how-to block ads
|
gilberca
join:2001-04-15
Darlington, MD
| reply to Bisk
Re: 4pm - 12am traffic shaping problem list
I have been experiencing poor browsing performance from about 3:30PM-11:59PM M-F consistantly since March 11. I placed 4 service calls to tech support for this issue and as of 3/31 when I spoke with tier3 tech support the issue remains unresolved. The tier3 tech I spoke with told me that 91W is maxed out with subscribers. No new subscribers will be added to the satellite. Moreover, no existing subscriber to 91W can even change thier plan without a dish repoint to another satellite. No estimate when I will see any improvement to this situation.
I have noticed that during the slow browsing periods, the HN7000S will have its flow control enabled. I suspect that this function is being manipulated in a effort to reduce congestion with the undesirable effect of severly limiting bandwidth.
I've been a hughesnet subscriber since 2004
Pro Plan
HN7000S Modem
91W
1410 Transponder
April 6th, 10:41PM - 417/119 kbps
April 5th, 06:25PM - 789/187 kbps
April 4th, 05:17PM - 613/145 kbps
April 4th, 05:14PM - 370/176 kbps
April 3rd, 11:26PM - 634/126 kbps
April 3rd, 11:24PM - 564/70 kbps
April 3rd, 08:55PM - 168/119 kbps
April 3rd, 09:00AM - 818/195 kbps
April 3rd, 08:58AM - 850/127 kbps
April 1st, 09:10AM - 844/163 kbps
April 1st, 08:51AM - 447/104 kbps
April 1st, 08:48AM - 772/137 kbps
March 31st, 10:48PM - 251/23 kbps
March 31st, 05:13PM - 257/69 kbps
March 31st, 05:10PM - 337/53 kbps
March 31st, 09:43AM - 839/195 kbps
March 31st, 09:33AM - 820/130 kbps
March 31st, 09:00AM - 819/146 kbps
March 31st, 08:30AM - 870/195 kbps
March 31st, 08:10AM - 877/195 kbps
March 31st, 08:08AM - 879/196 kbps
March 31st, 12:09AM - 745/113 kbps
March 31st, 12:07AM - 796/188 kbps
March 31st, 12:05AM - 595/151 kbps
March 30th, 09:39PM - 67/25 kbps
March 30th, 09:38PM - 76/27 kbps
March 30th, 09:28PM - 108/36 kbps
March 30th, 09:27PM - 94/59 kbps
March 30th, 05:32PM - 199/109 kbps
March 30th, 05:16PM - 221/72 kbps
March 30th, 05:14PM - 220/67 kbps
March 29th, 09:11AM - 785/176 kbps
March 28th, 08:13PM - 229/55 kbps
March 28th, 01:02AM - 863/195 kbps
March 27th, 08:44PM - 244/55 kbps
March 27th, 08:41PM - 243/136 kbps
March 27th, 11:26AM - 851/187 kbps
March 26th, 11:50PM - 511/74 kbps
March 25th, 07:35AM - 875/193 kbps
March 24th, 07:30PM - 193/116 kbps
March 24th, 07:07PM - 153/62 kbps
March 24th, 07:03PM - 119/39 kbps
March 24th, 06:36PM - 189/37 kbps
March 20th, 03:13PM - 844/193 kbps
March 20th, 11:13AM - 876/159 kbps
March 20th, 09:56AM - 870/167 kbps
March 20th, 09:53AM - 870/159 kbps
March 19th, 03:31PM - 302/140 kbps
March 19th, 03:05PM - 432/185 kbps
March 19th, 02:32PM - 361/127 kbps
March 19th, 02:30PM - 394/144 kbps
March 17th, 02:44PM - 820/158 kbps
March 17th, 11:27AM - 875/147 kbps
March 17th, 11:26AM - 875/150 kbps
March 16th, 05:01PM - 302/127 kbps
March 16th, 04:58PM - 358/95 kbps
March 16th, 04:18PM - 427/100 kbps
March 16th, 04:16PM - 435/110 kbps
March 16th, 11:03AM - 680/68 kbps
March 16th, 11:02AM - 806/74 kbps
March 15th, 08:43AM - 879/146 kbps
March 15th, 08:36AM - 822/165 kbps
March 14th, 03:23PM - 855/116 kbps
March 14th, 12:15PM - 768/110 kbps
March 14th, 08:30AM - 874/188 kbps
March 14th, 05:52AM - 868/191 kbps
March 14th, 05:50AM - 782/193 kbps
March 13th, 11:22PM - 105/25 kbps
March 13th, 11:21PM - 109/38 kbps
March 13th, 10:14PM - 87/34 kbps
March 13th, 08:15PM - 67/36 kbps
March 13th, 08:14PM - 68/32 kbps
March 11th, 11:58PM - 273/142 kbps
--
DW7000 G11 1410Mhz XPPro | |
grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY
edit:
April 7th, @08:51AM
