@rob--nl 

 

Why have you decided to add P9s, instead of adding M5s: do you have use of powerline backhaul on P9s? What you describe may be related to low quality electrical wiring, or high levels of electrical noise (such as dimmable lamps, etc.). 

Check in Deco app under Signal Source for P9 units, if they use powerline backhaul. See more information here: General Questions about the Powerline or PLC Backhaul Feature of Deco

 

If using powerline was not your intention, you may wish to return P9s and replace them with M5s. If you want/need powerline backhaul, Google articles on "troubleshooting powerline Ethernet adapters." Which is what P9 is: a Deco mesh unit, kind of like M5, plus embedded powerline Ethernet adapter.

 

 

>Also I've noticed the app is failing quite often to show accurate online/offline status or not showing which devices are linked to which deco. 

 

There is a delay in updating the information, because your Deco mesh reports to TP-Link servers which then send updates back to Deco app. This is normal for TP-Link Deco, as long as information is eventually properly updated. 

@Alexandre. 

 

indeed, I've added P9s to use the powerline feature to extend my network to places I couldn't reach otherwise. 

 

One of the P9s is using the powerline, the other one seems to prefer wifi or might have difficulties reaching the base unit over power line. 

 

Using the network in access point mode, might that have an impact?

 

I was wondering if there is any logging to identify when issues occur to see what the problem is and maybe try to fix it. 

 

Regards,

Rob

 

 

@TP-Link 

 

thanks, have updated the firmware

 

few things I notice

1) CPU load is very high all the time, both before and after the upgrade btw

2) get regular 'operation failed' message while watching network status

3) logging is showing these errors

 

Mon Oct 18 14:25:28 2021 daemon.err /usr/bin/apsd: apsd_get_active_ports:214: Error: No such interface 25

Mon Oct 18 14:25:31 2021 daemon.err /usr/bin/apsd: apsd_get_active_ports:214: Error: No such interface 25

Mon Oct 18 14:25:34 2021 daemon.err /usr/bin/apsd: apsd_get_active_ports:214: Error: No such interface 25

Mon Oct 18 14:25:37 2021 daemon.err /usr/bin/apsd: apsd_get_active_ports:214: Error: No such interface 25

Mon Oct 18 14:25:40 2021 daemon.err /usr/bin/apsd: apsd_get_active_ports:214: Error: No such interface 25

@TP-Link 

 

in addition seeing these errors in the log as well

 

Mon Oct 18 15:14:53 2021 daemon.err nrd[4704]: estimatorCmnHandleBeaconReportEvent: Beacon report for STA BC:7A:BF:63:7A:70 in unexpected state 2

Mon Oct 18 15:14:53 2021 daemon.err nrd[4704]: estimatorCmnHandleBeaconReportEvent: Beacon report for STA BC:7A:BF:63:7A:70 in unexpected state 2

 

Mon Oct 18 15:15:01 2021 daemon.err nrd[4704]: estimatorCmnHandleBeaconReportEvent: Beacon report for STA BC:7A:BF:63:7A:70 in unexpected state 2

Mon Oct 18 15:15:01 2021 daemon.err nrd[4704]: estimatorCmnHandleBeaconReportEvent: Beacon report for STA BC:7A:BF:63:7A:70 in unexpected state 2

 

 

Mon Oct 18 15:14:50 2021 daemon.err uhttpd[6224]: ath22 no frequency information.

Mon Oct 18 15:14:50 2021 daemon.err uhttpd[6224]:

 

Mon Oct 18 15:14:48 2021 daemon.err uhttpd[6224]: iptables: No chain/target/match by that name.

Mon Oct 18 15:14:48 2021 user.emerg syslog: tp215,22[12173]:

 

Mon Oct 18 15:14:27 2021 daemon.err nrd[4704]: estimatorHandleValidBeaconReport: No local BSS reported in beacon report from BC:7A:BF:63:7A:70

 

 

in addition BC:7A:BF:63:7A:70 is my mobile phone Samsung Galaxy A51 running the tplink mobile app to manage/control the network

@TP-Link any suggestions based on the logging provided?

 

especially the 'no such interface' message is filling the log a lot at the moment, not sure if that is harming anything. 

network seems to be pretty stable now, though sometimes the red led is blinking at my M5 close to my workstation (the others I don't monitor that actively, so don't know for sure)

 

@TP-Link 

 

have changed the network from access-point to router mode

not seeing the same errors anymore now

 

will have a look at the logs later to see if new issues pop up

 

 

one remark on the P9s:

I have one in reach of the main Deco, is it by design it prefers Wifi connection over powerline in this case? I did validate powerline link between two sockets is working (using other powerline adapters)

 

@TP-Link 

 

now seeing in the logs, the following recurring messages

 

Thu Oct 21 09:04:28 2021 daemon.err tmpsvr[5090]: Error: memfree less than 15000 KB

Thu Oct 21 09:04:28 2021 daemon.err tmpsvr[5090]: Error: tdp_trans_relay

 

Thu Oct 21 09:04:28 2021 daemon.err tmpsvr[5090]: Error: CGI read error

Thu Oct 21 09:04:29 2021 daemon.err tmpsvr[5090]: Error: CGI produced less bytes than expected

 

Thu Oct 21 09:21:20 2021 daemon.err uhttpd[7300]: ath22 no frequency information.

Thu Oct 21 09:21:20 2021 daemon.err uhttpd[7300]:

 

and many of these, but not sure if this is actual errors or 'just' client ip management

 

Thu Oct 21 09:22:55 2021 daemon.err udhcpd[16361]: Sending ACK to 192.168.68.101

Thu Oct 21 09:22:55 2021 daemon.err udhcpd[16361]: clear ip c0a84465

 

 

 

is the CPU load shown in http://tplinkdeco.net/ for each deco you select? is it refreshing after you select another deco node?

CPU is 'better' now, not 100% all the time, so in general it seems the network is doing way better in router mode compare to access point mode