We are in the process of migrating this forum. A new space will be available soon. We are sorry for the inconvenience.

BHS to AT&T Speed Peering Issues


HTMLtag
2016-12-18, 03:16 PM
Update 2016.12.18 These issues are pretty much back again.

HTMLtag
2016-10-04, 12:50 PM
BHS is fast again today...
When the issue happens, download goes down to 0.2mbps -7mbps, while upload speed actually goes up to 130+mbps.
Ironically, now that DL speed is fast again, the upload went down a bit.
The speed test seems to be maxing out my Wifi speed, I can test hardwired later.
Note: This is not due to network saturation on my side, I have gigabit fiber, and other providers would have high data when BHS would be poor.
This is a dramatic speed difference when problems occur and when all is good.
This is not my server I test against, but the proof.bhs to show the BHS peering in general.
http://i.imgur.com/VCA0yRR.jpg

Update: Hardwired can peak at nearly 700mbps today. Whatever issue was going on seems gone again. The upload still seems a bit slower than normal, but it is acceptable for most.
http://i.imgur.com/Ry8TM28.jpg

Phil @ OVH
2016-10-04, 08:59 AM
Hello,

Could you open a support ticket and provide MTR results, we'll gladly look into it.

Regards,
Phil C.

HTMLtag
2016-10-03, 10:19 PM
And speeds are back down and crappy again...

psilos
2016-09-29, 10:53 PM
Experience the same issues....how much longer for ovh to fix the issue? This is my 4thope post and still haven't allowed the post to be published ....but guess what.....I am going to switch to Leaseweb because you don't respect the money I pay every month

bloke
2016-09-14, 02:03 AM
I have also been having terrible network speed lately...
Here is what it looks like to my server...
(I am on TimeWarner 300MB/ps 20MB/ps)

Code:
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address

...
                                0/ 100 =  0%   |
  4   12ms     0/ 100 =  0%     0/ 100 =  0%  agg50.ausutxla01r.texas.rr.com [24.175.43.181]
                                0/ 100 =  0%   |
  5   20ms     0/ 100 =  0%     0/ 100 =  0%  agg22.dllatxl301r.texas.rr.com [24.175.41.46]
                                0/ 100 =  0%   |
  6   22ms     0/ 100 =  0%     0/ 100 =  0%  bu-ether14.dllstx976iw-bcr00.tbone.rr.com [66.109.6.88]
                                0/ 100 =  0%   |
  7   20ms     0/ 100 =  0%     0/ 100 =  0%  0.ae4.pr1.dfw10.tbone.rr.com [107.14.19.97]
                                0/ 100 =  0%   |
  8   17ms     0/ 100 =  0%     0/ 100 =  0%  dls-bb1-link.telia.net [213.248.93.189]
                                0/ 100 =  0%   |
  9   71ms     0/ 100 =  0%     0/ 100 =  0%  nyk-bb1-link.telia.net [213.155.137.50]
                                0/ 100 =  0%   |
 10   67ms     0/ 100 =  0%     0/ 100 =  0%  nyk-b2-link.telia.net [213.155.130.28]
                                0/ 100 =  0%   |
 11  ---     100/ 100 =100%   100/ 100 =100%  be100-154.nwk-5-a9.nj.us [192.99.146.38]
                                0/ 100 =  0%   |
 12  ---     100/ 100 =100%   100/ 100 =100%  be10-1018.bhs-g2-a9.qc.ca [192.99.146.101]
                                0/ 100 =  0%   |
 13  ---     100/ 100 =100%   100/ 100 =100%  vl20.bhs-g2-a75.qc.ca [192.99.146.51]
                                0/ 100 =  0%   |
 14  ---     100/ 100 =100%   100/ 100 =100%  be50-7.bhs-3a-a9.qc.ca [198.27.73.94]
                                0/ 100 =  0%   |
 15   77ms     0/ 100 =  0%     0/ 100 =  0%  **** [***.***.***.***]

psilos
2016-09-12, 06:49 PM
I experience the same issues with att uverse and it appears to be an ovh issue at least per tracerouter

psilos
2016-09-11, 11:13 PM
Quote Originally Posted by HTMLtag
Anyone on AT&T, preferably Gigabit, but any AT&T service, please post your general location and results from http://bhs.proof.ovh.net/ along with the result of pathping bhs.proof.ovh.net. I'm trying to show that there is a peering speed issue that started about a week ago.

This is to gather info from other users, not to open tickets... I'm trying to observe trends and which routes have the issues.
When you're used to seeing 160mbps up/down from a provider, and all of a sudden you're getting less than 2mbps peak 300k average down, even from the proof site... you know there's an issue... especially when over VPN's you realize those speeds again.

Other providers (Comcast, T-Mobile) soar 80+mbps depending on speed... On VPNs, I get > 100mbps. I am on fiber optic gigabit, and get very poor results... The download speeds are better than what they've been lately (80K-3Mbps) at around 12Mbps, but still poor to how I'm used to. The upload is in the 50Mbps range. I have had complaints about it from other people on AT&T. Ping is not an issue.
There is some packet loss seen from pathping. Support escalations have been made to both OVH and AT&T, but I am not getting past, we are reporting it, or not seeing issues. None of the weathermaps, status, etc show backbone issues.

This is from US-East:

If you post the results, please leave off the first few hops as to not reveal your IP in public.