| said by gilberca  :I have been experiencing poor browsing performance from about 3:30PM-11:59PM ... The tier3 tech I spoke with told me that 91W is maxed out with subscribers. Thank you Gil. That's positive reinforcement that the problem is simply one of selective over-subscription. Four to midnight approximates the peak internet usage period. Too many people trying to share shared bandwidth. There is no nefarious traffic shaping conspiracy.
//greg//
--
HN7000S/74cm/1w/ProPlus - G16/1250H/Germantown - NAT 66.82.187.152/Gateway 66.82.25.10/DNS 66.82.4.12 and 66.82.4.8 - Firefox - AVG+Firewall v7.5 | |
tori
@1dial.com
| reply to gilberca
said by gilberca :The tier3 tech I spoke with told me that 91W is maxed out with subscribers. No new subscribers will be added to the satellite. Moreover, no existing subscriber to 91W can even change thier plan without a dish repoint to another satellite. I don't think your tech was right, since last week I upgraded (sort of) from my DW3000 that is on 91w (I did not cancel my old service)to a HN7000S and my installer was told that he had to put my new dish on 91w. It took several phone calls to Hughes and at least an hour of trying to lock onto 91w unsuccessfully before they relented to a new satellite for my new dish. | |
MonstaDriva
join:2007-09-02
edit:
May 3rd, @09:58AM
| reply to grohgreg
said by grohgreg :said by gilberca :I have been experiencing poor browsing performance from about 3:30PM-11:59PM ... The tier3 tech I spoke with told me that 91W is maxed out with subscribers. Thank you Gil. That's positive reinforcement that the problem is simply one of selective over-subscription. Four to midnight approximates the peak internet usage period. Too many people trying to share shared bandwidth. There is no nefarious traffic shaping conspiracy. //greg// lol qoute you up above quote me from this thread "04-08-2008 I do know that somthing is going on at the hughesnet level for some users . Perhaps these users have it in common that they download to much and use to much of their allotted megs each day and they get flagged to a list but its def. being innitiated by hughesnet at a very precise schedule . I can't really see why they would punishish entire transponders or sats . The people affected seem to be all over so maybe its a flagged user thing . I do admit i use most of my megs each day so maybe this is hughes way of flagging those users to keep them from getting that 375 megs during primetime . Doesn't any of this seem logical ? They could have thresholds in place in their software that trigger any given user into a list . Its only logical to assume whatever software thats been written to run this conglomeration certainly has a lot of options , check boxes and flags that can direct the system to respond to various phenomena . "
Just admit it, you were way behind the curve on this one . Yor taking little snippets from quotes and they are evn out of context because you don't show whole quotes . i spent days having to convince you there even was an exact 4 to 12 problem . Instead of making progress with this issue everyone has been tangled with you shooting down everyones post . The proof is here for anyone to read just how this all transpired . | |
MMS747
@direcpc.com | There are patents and patent applications for methods and apparatus to do just that: Traffic Shaping. It is a management tool used throughout the industry, not just Hughes. | |
grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY
edit:
May 3rd, @02:59PM
| said by MMS747 :
There are patents and patent applications for methods and apparatus to do just that: Traffic Shaping. It is a management tool used throughout the industry, not just Hughes.
yessir, no one's arguing that. What I'm trying to slow down is the stampede of all the other non-technical customers who grab at every and all straws that might explain what the reason for THEIR particular connection issue.
