Here we go again! About a week ago I called phone support about the problems. Was issued ticket # 125796 184. Engineering never called back. I called support again on 12-21-2019. They checked my ticket and said I wanted to know how satellite internet worked. Wrong! I gave them a laundry list of symptoms that they apparently ignored. I ran through my list of symptoms again. I was then forced to jump through the usual hoops (again) before any action would be taken. I complied. They then issued a new trouble ticket, # 126031828 and said engineering would call. I hope they follow through this time!!!! Symptoms: * Although testmy.net has sometimes previously indicated =>20 Mbps the actual *throughput* has been abominable for a long time. * Testmy also sometimes reports download speeds well in excess of 1,500 Mbps which is impossible. * Today testmy.net was indicating a download of ~3 Mbps using 25 MB testing and the actual throughput is even worse than usual * Testmy.net connection chronically dropped and reconnected. * Testmy.net hung and failed the latency test. * There are very, very long pauses during a web page load. Simply no activity, waiting... * Youtube low res videos constantly buffering. I don't even try streaming large videos, movies, etc. That is hopeless. * Often a web page fails to load at all. Repeated "reload" clicks will usually finally load the web site. * Testmy.net results: https://testmy.net/quickstats/KZ4AK Power down resetting of the HT2000W and flushing DNS have not effect. Signal:112, IPGW ID J2SDO068HNSIGW0103, ESN 13179424, Diagnostic Code 0000-0000-0000-0005 I am waiting for the call back from engineering sometime next week . If they bother to call back... It seems that Hnet is happy to sell their service with a draconian contract but do not provide the service advertised. If any Hnet moderators can help with this it would be very much appreciated!!! Woody - KZ4AK @Liz @Damian
The moderators might not respond until Monday.
In the mean time, you can edit the message by clicking the 3 dots on the upper right. I think I have seen folks say that it is best not not have the SAN showing. I guess somebody can do something nefarious with it, but I am not sure what.
I think you may be on the same beam and gateway with @MarkJFine , maybe you can compare notes a bit?
Can you try a traceroute to microsoft.com? We can see if your routing is similar to what Mark has if you are the same gateway. Mark mentioned this a few days ago, so it might be useful.
Also, you mentioned a diagnostic code, was that code from before or after you rebooted? I ask because the diagnostic code is just a shorthand code for the condition of the last 3 hours of the checks/X graph that the system control center shows, and if the thing is rebooted then the code may not mean much until it has been running for 3 hours to actually have some history to show in the code.
I tried that testmy latency test a couple of days ago, but it does not seem useful right now to me. I ran it 9am and again at 9pm to compare. Since 9am is almost always fast because of light load, and 9pm is the rush hour so to speak, I was hoping to get a feel for understanding the meaning in context. The result looked about the same which did not give me a good feeling about it. I then connected to another machine with a hardwired high speed connection and ran that test, and again I was seeing graphs at 2000ms - 3000ms, with 'ping' results well above 1000ms for a hardwired connection that can do traceroute to places in 100ms.
The testmy page has some explanation links below it that shed some light on this new beta-test feature. The ping the test uses is not a normal ping, but rather a TCP packet send. The difference being that the TCP packet can be subject to funny routing and data verification, the difference in time from a real ping may represent some funny condition with the backbone or the graph might have a bug yet. Perhaps some of the folks that know more about this can chime in. For me, seeing that a hardwired machine had a similar graph with crazy average "ping" times made me think the test may not be ready for prime time. After all, if I looked at those three graphs, I don't think I would have been able to tell which graph went to the machine with the high speed/low latency hardwire from the graph for the high latencysatellite during the most crowded point in the evening, and that seems to be the very point of a latency test.
We've actually been through this before.
I did not realise you posted before regarding this, and I have not read those long threads, so I will just drop out.
Regarding the traceroute timeout, I don't think that really does mean much -- I understand many machines are configured to block ping with the firewall so it just times out. Since the traceroute reached the next hop, I think we can know it made that hop without issue, and the latency was likely something less than what shows for the next hop. There are probably folks familiar with networking that can confirm.
Since we're on the same beam, and since I'm not experiencing the same slowdown, I'm going to go with that there's either something on @hamradio's LAN, modem, or dish alignment that's causing it.
Thanks for you reply. Cannot locate an edit option (no dots), but your warning is good to know.
If you're using an ad blocker, try temporarily disabling it and refresh the page. I've had a problem with uBlock Origin removing some of the page options for me, including the option to edit posts.