[glidewave]

It's 2014... they restocked RBX0.1 - The planning meetings (ROFL) alone must use up at least 2-3 barrels each time:



New year, new mess.

[crowarth]

great prices!!!






Troll

This is Oles right now I suppose:




i recall the days of them saying your 100Mbps was shared over all your servers. Definitely not something OVH want to go back to.

[fzygo]

great prices!!!






Troll

[TheBritoid]

Yeah, I've resigned myself to the fact that oles is attempting to force turnover on the 10gbps servers.

I don't know anyone who would really choose to keep them if this bandwidth change is applied :/

I still have a 10Gbps server and have been able to pull over 3Gbps from a network outside of OVH before (even though OVH says I have 750Mbps). If OVH adds £1500 to my bill, I'm just going to leave.

[raxxeh]

♫ ♪
You better watch out,
You better not cry,
Better not pout,
I'm telling you why:
Oles does what-ever s'on mind.
♫ ♪

Yeah, I've resigned myself to the fact that oles is attempting to force turnover on the 10gbps servers.

I don't know anyone who would really choose to keep them if this bandwidth change is applied :/

[rickyday]

♫ ♪
You better watch out,
You better not cry,
Better not pout,
I'm telling you why:
Oles does what-ever s'on mind.
♫ ♪

lol

Made me chuckle that did Myatu!

[Myatu]

This had better not affect servers that are over one year old.

♫ ♪
You better watch out,
You better not cry,
Better not pout,
I'm telling you why:
Oles does what-ever s'on mind.
♫ ♪

[raxxeh]

This had better not affect servers that are over one year old.

[elcct]

According to Oles http://forum.ovh.com/showthread.php?...te-(en-option) by default they are Gold up to the limit.what Im not so sure about is why, if you wanted 1Gbps when the limit is 300Mbps on the £84/mo servers, you wouldnt just order 3 more servers for a total throughput of 1.2Gbps, hell for 800 euros you could order 8 more servers for a total throughput of 2.7Gbps

Exactly what I have been thinking

[Kode]

According to Oles http://forum.ovh.com/showthread.php?...te-(en-option) by default they are Gold up to the limit.what Im not so sure about is why, if you wanted 1Gbps when the limit is 300Mbps on the £84/mo servers, you wouldnt just order 3 more servers for a total throughput of 1.2Gbps, hell for 800 euros you could order 8 more servers for a total throughput of 2.7Gbps

[Shimon]

Andy's old thread: http://forum.ovh.co.uk/showthread.ph...gime-explained

[Shimon]

I'm currently getting about 10% - 30% packet loss from China Telecom to my server. I barely use 100GB of bandwidth per month. I would love to change my server to gold grade bandwidth if possible even if I have get a very small data quota.

1 192.168.0.1 (192.168.0.1) 2.549 ms 2.510 ms 2.995 ms
2 192.168.1.1 (192.168.1.1) 5.910 ms 5.913 ms 5.950 ms
3 1.8.173.61.broad.xw.sh.dynamic.163data.com.cn (61.173.8.1) 10.348 ms 19.441 ms 19.461 ms
4 124.74.58.197 (124.74.58.197) 10.217 ms 10.241 ms 10.218 ms
5 124.74.210.205 (124.74.210.205) 14.494 ms 14.446 ms 14.422 ms
6 61.152.86.42 (61.152.86.42) 14.402 ms 10.067 ms 10.070 ms
7 202.97.33.250 (202.97.33.250) 7.541 ms 7.486 ms 7.466 ms
8 202.97.34.102 (202.97.34.102) 7.451 ms 7.418 ms 10.311 ms
9 202.97.52.198 (202.97.52.198) 244.406 ms 244.433 ms 244.441 ms
10 ffm-b2-link.telia.net (213.248.91.45) 235.222 ms 235.256 ms 235.249 ms
11 ffm-bb2-link.telia.net (80.91.252.255) 226.922 ms ffm-bb1-link.telia.net (80.91.246.224) 235.219 ms ffm-bb2-link.telia.net (213.155.132.206) 290.267 ms
12 ffm-b10-link.telia.net (80.91.247.189) 226.765 ms ffm-b10-link.telia.net (80.91.251.120) 232.399 ms ffm-b10-link.telia.net (213.155.134.137) 224.120 ms
13 fra-5-6k.fr.eu (213.186.32.225) 307.415 ms * *
14 rbx-g2-a9.fr.eu (94.23.122.117) 279.344 ms 279.339 ms 279.316 ms
15 * * *
16 * * *
17 bhs-g2-6k.qc.ca (198.27.73.207) 316.958 ms 319.568 ms 312.465 ms
18 bhs-4b-6k.qc.ca (198.27.73.228) 285.823 ms bhs-4a-6k.qc.ca (198.27.73.224) 278.539 ms *

31 packets transmitted, 22 received, 29% packet loss, time 30055ms
rtt min/avg/max/mdev = 268.866/330.317/469.935/50.275 ms

Other Chinese ISPs seem to connect to BHS via the US and have next to no packet loss. So it is probably a case of bad routing.

