have been around since 2003. Do take a look at their Network page and Data Center page, and Abdicar’s explanation here. It seems that they have some local peers for Costa Rica connections, and for all the other out-bound it’s going through their “patent pending” “complex VPN” via Panama to their US end point. So I guess if you are connecting from anywhere else in Central/South America, you’ll still have to route through US. With 1Mbps it means it might have some limited applications though.
Related Posts:
- 5 Reasons Why You Want a Low End Box - May 26, 2021
- Dead Pool January 2012 - February 2, 2012
- exit(0); - January 19, 2012
Any test ip?
Do you offre native ipv6?
Hi Dirk, you can test us with 200.50.230.5 :)
Actually, we are not using the IPv6 yet, since in Costa Rica we have a lot of IPs yet.
So for end user, this is slow performing US VPS.
Hello Mina :)
Well It depends, with the $8 crVPS you get 1mbps, and you can download 300kbps easily. But yes, if you compare it with USA, we can not be better yet.
At 1Mbps that is 122KB/sec best case. That means it would take 8 seconds to transfer a single average web page ( http://www.httparchive.org/trends.php ) if that page was the sole traffic on the VPS.
You’d saturate that bandwidth for extended periods just to install security updates for your OS.
Pass.
International 1Mbps bw is usual everywhere around the world but North America and Europe.
1Mbps x 8s ~ 1MB. Hardly “average Web page”.
I’d like to have a test IP posted also. I wonder if South America connectivity is awesome with the “patent pending” “complex VPN”.
Hi Damian, Costa Rica is not in South America. We’re in Central America :)
Actually we’re the faster ISP in Central America.
Very strange company, they announce their space on the FDC AS in Denver (AS30058) and nowhere a sign of Latin America (except that the space is from LACNIC, which is not a surprise) – also this traceroute:
8 fdc-servers.ip4.tinet.net (77.67.71.6) 175.092 ms 175.734 ms 176.075 ms
9 * * *
10 openvz-atom01.10mbps.com (76.73.100.178) 172.415 ms 172.231 ms 173.625 ms
11 rdns01.abdicar.com (200.50.230.5) 164.032 ms 161.850 ms 168.433 ms
Which is clearly US and not Costa Rica.
Strange, very strange.
Hi William :D
Between the “10 openvz-atom01.10mbps.com (76.73.100.178) 172” and the other point “11 rdns01.abdicar.com (200.50.230.5) 164.032 ms 161.850 ms 168.433 ms” we have all our routing, vpn, proxy cache, and compress technology :)
If you have any questions, just let me know, but I suggest you to take a look at http://abdicar.com/our-network/
Thank you for your time and Pura Vida!
Ahm, Its impossible to from Denver/Chigaco to San Jose, CR (5000km) with a NEGATIVE ping addition unless you found a way to transfer data faster than the speed of light and break the room-time continuum to be faster there than data was in the USA….
I have high doubt’s about the server locations still.
Look here William http://new.imghst.info/1/7d49cr.jpg
World ip confirms it.
So since the traffic routes slowly through the US but physical servers are in Costa Rica, sounds like this service is only good for someone who wants to host something illegal in North America and Europe but legal in Costa Rica. Otherwise, sounds like you’d have better bandwidth hosting your site in the US or Europe.
We have a lot of clients from USA, actually more than the 90% of our customers. And well, yes, we have a lot of clients with gambling websites and similar, but we also have marketing clients and every kind of biz.
Do “note”, I guess.
Thank you very much for the post.
Yes, our VPS aren’t as faster as a USA VPS. Here at Costa Rica, we are not at same level as USA :)
But It’s always a great opportunity to have a VPS outside USA to backup, VPN, Proxy Reverse, or other kind of content services.
If you have any question just let us know.
Happy New Year!
I guess I’m starting to understand.
1. The VPS are located in Costa Rica.
2. Some local ISPs peer directly with abdicar so their customers should have the lower latency possible. Domestic users from others local ISPs use routes peering in a Panama POP.
3. All other routes use the Denver POP. Between Denver and Costa Rica there is a “transport” circuit implemented using the public Internet, VPN, cache, compression, etc. All requests and answers from/to users outside Costa Rica are done via Denver @ 1+Mbps.
