HughesNet Tier 4 support refused to help with unusual high latency problem

[lighthope1]

Here is a record of this current problem.

Record begins.

Case ##124526040

Had issue with low speeds in evening.  400K speeds.  Was switched to different satellite.  Speed problem solved, however, now am experiencing very high and wildly swinging ping times.  Between 800 ms and 44,000ms!!!!!!!!!!!!  (That 44,000 was only recorded once.  Usual upper limit is 5,000 ms)

At times, websites like Facebook won't load and any kind of streaming is impossibly stuttered.

Called Tier 1 support.

First call.  Weather was blamed.

Second call same day.  Weather was blamed.

Third call next day (Next day (10-18-19).  Disconnected

Called back.  Represetative clearly did not speak English natively and had some communication problems.  Representative said case has been escalated to Engineering department and that I would receive a call tomorrow in the AM. Reference #124564944.  Very oddly the representative asked me what I had for dinner.

Engineering called back 10-21 in the afternoon. Heavy storm overhead. Have to call back.

Called back 10-22 1:37 PM. Told Tier 4 Representative about ping problems.  I told him that ping was wildly swinging between 800ms and a record 44,000 ms.  I also said that I never had this problem on the old satellite.

Representative said he would not do anything about it, said 2000+ latency is just the way that it is. (Not a direct quote)

I acknowledged that he refused to do anything about it.  Representative thanked me for calling HugesNet Support.

End of record.

So bascially HughesNet Tier 4 told me to go suck an egg.  They won't do anything about it.

Great customer service.

Wish there was another company in my area.  Will keep looking and hoping.

[lighthope1]

Here is my latency from tonight:

Range was 979ms to 2848ms

https://testmy.net/latency?tID=uidalsi9s9

[MarkJFine]

I wouldn't go by that latency chart.

The only way to test your latency is to do your own pings/traceroutes to several major hosts.

[lighthope1]

---

@MarkJFine wrote:

I wouldn't go by that latency chart.

The only way to test your latency is to do your own pings/traceroutes to several major hosts.

---

I've done that.  But testmy.net had a chart I could post.

[GabeU]

The result the TMN test gave me, using the Dallas server, was 2471ms, which is WAY off.  I don't know exactly how it tests, but I also tried a few other servers, with each one giving me results that I know are not correct.

[lighthope1]

I've tried a couple more sites.  Some just time out.

I also tried the command prompt ping test. That comes out at 600ms fairly consistently.

So now we have to ask the question: Why does a ping test come with low ping, but streaming data have such problems.

The only thing I can think of is what someone else brought up: There is some sort of data clog somewhere.

Sometimes doing a download test from testmy.net (the site Hughesnet tells us to use!), shows it hanging for a long them, then sending the data in a whoosh. Upload tests have upload speeds all over the place during the timeline.

That could be the problem.  Not latency, but a data clog.

[lighthope1]

Oh, to add to the record, someone from HughesNet called me. Unrelated to this issue, it was for something else. But I did bring this problem up (as well as the extremely unhelpful Tier 4 support person), and she is investigating the issue.  She said she'd call back.

I did mention the possibility of a data clog being the problem.

I will report when (if) she does.

[lighthope1]

I don't know if this will be relevant, but this page

https://gallifreybase.com/gb/

Took 21 seconds to load.

This page

https://my.hughesnet.com

Took 23 seconds to load.

This page

https://www.facebook.com/

Timed out and failed to load.

[maratsade]

@GabeU was having issues accessing FB a while back.   It may all be related to CenturyLink's infrastructure problems. 

Gallifrey is so far away, this may be the reason it takes so long for the signal to reach it -- except for Time Lords, of course. 

lighthope1 wrote:

I don't know if this will be relevant, but this page

https://gallifreybase.com/gb/

Took 21 seconds to load.

This page

https://my.hughesnet.com

Took 23 seconds to load.

This page

https://www.facebook.com/

Timed out and failed to load.

---

[lighthope1]

---

@maratsade wrote:

Gallifrey is so far away, this may be the reason it takes so long for the signal to reach it -- except for Time Lords, of course. 

Ah, but they have time machines, so they can go back in time and send the message so it gets here the next instant.  🙂

[maratsade]

EXACTLY!!

---

@lighthope1 wrote:

maratsade wrote:

Gallifrey is so far away, this may be the reason it takes so long for the signal to reach it -- except for Time Lords, of course. 

