Forum Discussion
Xfinity Says I Need To Replace My CM3000
Hello, A couple of days ago I had a technical support call with Xfinity regarding a streaming issue. During the call, they checked modem status and claimed that the modem was generating errors. T...
The ISP is Xfinity.
Xfinity has sent a tech to the building who claims that nothing is at issue with the line in.
Xfinity has claimed that the issue is the modem (CM2000).
That is why I sourced a new Xfinity gateway as well as a new CM3000.
The current agenda is to ask Xfinity to test for modem issues from their servicing node to the current CM2000 - then implement the Xfinity gateway and test again. That will supposedly determine if modem issues/errors are/were with the CM2000 or an issue with the servicing node.
The current obstacle is the ability to reliably/repeatably launch a modem test protocol from a browser interface such that results can be observed and/or recorded.
I have had one phone call where the agent claimed modem errors, and a different phone call where the agent claimed no errors. This is not a reliable test.
That is why I would like to know if there is an official Xfinity connection test that can be launched at will by a user that will show modem errors/issues.
The ISP will and may say anything to avoid taking the right action on getting this fixed. Not the first time.
I bet if you take the CM2000 to a different location, might find a difference.
Those errors in the event log which is what that is for points to issues ON the ISP side of the line. The tech only testing up to the building? How about from the building debarkation to your modem? Huh?
If this is an apt building, then tech should have tested a bit more from the entry point in the building up to your modem.
Something to check with maybe the XF thread on reddit. May find someone there that has more depth on what to test for on the ISP side.Good Luck.
The correctables/uncorrectables columns under the downstream bonded channels, reset to zero when you reboot your modem.
Reboot your modem, let it run for a set time frame, such as a week, and download your values. That is the best you are going to get for a repeatable at will test you can document.Your original post indicated a problem with a CM3000. Now you are having problems with a CM2000?