In this case, the 4-12 thing seems very VERY selective. On this forum, there has been a teeny tiny (but very vocal) number of customers who have documented this particular Hughes phenomenon - and this is one of the LARGEST HughesNet self-help forums in existence. But even if I accept the figure of 500 complainants that somebody threw out here, that's STILL a drop in the half million customer bucket that is HughesNet.
So the KEY here is to come up with, is some common set of behaviors that might serve to trigger this 4-12 throttle for some - and NOT for others. And the survey info requested by the OP does NOT fulfill satisfy that requirement.
//greg//
--
HN7000S/74cm/1w/ProPlus - G16/1250H/Germantown - NAT 66.82.187.152/Gateway 66.82.25.10/DNS 66.82.4.12 and 66.82.4.8 - Firefox - AVG+Firewall v7.5 | |
MMS747
@direcpc.com
| The number of customers affected couldn't be all that "teeny tiny." If you compare the normal speeds within the 4-12 timeframe with speeds outside the 4-12 timeframe, you'll see there isn't that big a difference anymore. In my own case, 4-12 downloads are typically 680+/- compared to outside 4-12 downloads of 700+/-. Once upon a time, the differences were quite marked. Given the huge numbers of 4-12 users, it stands to reason that more than a few users are throttled in order to maintain performance. | |
MichaelRS
join:2006-06-10
Half Moon Bay, CA
 ·WildBlue
edit:
May 4th, @01:25AM
| reply to grohgreg
Just a few fact points:
* The 4-12 throttle affects accoridng to HN Exec Cust Care Tech Whiz 75% of customers.
* HN has per the CEO 350,000 customers and hopes to get to 1 Million.
* While this forum may have a lot of visitors, clearly to be representative of 350,000 you would need more complaints.
* Any number of internet usage statistics will show clearly (null hypothesis etc.) that traffic on the internet or gateway is NOT a step function going on at 4 and off at 12, on the second, repeateadly.
* Such step function, as any gateway/network engineer can confirm, is man-made throttling by HN (see above) at gateway level, gateway = transponder. In fact several VARs have confirmed this practice being in place since years.
* Similar behavior was observed on DW4020 systems in 2006 and on Spot 21 of Anik/F-2 when WildBlue got wild in the west (and crashed) in 2007.
* Customers do NOT appreciate this business (! - not technical) practice, neither by HN, WB, ComCast or anyone.
* Unclear is if this practice is user specific throttling or gateway specific throttling. Any network engineers with a thought her ?
* Corporations pay people to observe and post on blogs on their behalf and undercover
* The following transponders have been reported as affected:
G17/74 - 1130/1250
AMC9/83 - 1430
AMC3/87 - 1330
G28/89 - 1070/1150/1270
G11/91 - 1410
G26/93 - 1216/1433
G16/99 - 970/1110/1210/1250/1370
SM6/113 -1150/1210
SM5/117 - 1070/1090/1130/1250/1270
* This 22 out of 70 transponders confirmed to be knocked out.
--
HN7000S - Satmex 6 - 1210 - 0.98m / 2W dish - HughesNet ProPlus | |
MMS747
@direcpc.com | Interesting. Looks like someone's being doing their homework. Thanks. | |
MichaelRS
join:2006-06-10
Half Moon Bay, CA
·WildBlue
edit:
May 3rd, @11:27PM
| Yeah, welcome. Me engineer too.