Ah, but they have time machines, so they can go back in time and send the message so it gets here the next instant.  🙂

---

[GabeU]

@lighthope1 

If you use Chrome, I suggest using the extension "Page load time", then taking a snapshot of the pages, with both the load time and your desktop clock showing, and indicating what website it was.  This way HughesNet not only has proof of the pages loading slowly, but also the times you tried the pages, which might be of value to them while attempting to determine the cause of the issue.  If you use another browser, there is likely a similar extension for it.  

This was for the Tech Support section page.  The blue highlight in the breakdown box isn't me trying to direct attention to a certain part, but rather something "Page load time" does automatically. The number in darker blue to the upper right of the breakdown box is the total page load time in seconds.  In this case, 4.69s.  

[maratsade]

Your latency (via ping) is right for satellite. 

I think what you're experiencing is a data clog, like you say.  The data clogs are happening somewhere else, not at HN.  They are a national issue having to do with CenturyLink  (@MarkJFine knows a lot about this topic). 

It is likely that until the government does something to address this issue with CenturyLink, we will just have to deal. 

I'm sure companies that receive service from CL are working to address the matter (which can be a lengthy process),  but they don't owe the customers any explanation about this. 

lighthope1 wrote:

I've tried a couple more sites.  Some just time out.

I also tried the command prompt ping test. That comes out at 600ms fairly consistently.

So now we have to ask the question: Why does a ping test come with low ping, but streaming data have such problems.

The only thing I can think of is what someone else brought up: There is some sort of data clog somewhere.

Sometimes doing a download test from testmy.net (the site Hughesnet tells us to use!), shows it hanging for a long them, then sending the data in a whoosh. Upload tests have upload speeds all over the place during the timeline.

That could be the problem.  Not latency, but a data clog.

---

[lighthope1]

---

@maratsade wrote:

Your latency (via ping) is right for satellite. 

I think what you're experiencing is a data clog, like you say.  The data clogs are happening somewhere else, not at HN.  They are a national issue having to do with CenturyLink  (@MarkJFine knows a lot about this topic). 

It is likely that until the government does something to address this issue with CenturyLink, we will just have to deal. 

I'm sure companies that receive service from CL are working to address the matter (which can be a lengthy process),  but they don't owe the customers any explanation about this. 

---

I agree that it is a data clog.

I am not sure that government should be the go-to solution for this.

I disagree with your position that "they don't owe the customers any explanation about this."  That is not good business practice, especially when HughesNet is a vitual monopoly in certain areas.

Coming out and saying: "There is a problem.  This is what the problem is. This is what we are doing about the problem.  Here is where it is out of our hands and we are doing such and such about it" would go a very long way to keep customers from becoming upset.

Ignoring the problem, not acknolwledging the problem, and saying "that's just the way it is" puts very little credence into your belief that "I'm sure companies that receive service from CL are working to address the matter".  Why would HughesNet be working with CenturyLink when HughesNet doesn't believe there is a problem?  What would be the working on when there is no problem?

[maratsade]

I'm not sure what other choices there are for internet backbone, so HN may not have a choice.  Can't say one way or another...

I'd love to see the government and CL corporate customers do something, like storm CL HQ with pitchforks.  A government agency that shall remain nameless has already started yelling at CL.  We'll see where that goes. 

I do hear you about telling customers, but see, corporations don't normally reveal this kind of stuff, and they're entitled not to (whether we like it or not).  

lighthope1 wrote:

---

---

I agree that it is a data clog.

I am not sure that government should be the go-to solution for this.

I disagree with your position that "they don't owe the customers any explanation about this."  That is not good business practice, especially when HughesNet is a vitual monopoly in certain areas.

Coming out and saying: "There is a problem.  This is what the problem is. This is what we are doing about the problem.  Here is where it is out of our hands and we are doing such and such about it" would go a very long way to keep customers from becoming upset.

Ignoring the problem, not acknolwledging the problem, and saying "that's just the way it is" puts very little credence into your belief that "I'm sure companies that receive service from CL are working to address the matter".  Why would HughesNet be working with CenturyLink when HughesNet doesn't believe there is a problem?  What would be the working on when there is no problem?

---

[MarkJFine]

I administer two sites, one of which is a web space on a server in Pennsylvania, hosted on Pair. On that server I have a 'bot' that creates an image from he last 10 posts on another site I administer in Germany, but based in Northern Ireland. The 'bot' then takes the image and creates an automated tweet that gets posted on Twitter. This is a process that is done three times a day.