@Daniel:
Yes, worldip is not a measurement against physic laws though – Just ping their website, which according to WorldIP is in Costa Rica, from the FDC Servers looking glass: http://lg.fdcservers.net/
A 0,5ms(!) Ping from Denver to Costa Rica is PHYSICALLY IMPOSSIBLE.
PING 200.50.230.5 (200.50.230.5) 56(84) bytes of data.
64 bytes from 200.50.230.5: icmp_seq=1 ttl=63 time=0.457 ms
64 bytes from 200.50.230.5: icmp_seq=2 ttl=63 time=0.359 ms
64 bytes from 200.50.230.5: icmp_seq=3 ttl=63 time=0.353 ms
64 bytes from 200.50.230.5: icmp_seq=4 ttl=63 time=0.378 ms
The Servers are in Denver, and nowhere near Latin/Central America.
I think we can safely say it is located in USA FDC with their own cost rican ips.
Costa*
The other option is to pay $40 per month elsewhere. I’ll let you know how mine goes.
The hosts are clearly mirrored/cached in Denver. Here’s what I think. They probably set up servers in both Costa Rica and in Denver. The Denver servers mirror or “cache” the data from the Costa Rica servers with file-level or byte-level caching. When you hit their IP from somewhere outside of Costa Rica, it routes to the USA and stops there, probably never goes to Costa Rica. Since this would be considered caching, they can continue to operate the IPs as Costa Rica IPs even though the actual response comes from Denver.
I would guess this is similar to what Cachefly does. If I look up the IP of cachefly.cachefly.net, it returns either 205.234.175.175 or 140.99.93.175, traceroute to those IPs and you’ll hit servers in either Los Angeles, Dallas, Chicago, France, Amsterdam, etc… whichever server is closest to you and that server responds. Do a “whois” (or ARIN lookup or GeoIP lookup at http://api.hostip.info/get_html.php?ip=205.234.175.175 ) and it comes up as Chicago, IL even if your traceroute to that IP lands on a server in another city. Since those are “caching” servers, they can continue to operate the IP’s with an origin of Chicago.
I have a server in FDC Denver, and I tested a bunch of pings in their 200.50.224.0/21 range. For the most part, pings averaged 0.5ms to 10ms depending on the IP, 200.50.224.1 was the slowest. But since the pings to every IP was as fast as 0.5ms, I highly doubt any of my pings traveled to Costa Rica and back in that amount of time. 0.5ms is faster than my laptop communicates to my router over Wifi-N at 300 megabits and 50 feet away.
So, if I sign up hoping for a server in Costa Rica that I can ping and test my bandwidth to/from, I will be disappointed. But if I want to host a website with presence in both Costa Rica and Denver, like perhaps a Costa Rica tourism site or DNS server or gambling site, this would be a deal I’d consider.
“probably never goes to Costa Rica.”
It is unlike because dynamic pages require VPS processing as well as FTP, SMTP, ….
As explained by the provider Denver is just a PoP using caching (reverse proxy), VPN, compression, …
I would suspect CPU processing happens in Denver, but the filesystem is in Costa Rica. If the processing didn’t happen in Denver, my pings would never be 0.5ms.
Why not? The provider said they use NAT too. At Denver PoP they may be forwarding the other protocols to Costa Rica via the VPN and replying ICMPs requests in Denver.
true… Maybe you’re right, it’s probably just caching ICMP. Now I’m running HTTP tests on their IP, and it is responding a bit slower. I’m curious enough to think about signing up for a month just to see how their system works. I really have no need for a presence in Costa Rica, though if I did, this looks kinda cool.
I’m curious too and I don’t have audience in Costa Rica enough to justify a VPS there as well. :-(
However considering everything it was posted I guess the provider intended audience for the VPS is the US.
Unfortunately I couldn’t find a looking glass in Costa Rica to roughly assess the latency for the local visitors. From Panama the latency is 100ms using the Global Crossing lg even though the provider has a PoP there. I did conclude the VPS has poor connectivity and transit capacity.
No it isnt.
Cachefly is ANYCAST (which is not only special IP space but also cleary seen in a router bgb log as beeing announced on multiple locations) – Abdicar runs on the FDC AS on normal IP space which makes this impossble.
If they would have this space announced in Costa Rica i would see this in my router as beeing originated on 2 ASNs (or smaller prefixes originated from another AS).
Since i do not see any announcements except AS30058 (which is FDC) this means the space is ONLY routed in the USA.
