08-06-2013 11:45 AM
Hi - found another post on this topic whereby some blame the 2Wire RG; others believe it to be a line problem (bridge taps/interference). A bad vrad port seemed to be the problem for one of the posters.
I have 12.5/1.5 service plan. When I am doing nothing to consume bandwidth, Speedtest.net shows I am getting what I am paying for (11.6/1.5).
When I kick off an online backup (Iron Mountain Connected backup), speedtest.net shows between 0.25/0.3 and 1.4/0.4
I've been getting nowhere slowly with AT&T.
I've spoken to several level 1 techs since last week; they overnighted me a new modem (despite my assurances that this was NOT the problem...); had a prem tech out here this morning saying all was well. I objected and he called his manager Robert Williams who told him to tell me he was in a meeting and would call me in 10 - 15 minutes.
That was about 4 hours ago, during which time an admin called me to confirm all was well. When I told her "No", she cold transferred me to Level 1 again! I objected and they bounced me to Level 2. Level 2 called the prem tech twice over a 45 minute period reminding him that I hadn't yet heard from his manager -- but still no return call.
Level 2 confirmed what I have been told in the past: that upstream and downstream services use separate protocols and should not affect each other. Further, Level 2 said that the "Quality Check" tool shows overall status failure and line status failure, meaning that this is a line quality issue. I had been told this by the Level 1 tech yesterday who scheduled the prem tech, but the prem tech couldn't see these status failures onsite. Interestingly, it seems the status failures may occur only while I am running the online backup and maxing out my upload bandwidth which surprises me; I would think their quality tests are checking hardware characteristics like signal, not data packets...?
Level 2 also said that I was using negligible bandwidth when speedtest.net was reporting such low bandwidth metrics; he determined this via the NAT sessions:
e.g. RG Settings, Diagnostics, NAT tab shows
session table 989/1024 available, 0/512 used in inbound sessions:
So I still have 989 sessions available for download which the tech explained was evidence of a problem since speedtest.net was showing such low download bandwidth available (it seems the backup software doesn't use NAT because UV Realtime shows my upstream maxed out at around 1.5 even though the NAT table shows 0/512 used)
And then I just got an automated call saying something like "your problem was a network issue that has identified and fixed"... but nothing is different on my end.
I think it must be every single time I have had a problem over the past several years with Uverse, it has been a support nightmare that consumed man days of my time... (sorry... ranting....)
I don't know how to proceed... hoping someone can offer some suggestions. Screenshots in next post for those who are still reading...
08-08-2013 7:28 AM
I normally would suggest a figure less than 95%, but yes
Sorry, I read your post rather quickly the first time through. You need to be sure that there's enough "idle" time for the returned packets to get through. Different applications will throttle their sending differently, and its best to leave a little room for other bursty devices to get their traffic through as well. I'd recommend experimentation.
08-12-2013 3:37 PM
08-12-2013 4:40 PM
Asus is making some routers with pretty impressive features and specs. I've been planning to buy one, but haven't done so yet. Here is a link to a page showing what QoS settings are available on one particular ASUS router:
It looks like Automatic setting may give you what you're looking for without a lot of muss or fuss. But if not, you have some tuning options to play with.