Thanks asbokid, I'll take a more in depth look later and see what I can find. Would be nice to be able to issue a command to get this since it is there somewhere for the web server to get the correct value.
Hi Stuart,
Sorry, I didn't read what you said properly.
IIRC, the embedded web server, at least on the HG612 connects via a Unix socket/pipe/IPC to several different binaries in a middleware layer to obtain this sort of data.
That middleware polls the DSL driver in kernelspace to check the status of the DSL layer. If it is found to be down, the uptime counter is reset. So it would seem that the middleware maintains the connection uptime.
In the HG612 file system there is an XML-based MIB document tree called
/etc/t_tree.xml.
It defines two
Uptime nodes, one for system uptime, one for uptime of the
WANPPPConnection. The latter node looks like this:
<Uptime CMO="0x8000C50D" Type="VALUE_TYPE_ULONG" AppName="cms|cwmp" MaxLength="0" DefaultValue="0"/>The
AppName attribute defines two binaries concerning the WANPPPConnection Uptime object':
cms and
cwmp.
cwmp (or CPE WAN Management Protocol) is a component of TR069 remote management.
cms is apparently the binary for the 2Wire customer premises management system.
WIth no docs, neither binary wants to do anything very useful:
# cwmp --help
CWMP app version: V100R002C05B021 cwmp app V1.2.3.0.0
CWMP stk version: V100R002C05B021 cwmp stk V1.2.3.0.0
Not save certification!!!
ConfigDefaultSsl return: 0
^@
# cms --help
Failed in VTOP_MSG_Init!! ulRet = 0x8016801d
I've got a feeling there's yet another layer of middleware (at least on the HG612), and that is provided by a binary called
MidServer. MidServer (I suspect) proxies queries, reads that MIB file (/etc/t_tree.xml) and then calls the relevant binary to actually obtain the requested 'object' for the client. Could be wrong though.
cheers, a