SNMP "smiv2" yang extension doesn't work on augmented nodes

[vvsosed]

I use IF-MIB.yang in order to add support for SNMP ifTable. All nodes inside IF-MIB.yang are already properly tagged with smiv2:oid, smiv2:max-access yang extension tags.
SNMP ifTable container (in IF-MIB.yang) contains ifSpeed leaf which holds current bandwidth in bits per second. Also, ifHighSpeed leaf is added as yang extension to hold bandwidth bigger than 4,294,967,295 bit/sec. So in my case, I filled both nodes as stated in the description and checked that they (base ifSpeed and augmented ifHighSpeed) held proper values through NETCONF. After, I tried to get the values with help of SNMP by using snmpwalk CLI tool. Unfortunately, snmpwalk's output consisted of values from the base ifTable container (i.e. ifSpeed) and none of the augmented nodes (i.e. mentioned earlier ifHighSpeed augmented node).

This issue was reproduced at latest clixon master revision 5f38b50.

[olofhagsand]

Re-created.
But it is somewhat unclear what is expected behavior. From OID perspective, those fields are in table:

  smiv2:alias "ifXTable" {
    description
     "A list of interface entries.  The number of entries is
      given by the value of ifNumber.  This table contains
      additional objects for the interface table.";
    smiv2:oid "1.3.6.1.2.1.31.1.1";
  }

That is in the range 1.3.6.1.2.1.31.1.1.1.1 - 1.3.6.1.2.1.31.1.1.1.19
Which is different from ifTable being in the range: 1.3.6.1.2.1.2.2.1.1 - 1.3.6.1.2.1.2.2.1.22
This seems to be the only instance of MIBs of augmentation.
Do you have a reference on how the walk should behave in an augmented table?
In particular which OID the ifXTable entries have and how they behave in a walk/getnext?

[olofhagsand]

Fixed snmp-get for augmented tables.
May remain work for getnext/walk

[olofhagsand]

getnext + walk should also work, and regression tests added.
Please verify

[vvsosed]

Thanks a lot! We will check it soon.