Re: [Linuxptp-users] PMC Reporting portState: Slave When Interface Not Active

[Vladimir O. <ol...@gm...>]

Hi Daniel,

On Fri, Sep 25, 2020 at 03:34:11PM +0000, Zuckerbrod, Daniel wrote:
> Hi,
>
> I am running two instances of ptp4l on the same machine (eth1 and
> eth2). It looks like the pmc utility is not choosing which interface
> correctly when trying to query it.
>
> If I only run ptp4l on eth1 but execute the following command:
>
> pmc -u -b 0 -i eth2 'GET PORT_DATA_SET'
>
> then I get the following output:
>
>       023580.fffe.00370c-1 seq 0 RESPONSE MANAGEMENT PORT_DATA_SET
>                 portIdentity            023580.fffe.00370c-1
>                 portState               SLAVE
>                 logMinDelayReqInterval  2
>                 peerMeanPathDelay       0
>                 logAnnounceInterval     1
>                 announceReceiptTimeout  3
>                 logSyncInterval         0
>                 delayMechanism          1
>                 logMinPdelayReqInterval 0
>                 versionNumber           2
>
> I would expect it to come back as blank. I need to be able to
> separately detect the portState of each of the two interfaces I have
> running.

You are misunderstanding the '-i' option, have you read the manual
description?

-i interface
       Specify the network interface. The default is /var/run/pmc.$pid
       for the Unix Domain Socket transport and eth0 for the other
       transports.

This is about the communication channel between pmc and ptp4l, not about
the interface to query. Consider the general case where ptp4l is running
on a remote host, you have no idea if it's called eth0 over there or
something else.

Thanks,
-Vladimir