A friend on DSL in San Jose confirmed this for me by pinging their site/range with FAR over 150ms ping, both on his DSL and on 3G with traceroutes going back to Denver again.
Of course this could just be some sort of NAT – but what sense would it make?
Costa Rican users go to the USA anyway for routing and a L2/Transport link between CR and the US is far more expensive than just hosting everything in the US with no difference.
I should also note that due to the public endpoint in the US and the company used US JURISDICTION applies to them – NOT COSTA RICAN.
They may be using another AS in Costa Rica and the VPS has 2 IPs, one “local” and one “international” …
Since they said we can use their website as testip this is surely not the case.
Hopefully the provider or someone from Costa Rica will post a local traceroute to help us understand how the routing works.
You post comments really goods.
Let’s try to help here:
1.- We don’t care the traffic from Costa Rica. Only the 1-2% of our clients are from Costa Rica. Since the 2003 we’ve been getting clients from USA and Europe. What we like from Costa Rica? We are here, and actually it’s pura vida to life.
2.- We don’t nat for ingoing connections from Costa Rica. It is not important at all, since all the Internet of Costa Rica come from Miami, so if we are fast in USA connections, we’re fine in Costa Rica :)
That’s why we prefer to nat all using usa providers… Our clients are there.
We are going to use all your comments to improve our network page, and explain our technology better in a coming future.
Thank you again!
What’s the point with the regulation?
http://abdicar.com/company-policies/
“8. Compliance with law:
In as much as the service that abdicar provides to the Subscriber is controlled and operated by abdicar from its facilities in the Costa Rica, the Subscriber shall comply with all International, Costa Rican, US, local, state, federal and national laws, statutes, ordinances and regulations that apply to the Subscriber’s use of the service that abdicar provides to the Subscriber, as well as with all laws in all other jurisdictions that apply to the Subscriber’s use of the service that abdicar provides to the Subscriber.”
We dont allow illegal content in our network :)
wow thank you for all your interest guys :)
we have a great affiliate program – 10% recurring, per month. Just in case that you want to sign up!
Happy New Year
their answer on wht was “We cache the ping request and we serve them from the last point of the network” rofl
I ordered a VPS with XEN for my proxies an hour ago, and I just got the welcome email.
So, I decided to ping one of the ISPs in Costa Rica to check the speed with other servers in CR.
And that’s what i got:
[root@cres /]# ping http://www.grupoice.com
PING http://www.grupoice.com (201.200.105.6) 56(84) bytes of data.
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=1 ttl=64 time=0.085 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=2 ttl=64 time=0.081 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=3 ttl=64 time=0.076 ms
From server in serverpronto:
[root@ebark /]# ping http://www.grupoice.com
PING http://www.grupoice.com (201.200.105.6) 56(84) bytes of data.
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=1 ttl=242 time=102 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=2 ttl=242 time=102 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=3 ttl=242 time=102 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=4 ttl=242 time=102 ms
I think it’s considerably faster than the server.
Please post a traceroute, 0,08ms is even in the same datacenter unlikely…
Sure.
[root@ebark ~]# tracert http://www.grupoice.com
traceroute to http://www.grupoice.com (201.200.105.6), 30 hops max, 40 byte packets
1 10.10.95.2 (10.10.95.2) 1.085 ms 1.152 ms 1.196 ms
2 ice.abdicar.com (200.50.224.128) 0.783 ms 1.003 ms 1.164 ms
3 201.200.105.2 (201.200.105.2) 0.782 ms 0.871 ms 1.024 ms
4 http://www.grupoice.com (201.200.105.6) 1.165 ms 1.227 ms 1.294 ms
[root@ebark ~]#
[root@ebark ~]# ping http://www.grupoice.com
PING http://www.grupoice.com (201.200.105.6) 56(84) bytes of data.
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=1 ttl=63 time=1.93 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=2 ttl=63 time=0.784 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=3 ttl=63 time=0.621 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=4 ttl=63 time=0.607 ms
64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=5 ttl=63 time=0.634 ms
^X64 bytes from http://www.grupoice.com (201.200.105.6): icmp_seq=6 ttl=63 time=0.644 ms
[2]+ Stopped ping http://www.grupoice.com
[root@ebark ~]
Tomorrow I will order a OpenVZ, and we can do another test.
Thank you for your order Amy :)
Costa Rica is really small country, and when you have a peer with other ISP, using wired connection (Optical fiber), 0.08 is like be in the same datacenter as william wrote.