Code:
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
                                4/ 100 =  4%   |
  5    6ms    98/ 100 = 98%    94/ 100 = 94%  12.83.82.161
                                0/ 100 =  0%   |
  6  ---     100/ 100 =100%    96/ 100 = 96%  gar24.attga.ip.att.net [12.122.141.181]
                                0/ 100 =  0%   |
  7  ---     100/ 100 =100%    96/ 100 = 96%  192.205.33.42
                                0/ 100 =  0%   |
  8   35ms     4/ 100 =  4%     0/ 100 =  0%  ash-bb4-link.telia.net [213.155.134.132]
                                2/ 100 =  2%   |
  9   54ms     6/ 100 =  6%     0/ 100 =  0%  nyk-bb2-link.telia.net [62.115.120.215]
                                0/ 100 =  0%   |
 10   42ms     6/ 100 =  6%     0/ 100 =  0%  nyk-b2-link.telia.net [62.115.134.108]
                                5/ 100 =  5%   |
 11  ---     100/ 100 =100%    89/ 100 = 89%  be100-154.nwk-5-a9.nj.us [192.99.146.38]
                                0/ 100 =  0%   |
 12  ---     100/ 100 =100%    89/ 100 = 89%  be10-1018.bhs-g2-a9.qc.ca [192.99.146.101]
                                0/ 100 =  0%   |
 13  ---     100/ 100 =100%    89/ 100 = 89%  vl20.bhs-g2-a75.qc.ca [192.99.146.51]
                                0/ 100 =  0%   |
 14  ---     100/ 100 =100%    89/ 100 = 89%  be50-7.bhs-3b-a9.qc.ca [198.27.73.98]
                                0/ 100 =  0%   |
 15   48ms    11/ 100 = 11%     0/ 100 =  0%  bhs.proof.ovh.net [192.99.19.165]
I do experience the same issues with att uverse....and I know more people that experience issue with att uverse and ovh....and yes issues started two weeks or so ago.....

HTMLtag
2016-09-11, 03:47 PM
Update 2016.12.18 These issues are pretty much back again.

OLD UPDATE: Speeds seem to have returned to normal after a few days.

Anyone on AT&T, preferably Gigabit, but any AT&T service, please post your general location and results from http://bhs.proof.ovh.net/ along with the result of pathping bhs.proof.ovh.net. I'm trying to show that there is a peering speed issue that started about a week ago.

This is to gather info from other users, not to open tickets... I'm trying to observe trends and which routes have the issues.
When you're used to seeing 160mbps up/down from a provider, and all of a sudden you're getting less than 2mbps peak 300k average down, even from the proof site... you know there's an issue... especially when over VPN's you realize those speeds again.

Other providers (Comcast, T-Mobile) soar 80+mbps depending on speed... On VPNs, I get > 100mbps. I am on fiber optic gigabit, and get very poor results... The download speeds are better than what they've been lately (80K-3Mbps) at around 12Mbps, but still poor to how I'm used to. The upload is in the 50Mbps range. I have had complaints about it from other people on AT&T. Ping is not an issue.
There is some packet loss seen from pathping. Support escalations have been made to both OVH and AT&T, but I am not getting past, we are reporting it, or not seeing issues. None of the weathermaps, status, etc show backbone issues.

This is from US-East:

If you post the results, please leave off the first few hops as to not reveal your IP in public.

Code:
            Source to Here   This Node/Link
Hop  RTT    Lost/Sent = Pct  Lost/Sent = Pct  Address
                                4/ 100 =  4%   |
  5    6ms    98/ 100 = 98%    94/ 100 = 94%  12.83.82.161
                                0/ 100 =  0%   |
  6  ---     100/ 100 =100%    96/ 100 = 96%  gar24.attga.ip.att.net [12.122.141.181]
                                0/ 100 =  0%   |
  7  ---     100/ 100 =100%    96/ 100 = 96%  192.205.33.42
                                0/ 100 =  0%   |
  8   35ms     4/ 100 =  4%     0/ 100 =  0%  ash-bb4-link.telia.net [213.155.134.132]
                                2/ 100 =  2%   |
  9   54ms     6/ 100 =  6%     0/ 100 =  0%  nyk-bb2-link.telia.net [62.115.120.215]
                                0/ 100 =  0%   |
 10   42ms     6/ 100 =  6%     0/ 100 =  0%  nyk-b2-link.telia.net [62.115.134.108]
                                5/ 100 =  5%   |
 11  ---     100/ 100 =100%    89/ 100 = 89%  be100-154.nwk-5-a9.nj.us [192.99.146.38]
                                0/ 100 =  0%   |
 12  ---     100/ 100 =100%    89/ 100 = 89%  be10-1018.bhs-g2-a9.qc.ca [192.99.146.101]
                                0/ 100 =  0%   |
 13  ---     100/ 100 =100%    89/ 100 = 89%  vl20.bhs-g2-a75.qc.ca [192.99.146.51]
                                0/ 100 =  0%   |
 14  ---     100/ 100 =100%    89/ 100 = 89%  be50-7.bhs-3b-a9.qc.ca [198.27.73.98]
                                0/ 100 =  0%   |
 15   48ms    11/ 100 = 11%     0/ 100 =  0%  bhs.proof.ovh.net [192.99.19.165]