Waiting for grohgreg to tell me it is all wrong, and he personally checked all transponders and there is no issue, but my checking satellites is using too much bandwidth, which is why I have the problems and am causing them for others 
BTW if anyone can actually confirm more satellites/transponders that would be VERY helpful.
--
HN7000S - Satmex 6 - 1210 - 0.98m / 2W dish - HughesNet ProPlus | |
MonstaDriva
join:2007-09-02
edit:
May 4th, @12:31AM
| I was always on 89w 1270 till the 4-12 started . A couple of weeks ago they swapped me to 1070 and it was fine for 7 days . Then it started again . A few days later I got swapped to 1150 and it was fine for one full day then it did it too . So thats 3 of them . | |
grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY
edit:
May 4th, @12:49AM
| reply to MichaelRS
said by MichaelRS : gateway = transponder. Incorrect. At least the way that Hughes uses the term "gateway", it does not "= transponder". They use it in the sense of "gateway server", where each gateway has an IP address. There are typically six or more Hughes gateway servers per transponder. So the conclusions you based upon this misconception are not valid. Note PLEASE that I have listed my gateway server as part of my signature line. It's been there for years, and for good reason. You asked for people to identify their transponders - that will get you nowhere. If you want meaningful data, you need them to identify their GATEWAY address. And if members would compose a more informative sig line, you wouldn't even HAVE to ask.
A VAR may have one or more gateway address (servers), but I don't know of ANY that occupy a full transponder by themselves. They share, just like everybody else.
And with 2 thousand or more customers per gateway - and six or more gateways per transponder - I believe you're looking at individual customer throttling. Which brings us right back around to individual usage patterns.
Other than a few cases of overstating the obvious, I take little exception to the rest of your "facts". Although I follow Wildblue also, and don't have any idea of what you've identified as a 2007 "crash".
//greg//
--
HN7000S/74cm/1w/ProPlus - G16/1250H/Germantown - NAT 66.82.187.152/Gateway 66.82.25.10/DNS 66.82.4.12 and 66.82.4.8 - Firefox - AVG+Firewall v7.5 | |
MichaelRS
join:2006-06-10
Half Moon Bay, CA
·WildBlue
| WB crash = inability to add more customers (per spot), resulting in installers quiting business for them until their new bird is up. Exact same deal as with HN as they ran out of capital to add more capacity (ie more satellite). They claimed that their new satellite would fix it, but apparently it did not. What is scarry is that HN is using the same technology as WB does for SpaceWay and WB has issues delivering. Topline also the 9000 for HN does not seem to impress the customers.
--
HN7000S - Satmex 6 - 1210 - 0.98m / 2W dish - HughesNet ProPlus | |
MMS747
@direcpc.com | reply to grohgreg
I'm pretty much convinced it's individual, based on usage patterns. In my case, I believe it was triggered by video streaming, especially YouTube. I also believe that not all video streaming will trigger it, but that there is a threshold. | |
grohgreg
Dunno. Ask The Chief
join:2001-07-05
Dawson Springs, KY
edit:
May 4th, @08:58AM
| reply to MichaelRS
said by MichaelRS :WB crash = inability to add more customers (per spot), resulting in installers quiting business for them until their new bird is up. Exact same deal as with HN as they ran out of capital to add more capacity (ie more satellite). They claimed that their new satellite would fix it, but apparently it did not. What is scarry is that HN is using the same technology as WB does for SpaceWay and WB has issues delivering. Topline also the 9000 for HN does not seem to impress the customers. That is the least supportable collection of sentences you have compiled yet. Such inaccurate representations of how this stuff works have now simply become counterproductive. There is not one technically rationally supportable sentence in that entire paragraph. But none of it has a single thing to do with this topic, I think it's probably best left to die from natural causes.
//greg//
--
HN7000S/74cm/1w/ProPlus - G16/1250H/Germantown - NAT 66.82.187.152/Gateway 66.82.25.10/DNS 66.82.4.12 and 66.82.4.8 - Firefox - AVG+Firewall v7.5 | |
|