There are a total of only two hops between the server in Pennsylvania and Twitter, and both hops are to CenturyLink servers on their backbone. Invariably, the login to Twitter times out past 60 seconds at least once per day. Note that HughesNet isn't involved anywhere in this process, and this is a CL-only networkng problem.

I have personally contacted CenturyLink several times over the past year and *they* deny they have problem. What makes you think HughesNet would have better luck convincing them? Or, Pair, my server hosting company in Pennsylvania, who is also affected?

Plus, as @maratsade has already said, HN may be locked into a long term grandfathered contract with Level3 (whom CL bought) and can't go anywhere else or there might not be anyone else.

If you look back a couple of years, (prior to the Level3 buyout) CenturyLink's *backbone crashed* for a week leaving the entire internet in a mess on the West Coast. It was heavily covered on the news.

CenturyLink is a disaster, and they are making everyone using them look bad, but I doubt Pei's government organization will get involved.

[maratsade]

"one of which is a web space on a server in Pennsylvania, hosted on Pair. On that server I have a 'bot' that creates an image from he last 10 posts on another site I administer in Germany, but based in Northern Ireland."

Intriguing!

"CenturyLink is a disaster, and they are making everyone using them look bad, but I doubt Pei's government organization will get involved."

Pei and his org have already gotten involved with this. The trigger was that people couldn't phone 911.

Can't say if this will end up going anywhere at all, but they have gotten involved. 

[MarkJFine]

---

@maratsade wrote:

Intriguing!

---

Yeah, it's a simple little script that polls the site's rss feed, takes the most recent titles and posts this image/caption to Twitter:

Today on Daily Cannon - 25th Oct 17.01 BST. All this and more at DailyCannon.com/latest/
#ArsenalDaily

[lighthope1]

When I was on a different satellite, I am not sure I had this problem. The other satellite was where I was getting 400K download speeds.  That came with some pretty high latency as welll, but it is impossible to tell if one problem was affecting the other.

My speeds are good now, but the latency is problematic.

I think we're all on board that there is a data clog.  But since HughesNet has not acknowledged the problem, we can not say for sure that it is the problem or if they are even working on it.

As long as the problem is not acknowledged, it will be classifed as a Problem Ignored.

[maratsade]

You may very well not have had this problem. As technology progresses issues are solved and/or new problems are created, and there are so many variables involved that's it's maddening to try to unravel the knot, especially since we don't have all the info.

Also, because we have little control over the issue, we can only hope that those who do will take action, and soon.

lighthope1 wrote:

When I was on a different satellite, I am not sure I had this problem. The other satellite was where I was getting 400K download speeds.  That came with some pretty high latency as welll, but it is impossible to tell if one problem was affecting the other.

My speeds are good now, but the latency is problematic.

I think we're all on board that there is a data clog.  But since HughesNet has not acknowledged the problem, we can not say for sure that it is the problem or if they are even working on it.

As long as the problem is not acknowledged, it will be classifed as a Problem Ignored.

---

[lighthope1]

So I installed the page-lookup addon.  Here are some page times.

Hughesnet Community - https://community.hughesnet.com/

Redirect	0	0
App cache	0	0
DNS lookup	0	0
TCP connection	0	0
TCP request	20	13634
TCP response	13654	10
Processing	13658	3082
onload event	16740	1

Gallifrey Base - https://gallifreybase.com/gb/

Redirect	0	0
App cache	1	4
DNS lookup	5	1
TCP connection	6	988
TCP request	994	1746
TCP response	2740	397
Processing	2744	1658
onload event	4402	6

My HughesNet

Redirect	0	0
App cache	1705	0
DNS lookup	1705	0
TCP connection	1705	0
TCP request	1712	748
TCP response	2460	1
Processing	2462	4982
onload event	7444	1

World of Warcraft forums - https://us.forums.blizzard.com/en/wow/

Redirect	0	0
App cache	1	0
DNS lookup	1	0
TCP connection	1	0
TCP request	15	6321
TCP response	6336	1
Processing	6340	8743
onload event	15083	1

During these tests, testmy.net was giving the following latency pattern:

https://testmy.net/latency?tID=wrawnynbgt

which varied between 887ms and 2608ms

To date, I have not heard back from HughesNet on this issue.