My ip is 61.173.11.27 and the reverse traceroute from the server is:

traceroute to 61.173.11.27 (61.173.11.27), 30 hops max, 60 byte packets
1 192.99.10.253 (192.99.10.253) 0.315 ms * *
2 bhs-g2-6k.qc.ca (198.27.73.225) 1.393 ms 1.693 ms 1.694 ms
3 * mtl-1-6k.qc.ca (198.27.73.2) 2.026 ms 2.023 ms
4 * * *
5 * * *
6 nyk-bb1-link.telia.net (213.155.130.29) 13.165 ms 9.253 ms nyk-bb2-link.telia.net (213.155.130.31) 9.779 ms
7 sjo-bb1-link.telia.net (213.155.133.239) 82.424 ms 83.123 ms sjo-bb1-link.telia.net (80.91.245.96) 83.097 ms
8 218.30.54.181 (218.30.54.181) 78.860 ms 218.30.54.89 (218.30.54.89) 79.954 ms 218.30.54.181 (218.30.54.181) 77.400 ms
9 * * *
10 202.97.58.189 (202.97.58.189) 252.071 ms 251.689 ms 251.322 ms
11 202.97.35.109 (202.97.35.109) 276.712 ms 277.062 ms 277.013 ms
12 202.97.33.37 (202.97.33.37) 277.566 ms 276.460 ms 276.251 ms
13 202.101.63.121 (202.101.63.121) 272.017 ms 271.985 ms 272.924 ms
14 124.74.211.138 (124.74.211.138) 269.721 ms 268.909 ms 269.194 ms
15 124.74.58.190 (124.74.58.190) 279.921 ms 278.911 ms 281.205 ms

[elcct]

Check in your OVH Manager but most have 100Mbps - 300Mbps Outgoing Guaranteed this is for customers who want more than that.

Hmm... maybe I wasn't clear - as options are divided by Burstable, Premium and Gold, default traffic belongs to which category?

[Neil]

What type of network is by default assigned to server?

Check in your OVH Manager but most have 100Mbps - 300Mbps Outgoing Guaranteed this is for customers who want more than that.

[elcct]

What type of network is by default assigned to server?

[Neil]

Indeed we had them in the past but I would like to point out that was part of the server packages, this is an option on top of the bandwidth you have with the new server ranges. This is aimed at customers that use more than what is guaranteed with the server, for example streaming or image hosting websites.

[Myatu]

Sounds like OVH is heading back in the wrong direction.

+1! OVH did this years ago, and it was a mess then and wouldn't expect it to be any different now. Your "How to" is still somewhere on here, Andy. Haha!

Way to go OVH. So much for listening to customer's input... One ear in, other one out.

[Andy]

Sounds like OVH is heading back in the wrong direction.

[NatdaS]

Hello,

We will be offering 3 types of bandwidth on OVH servers for intensive use of the network. On servers with 1 Gbps connections, you will be able to use 1 Gbps bandwidth instead of 300 Mbps/400 Mbps, and on servers with 10 Gbps connections, you will be able to use 1 Gbps/2 Gbps/3 Gbps instead of 500 Mbps.

The 3 types of bandwidth are:

A) Burstable
1 Gbps at €150/month
2 Gbps at €290/month
3 Gbps at €430/month

B) Premium
1 Gbps at €400/month
2 Gbps at €780/month
3 Gbps at €1160/month

C) Gold
1 Gbps at €800/month
2 Gbps at €1570/month
3 Gbps at €2340/month

There are no changes to the inbound traffic. Regardless of the type of bandwidth, you will be using our entire global network. There are no limitations. You will also all benefit from Anti-DDoS protection based on the entire network of inbound traffic.

On the outbound traffic, to provide the 3 types of bandwidth, we have built 3 sub-networks in each datacentre, within our main network.

This enables us to connect the links between OVH and the internet in a different and more direct/shorter manner, and thus to produce the bandwidth at different costs per Mbps. And in the event of failure of one of the sub-networks, the global network will takeover all of the traffic.

Burstable bandwidth.
Instead of using the entire OVH network with all of its optimisations, the outgoing traffic is redirected to 2 routers of the Burstable sub-network. The transit providers and the peering links are directly connected on these 2 routers. Each datacentre has a Burstable network connected to 2 or 3 PoPs.

SBG = Paris/Frankfurt transit + peering in Europe
RBX = Paris/Amsterdam transit + peering in Europe
GRA = Paris/London transit + peering in Europe
BHS = New York/Chicago/Montreal transit + peering in USA/Canada

Transit providers:
Cogent (we are negotiating with another 3 transit providers)

Premium bandwidth
Instead of using the entire OVH network with all of its optimisations, the outgoing traffic is redirected to 2 routers of the Premium sub-network. The transit providers and the peering links are directly connected on these 2 routers. Each datacentre has a Premium network connected to 2 or 3 PoPs:

SBG = Paris/Frankfurt transit + peering in Europe/USA/Canada
RBX = Paris/Amsterdam transit + peering in Europe/USA/Canada
GRA = Paris/London transit + peering in Europe/USA/Canada
BHS = New York/Chicago/Montreal transit + peering in Europe/USA/Canada

Transit providers:
Cogent, Tata, Level 3, Global Crossing, Seabone, NTT, Telia, Opentransit, T-System.

Gold bandwidth
The entire OVH network is used with all of its optimisations, regardless of the destination:
transit in Europe/USA/Canada (20 PoPs around the world)
peering in Europe/USA/Canada (20 PoPs around the world)

Transit providers: All (20 PoPs around the world)
Cogent, Tata, Level 3, Global Crossing, Seabone, NTT, Telia, Opentransit, T-System.

If anything is not clear, please let us know!

Regards,

Octave