I’ve been trying to get to the bottom of a fairly subtle network issue we’ve been seeing recently. We’ll likely have to book an engineer, but I wouldn’t mind understanding what the problem could be. Any input would be greatly appreciated.
Basically the LAN clients are seeing frequent TCP drops and handshake timeouts. So video streaming generally works okish, but going on the web is a pain, so is VPN and video conferencing, which makes wfh difficult.
The setup is a Hub 3 in modem mode with an Asus router. The router is not the problem, same issue when the Asus if off and the Hub 3 is in router mode instead.
Looking at the downstream:
Power levels: lowest is 9.3dB, highest is 14.1. That is too high surely? SNR is 38 for most channels, 40 for some. That seems ok? RxMER is 38.9 … 40.1 for all channels. Very few pre RS errors and not going up, post RS error count stable at zero for all channels.
Upstream: posting as image:
Looking at the log, there are regular (frequent?) T3 timeouts, about 10/hr and always on the same channel. Also I have seen some channels drop down to 16QAM. Seems the upstream is limping?
The coax from the VM outlet to the Hub is squeezed into the conduit rather tightly, so I thought there might be some kinks in there causing issues. However the problem only started recently, the same installation worked fine before. I also tested with a free coax cable instead and that didn’t really make any difference.
Connection is directly from the wall socket to the hub, no splitters or attenuators, just a galvanic isolator in between.
10 is the max downstream power level, so at least one is out of spec. Please post full up/downstream and network stats. You will need a technician’s visit to rectify, either phone customer services or wait a few days for a VM staff member to get to your post.
Tudor There are 10 types of people: those who understand binary and those who don't and F people out of 10 who do not understand hexadecimal c1a2a285948293859940d9a49385a2
Set up a BQM monitor which showed massive packet loss to the router until about 8pm yesterday, at which point the problem magically disappeared. Clearly a VM network problem upstream. The massive T3 timeouts have gone now. Nothing to do with the downstream power levels, these are completely unchanged.
VM are hilarious. Had an engineer booked, booking was cancelled. Finally got through to someone on the phone (!) after about an hour in the loop, had them schedule a new booking for today, only to have that silently cancelled - no engineer showing up, no phonecall, no text no mail. Great.
VM customer service truly is ... unique. Not in a good way. Luckily our contract is running out soon and will most decidedly not be extended. Fingers crossed there will be no more outages like that until then.