[review] LTE attach config and status
matthew stanger
stangerm2 at gmail.com
Tue Nov 27 18:28:34 UTC 2018
>
> I don't think that CID=1 is always the one that the
> module expects to use as initial EPS bearer
I'd agree, but so far the small set of carriers we've seen requiring
specific PDP's have all been #1.
Didn't Verizon require to use CID=4 for the same purpose in their
> modems?
Vzw is a little odd, they require their default APN in context #3, but
apparently, and we've yet to make this work,
they(Vzw) say that if you use context #1, which auto populates with
'vzwadmin' will, after the first attach the carrier,
self provision the rest of the context's.
Docomo in Japan wouldn't do 4G without use of context #1 populated.
SK Telecom in S. Korea wouldn't connect without using context #1, using
blank context #1 populated other context's
I recall a patch were we allowed specifying the CID to use explicitly
yes, and the non-simple interface. We still use it, and without it those
cases above would not work.
On Tue, Nov 27, 2018 at 9:37 AM Aleksander Morgado <aleksander at aleksander.es>
wrote:
> > 4 In 4G the 1 context (where default EPS bearer is created) generally
> does not require an APN, APN is required on other context to set up data
> sessions instead . In any case the rules are made by network operator
> >
>
> Unfortunately, I don't think that CID=1 is always the one that the
> module expects to use as initial EPS bearer, that may depend on the
> modem. So, if we were to add a logic that assumes "initial LTE bearer
> settings are those in CID=1" then that would need to be tied to a
> specific device.
>
> Didn't Verizon require to use CID=4 for the same purpose in their
> modems? (or something like that) @matthew stanger what was that about?
> I recall a patch were we allowed specifying the CID to use explicitly
> via Simple.Connect() for this very same purpose.
>
> --
> Aleksander
> https://aleksander.es
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/modemmanager-devel/attachments/20181127/79826aa1/attachment.html>
More information about the ModemManager-devel
mailing list