search

lorabasics / basicstation

LoRa Basics™ Station - The LoRaWAN Gateway Software
https://doc.sm.tc/station
Other
352 stars 181 forks source link

Time sync rejected --- #41

Open sgodey opened 5 years ago

sgodey commented 5 years ago

I am seeing the below messages when I am bringing up Gateway. Any ideas what is going wrong here?

2019-06-10 07:07:04.840 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:07:09.041 [SYN:VERB] Time sync rejected: quality=266 threshold=230 2019-06-10 07:07:11.142 [SYN:VERB] Time sync rejected: quality=399 threshold=230 2019-06-10 07:07:28.994 [SYN:INFO] Time sync qualities: min=70 q90=220 max=399 (previous q90=230) 2019-06-10 07:07:49.996 [SYN:INFO] MCU/SX1301 drift stats: min: -1.9ppm q50: -4.3ppm q80: -5.7ppm max: -18.6ppm - threshold q90: -10.5ppm 2019-06-10 07:07:49.997 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:08:32.001 [SYN:INFO] Time sync qualities: min=117 q90=139 max=146 (previous q90=220) 2019-06-10 07:08:32.001 [SYN:INFO] MCU/SX1301 drift stats: min: -1.0ppm q50: -4.3ppm q80: -5.2ppm max: -7.6ppm - threshold q90: -7.1ppm 2019-06-10 07:08:32.001 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:09:01.404 [SYN:VERB] Time sync rejected: quality=180 threshold=139 2019-06-10 07:09:16.106 [SYN:INFO] MCU/SX1301 drift stats: min: -1.4ppm q50: -4.3ppm q80: -5.2ppm max: -6.2ppm - threshold q90: -5.7ppm 2019-06-10 07:09:16.106 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:09:35.008 [SYN:INFO] Time sync qualities: min=118 q90=137 max=180 (previous q90=139) 2019-06-10 07:09:51.810 [SYN:VERB] Time sync rejected: quality=138 threshold=137 2019-06-10 07:10:00.211 [SYN:INFO] MCU/SX1301 drift stats: min: -3.3ppm q50: -3.8ppm q80: -4.3ppm max: -5.2ppm - threshold q90: -4.8ppm 2019-06-10 07:10:00.211 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:10:13.863 [SYN:VERB] Time sync rejected: quality=148 threshold=137 2019-06-10 07:10:26.464 [SYN:VERB] Time sync rejected: quality=142 threshold=137 2019-06-10 07:10:36.965 [SYN:INFO] Time sync qualities: min=75 q90=138 max=148 (previous q90=137) 2019-06-10 07:10:45.366 [SYN:INFO] MCU/SX1301 drift stats: min: -2.9ppm q50: -4.3ppm q80: -4.8ppm max: -14.3ppm - threshold q90: -4.8ppm 2019-06-10 07:10:45.366 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:10:49.567 [SYN:VERB] Time sync rejected: quality=161 threshold=138 2019-06-10 07:10:57.968 [SYN:VERB] Time sync rejected: quality=139 threshold=138 2019-06-10 07:11:21.071 [SYN:VERB] Time sync rejected: quality=162 threshold=138 2019-06-10 07:11:23.171 [SYN:VERB] Time sync rejected: quality=146 threshold=138 2019-06-10 07:11:29.472 [SYN:VERB] Time sync rejected: quality=140 threshold=138 2019-06-10 07:11:35.773 [SYN:VERB] Time sync rejected: quality=140 threshold=138 2019-06-10 07:11:37.873 [SYN:INFO] Time sync qualities: min=120 q90=146 max=162 (previous q90=138) 2019-06-10 07:11:37.873 [SYN:INFO] MCU/SX1301 drift stats: min: -2.4ppm q50: -4.3ppm q80: -4.8ppm max: -6.7ppm - threshold q90: -5.7ppm 2019-06-10 07:11:37.873 [SYN:INFO] Avg MCU drift vs SX1301#0: 1.0ppm 2019-06-10 07:12:13.577 [SYN:VERB] Time sync rejected: quality=148 threshold=146 2019-06-10 07:12:17.778 [SYN:VERB] Time sync rejected: quality=155 threshold=146

sgodey commented 5 years ago

Hello All:

Where can I read about this 'TimeSync' ? I am currently stuck here. Does station send any TimeSync messaged to LNS as specified in https://doc.sm.tc/station/tcproto.html#time-synchronization. I am not seeing any WSS messages to MUXS. Any thoughts ??? Please help.

-Sreekanth

beitler commented 5 years ago

In regular intervals Station collects some measurements for MCU/SX1301 clock synchronization. Each measurement round is assigned a timesync quality value which is defined by the time in microseconds it took to fetch the instantaneous xticks counter value from the SX1301, i.e. smaller is better. Station keeps statistics on the timesync quality and discards measurements whose quality is considered an outlier. Timesync quality outliers can occur for example if the Station process is preempted during the SPI transaction where the xticks counter is fetched. This could impact the MCU/SX1301 drift estimation and outliers are therefore discarded.

It should be noted that good MCU/SX1301 clock synchronization is not a strict requirement for Station to function properly. However, frequent time sync failures due to extreme jitter can be an indicator for certain failure modes. In your particular case you see that time sync measurements are being rejected because their quality is slightly worse than the 90th percentile of past measurements. Over time the threshold should slowly adapt and those warnings will cease. Therefore, this is not a problem at all. We may consider moving those log messages to the DEBUG level.

We are currently working on more extensive documentation on timing related aspects. In the next release you will find these sort of questions answered in the documentation.

smtc-bot commented 1 year ago

Thank you for your inquiry.

Customers are encouraged to submit technical questions via our dedicated support portal at https://semtech.force.com/ldp/ldp_support.

We invite all users to visit the LoRa Developer Portal Forum at https://forum.lora-developers.semtech.com and to join the thriving LoRa development community!