Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

PTP enabled switch as a Slave #18

Open
kondou opened this issue Apr 25, 2023 · 3 comments
Open

PTP enabled switch as a Slave #18

kondou opened this issue Apr 25, 2023 · 3 comments

Comments

@kondou
Copy link

kondou commented Apr 25, 2023

Hi,

This is not an issue, but just fyi.

I have two grandmaster on the same network, and one of them is for backup.
Now I tried to make IES3110-8TF-R as Slave with this backup just as an experiment.

  1. create Slaveonly clock
    image
  2. enable port where grandmaster is connected, and go to Ports Configuration
    image
  3. change Not Slave to True
    image
  4. IES3110-8TF-R now runs as slave
    image

Offset From Master looks like 120μs, but help page says 'measured in ns' which I don't think so... If the offset is really 120μs, it seems large number. This may due to Clock Adjustment method, but I have no idea how to use internal Timer as Clock Adjustment method like E2eTransp does, but it's always Software.

@jclark
Copy link
Owner

jclark commented Apr 26, 2023

Thanks for the info.

The Offset from Master is 0.000,120,190 which I think is just a weird way for formatting 0.000120190, which is presumably a number of seconds with nanosecond accuracy, as you say 120μs. That interpretation is consistent with the mean path delay which would be 2.7μs, which seems plausible.

120μs does seem high, but plausible if it is using an unintelligent software-based adjustment method.

I don't really understand what the Slaveonly device is.

Have you tried the CLI? That sometimes has details/options that the GUI omits.

@kondou
Copy link
Author

kondou commented Apr 26, 2023

For CLI, I didn't try so much related to PTP at the moment. Before fumbling CLI configuration, I do want to understand what clock means in the configuration context. That clock looks like meaning which clock the switch behaves as, but I'm often confused and assume the clock type of connected device... And I'd also want to know which parameter is valid for which clock. The document and help page look not enough to understand.

For the past week, I've been asking sales guy in fs.com, over email, to clarify what I don't understand. But he looks like not understanding what I'm asking nor what he wants to tell me, which comes from engineers in fs.
Tomorrow, I will have a online meeting with engineers at fs.com. I hope I can get much more useful information.

@kondou
Copy link
Author

kondou commented Apr 27, 2023

I expected it would be an online meeting today, but actually it was chat with an engineer. Here are what was new or clear to me during the chat. Please note that these are applied to IES3110-8TF-R.

  • Clock Instance needs to be only one.
  • E2eTransp is the right one to be selected.
  • Do not change any of Ports Configuration regardless of clock type of the PTP device connected to the port.
  • No change needed in PTP Clock's Configuration and Status except for Port Enable. (He didn't mention Protocol, but this needs to be set to IPv4Multi as James suggested)
  • PTP cannot be used to synchronize internal clock of the switch if E2eTransp is selected.
  • SNTP needs to be configured to have correct time in the switch.
  • There is no RTC (Real Time Clock) in the switch, hence the clock always starts from Epoch (1970/01/01 00UTC) when power cycled.

It was time consuming chat, and the engineer didn't explain the background or the concept of the configuration, but gave answers only to what I asked. I could only understand what is needed for my situation, but there is almost no information related to PTP other than that.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants