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.
Vultr now in Frankfurt (Germany)
Vultr has finally opened their 12th location, Frankfurt, Germany.
Great News!
Vultr.com has added additional capacity in our Frankfurt location!
Launch your high performance instance today by going to:
As of this post, all of their locations are available and in-stock. It sure is better than having only 4 locations. Check them out at http://Vultr.com if you haven't already.
Thanked by 1perennate
Comments
vultr is so much better than DO..
Is it singlehomed like their UK location?
They don't provide much info on transit etc. Got a tonne of credit at vultr and I don't know how to use it.
Create one for an hour and test? lol
Use this tool and see how many different routes does it take: http://tracert.com/trace_exe.html
I still don't have my new card to sign up for the trial... hopefully it'll be here next week. At first I was pretty enamoured by DO but it's pretty standard service once you get used to the panel and stuff, just slightly higher quality than most at the same price. I really like the number of locations Vultr has available. That name, though...
e: seems I missed it. Oh well.
The routing is still very bad, get 100ms from Germany. Wtf?
how can you get 100ms from Germany? i get under 40ms from Romania
@Vladorz
It routes German connections to the US, see here:http://www.dnstools.ch/visual-traceroute.html
Test IP from Vultr: 108.61.112.204
Ping:
Ping wird ausgeführt für 108.61.112.204 mit 32 Bytes Daten: Antwort von 108.61.112.204: Bytes=32 Zeit=111ms TTL=53 Antwort von 108.61.112.204: Bytes=32 Zeit=111ms TTL=53 Antwort von 108.61.112.204: Bytes=32 Zeit=119ms TTL=53 Antwort von 108.61.112.204: Bytes=32 Zeit=111ms TTL=5340 ms to frankfurt is the norm here. However, if routing goes to london, then paris then vienna then back to germany, you might get 100 ms.
i am aware of the old 'long routes => cheap routes' but i wasn't aware of such bad routes that would actually leave the country just to get back
As a romanian you should remember the upc/rds wars with routing through vienna, I remember one time even mediasat going over alternet in israel and coming back through romtelecom, but that was more than 10 years ago during the collapse of KPNQwest when internet used to cost an arm and a leg.
Used ping.ms to run a traceroute from locations all over the globe and they all have tinet as last hops before Vultr.
I'm well aware but that takes effort..
Have to find my login, wait for instance to deploy and that.. I'm a lazy sod.
Great!!
I just said I used the tool and they use only Tinet.
Oh, see I'm not even bothered to read. ;p
Thanks for that.
10ms from unitymedia germany.
not bad
Yes looks like that, no direct peering on DE-CIX or something only upstream is tibet:
1 gateway.ipv4.fra.filemedia.net (62.113.205.1) 0.198 ms 0.186 ms 0.198 ms 2 193.159.224.81 (193.159.224.81) 0.407 ms 0.403 ms 0.392 ms 3 193.159.167.158 (193.159.167.158) 0.588 ms 0.583 ms 0.610 ms 4 xe-2-0-0.fra61.ip4.tinet.net (89.149.184.65) 12.885 ms xe-8-3-0.fra61.ip4.tinet.net (141.136.108.46) 0.632 ms 0.639 ms 5 choopa-gw.ip4.tinet.net (77.67.64.86) 0.856 ms 1.013 ms 1.165 ms 6 108.61.112.204.choopa.net (108.61.112.204) 0.922 ms 0.917 ms 0.905 msThat is really upstream
not bad for $5/month
I hope they will open one in singapore
maybe the gameservers.com Africa/South America locations are next?
Wishing the same here. There's a huge market for cheap VPS in Singapore.
At $150 & $250 per terabyte of b/w overages in JP and AU, it's hard to be excited for any further exotic locations from Vultr. At least they should have a setting to lock the VPS rather than burning any extra credit you have due to some config mistake or attack/flood at your server.
If you had just a few files on a site, what is to stop you taking a snapshot and just creating a new VM to get a new 100GB of bandwidth each time you come close to the limit?
You'll still be charged for the VM you destroyed, but only for the period you used.
If you're using say 80GB a day with static files, it would be cost effective to do that.
That's one of the complex questions these "cloud" providers have to face. Well you do have a point, so maybe such locking should be per account per location. Also what you described currently seems to be a workable way to get "free" extra bandwidth regardless of whether b/w overage is charged, or gets you locked.
In any case, I consider it unacceptable that someone with an unmetered gigabit server can easily burn hundreds of dollars from my Vultr account just by running wget in a loop overnight (after finding some large file on my website first).
Does not help that Vultr encourages having those hundreds of USD on your account balance in the first place, via the "double your first payment" deal.
bingo.
You both doooooooooooon't get it, for example in the Australia location each VM comes with 100 GB of bandwidth; create a VM, use 99GB, destroy the VM; then repeat with a new one. Can create and destroy 5 VMs and that way use about 500 GB, just paying the hourly rate for each. But if you used 500 GB on one VM without destroying, that'd cost you 400*$0.25 = 100 USD instantly just for the bandwidth overage alone.
That's a loophole shared by many hourly rate cloud providers, I'm not sure if there's curently any countermeasure against this at Vultr.
I am aware of this. I misunderstood @1e10 and thought he was talking about the server itself, not the bandwidth.