I was seeing that error occasionally on different nics at steady state. In the cases I saw, ptp4l would recover within about a minute. I did push a fix to starlingx 3.0 to increase tx_timestamp_timeout which seems to have stopped the issue from happening in my setups. https://review.opendev.org/#/c/691113/ Of course, it was a bit unpredictable so it's hard to be sure it's completely gone. Gerry From: Philip_Wang@alphanetworks.com [mailto:Philip_Wang@alphanetworks.com] Sent: Thursday, November 21, 2019 11:02 PM To: starlingx-discuss@lists.starlingx.io Subject: [Starlingx-discuss] PTP slave send delay request packet fail Dear group, I use starlingx 2.0 to test PTP slave at AIO-SX. I check log then find error messages as below: 2019-11-21T14:38:25.000 controller-0 ptp4l: notice [77834.645] port 1: UNCALIBRATED to SLAVE on MASTER_CLOCK_SELECTED 2019-11-21T14:38:25.000 controller-0 phc2sys: info [77834.868] port 0015b2.fffe.a92e24-1 changed state 2019-11-21T14:38:25.000 controller-0 phc2sys: info [77834.868] port 0015b2.fffe.a92e24-1 changed state 2019-11-21T14:38:25.000 controller-0 phc2sys: info [77834.868] reconfiguring after port state change 2019-11-21T14:38:25.000 controller-0 phc2sys: info [77834.868] selecting CLOCK_REALTIME for synchronization 2019-11-21T14:38:25.000 controller-0 phc2sys: info [77834.868] selecting enp0s25 as the master clock 2019-11-21T14:38:53.000 controller-0 ptp4l: info [77862.895] rms 4917165483502 max 70368744178111 freq -74459 +/- 2613 delay 14401 +/- 230 2019-11-21T14:38:54.000 controller-0 phc2sys: info [77863.871] rms 2218 max 15666 freq -73519 +/- 175 delay 1669 +/- 115 2019-11-21T14:39:06.000 controller-0 ptp4l: err [77875.993] timed out while polling for tx timestamp 2019-11-21T14:39:06.000 controller-0 ptp4l: err [77875.993] increasing tx_timestamp_timeout may correct this issue, but it is likely caused by a driver bug 2019-11-21T14:39:06.000 controller-0 ptp4l: err [77875.993] port 1: send delay request failed 2019-11-21T14:39:06.000 controller-0 ptp4l: notice [77875.993] port 1: SLAVE to FAULTY on FAULT_DETECTED (FT_UNSPECIFIED) 2019-11-21T14:39:07.000 controller-0 phc2sys: info [77876.872] port 0015b2.fffe.a92e24-1 changed state 2019-11-21T14:39:07.000 controller-0 phc2sys: info [77876.872] reconfiguring after port state change 2019-11-21T14:39:07.000 controller-0 phc2sys: info [77876.872] selecting enp0s25 for synchronization 2019-11-21T14:39:07.000 controller-0 phc2sys: info [77876.872] nothing to synchronize Is it a driver bug?? OAM NIC is Intel I210. Best Reagrds, - Philip Wang --- Alpha Networks Inc. TEL: 886-3-5636666 EXT:6403 This electronic mail transmission is intended only for the named recipient. It contains information which may be privileged,confidential and exempt from disclosure under applicable law. Dissemination, distribution, or copying of this communication by anyone other than the recipient or the recipient's agent is strictly prohibited. If this electronic mail transmission is received in error, Please notify us immediately and delete the message and all attachments of it from your computer system. Thank you for your cooperation.