Guys, if you need any other tracert or more information, just contact us :)
I guess the local ISPs are using static routes.
I don’t know why this site refused to post comments I just wrote (maybe my traceroutes were too long?), so I’m writing this from another computer. I’m the same “jon” as above, I signed up just to test for a month, and I’m not getting results like I expected. Traceroutes to several costa rican IPs all go through FDC Denver with 100+ ms, including the IP mentioned in Amy’s post. For all practical purposes, this seems like just a box in Denver with a 1mbps Internet connection.
When I scan their TOS and AUP, it surprises me that they also exclusively mention that USA laws also apply.
So if you thought that you could set up a IRC bot / Seedbox / Wikileaks mirror / Gov protest site because it’s based in Costa Rica, you could better change your mind and search for a different host.
Why do you mention IRC bot? :-) What IRC bot have to do with laws?
An IRC bot is a program that sits in an IRC channel around the clock, keeping it open 24 hours a day. It looks just like a normal user on the channel, but is usually idle until it’s called upon to perform a particular function. A bot can contribute greatly to the stability of a channel by protecting it from takeovers and abusers, and providing a central location for storage of the channel op list and giving ops to those users when they join the channel.
It can also perform many other useful functions, such as logging channel events, providing frequently-requested information, hosting trivia games, and so on. IRC bots are particularly important on IRC networks without channel registration services, such as EFnet and IRCnet, and on networks that may prevent your channel being registered due to certain registration requirements, such as Undernet. On these networks, keeping a channel running smoothly without some kind of IRC bot would often be impossible.
It’s not against law of any country to run IRC bot :)
http://en.wikipedia.org/wiki/Warez#Automated_warez_distribution_via_IRC_robots
We’re in 21th century Damian. When did you saw file sharing xdcc bot last time? :) We went through all this at LEB so many times… It’s almost insultive for ordinary IRC bot users to connotate our daily IRC bots usage, open source IRC channel scripting, etc… with some obscure marginal not worth to be mention percentage of potential IRC bots abuse. Not even 0.1% of IRC bots are used for file sharing and even ordinary web pages are more used by abusers than IRC bots, so why not prevent or mention also WEBHOSTING as illegal? Nonsense, a? Well.. same is with IRC bots. IRC bots aren’t illegal by any government but yes.. they can be abused same as everything in hosting industry.
Is your “complex VPN” to the United States your only peering direction? Do you peer with anyone else in Central or South America? If you do, can I have traffic routed that way?
I’m looking for good South America connectivity. Despite not having an ocean in the midst, like with Asia and Europe, US connectivity to Brazil, etc, kinda sucks.
:(
Complete sham. They host within the US. When confronted, they immediately killed the VPS and refunded but would not admit they lied even after I called them on it with proof. Don’t do it, y’all.
Traceroute follows from the server to a well known DNS cluster.
—
[root@cr ~]# traceroute 4.2.2.1
traceroute to 4.2.2.1 (4.2.2.1), 30 hops max, 60 byte packets
1 76.73.2.1 (76.73.2.1) 1.097 ms 1.285 ms *
2 10.10.95.2 (10.10.95.2) 0.788 ms 0.758 ms 0.764 ms
3 be-10-701-cr01.denver.co.ibone.comcast.net (75.149.229.9) 2.239 ms 2.232 ms 2.221 ms
4 4.79.82.53 (4.79.82.53) 0.968 ms 0.965 ms 0.955 ms
5 vlan51.ebr1.Denver1.Level3.net (4.69.147.94) 10.647 ms 0.920 ms 10.596 ms
6 ae-2-2.ebr2.Dallas1.Level3.net (4.69.132.106) 15.326 ms 14.662 ms 14.648 ms
7 ae-72-72.csw2.Dallas1.Level3.net (4.69.151.141) 14.855 ms ae-92-92.csw4.Dallas1.Level3.net (4.69.151.165) 16.604 ms ae-82-82.csw3.Dallas1.Level3.net (4.69.151.153) 24.560 ms
8 ae-4-90.edge3.Dallas1.Level3.net (4.69.145.200) 16.302 ms ae-2-70.edge3.Dallas1.Level3.net (4.69.145.72) 16.563 ms 16.554 ms
9 a.resolvers.level3.net (4.2.2.1) 16.719 ms 16.680 ms 16.676 ms
[root@cr ~]#