CrowdSupport®
jsbrown724
Level 3: Gumshoe

Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi, the DJA0230TLS was updated recently to firmware 18.1.c.0443-950-RA, and since that point have noticed an increase in the number of drop outs and re-connections to NBN via FTTN (100/40).  Sync speed is the same roughly at 109/40 and has not degraded.  Previously I could run between 25-30 days without a resync and now it seems to happen at least once every day.

 

have also noticed there are a lot of errors and warnings in the Log Viewer with the wifi doctor agent...

 

Has anyone else seen any issues?

Was this helpful?

  • Yes it was, thank you
  • No, I still need help
18 REPLIES 18
jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Further to the Event Log; below is what is shown in the Modem.  Strangely that is the last entry as well (nothing for 12/11/19):

 

Nov 11 23:42:07 user.warn wifi-doctor-agent[30496] [fhc_c
Nov 11 23:42:06 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:06 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:42:05 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:05 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:42:04 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:04 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:42:03 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:03 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:42:02 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:02 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:42:01 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:01 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:42:00 cron.info crond[9046] USER root pid 15042 cmd /sbin/trafficmon.lua
Nov 11 23:42:00 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:42:00 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:41:59 user.err wifi-doctor-agent[30496] [fhc] Failed sending stats object
Nov 11 23:41:59 user.warn wifi-doctor-agent[30496] [fhc_common] Trying to publish MQTT message when fhc state is "CONNECTING"
Nov 11 23:41:58 user.err wifi-doctor-agent[30496] [fhc_common] Connect failed, reconnection is scheduled
Nov 11 23:41:58 user.err wifi-doctor-agent[30496] [mqtt] MQTT error 8: in connect asynch - A TLS error occurred.
Nov 11 23:41:58 user.info wifi-doctor-agent[30496] [fhc_common] Trying to connect to broker at 192.168.0.1:8883

Level 25: The Singularity
Level 25: The Singularity

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

@Yastiandrie, looks like a problem with the wifi interfacing. Might be worth forwarding on to your contacts.

Never be afraid to back yourself when trying new things, just always make sure you have 3 escape routes if things go wrong.
Level 24: Supreme Being
Level 24: Supreme Being

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Wifi doctor was updated in this FW version. Will have to keep an eye on this and pass on the log excerpt

I am a Premium Services Support Consultant (Telstra Platinum, Telstra Smart Home, Locator/Device Locator and Neto POS). However, I am not an official representative on CrowdSupport.

IT Helpdesk and Technicial Support by Telstra Platinum
Smart Home Automation & Monitoring from Telstra
Helping Australians Find the Things that Matter Most with Telstra Locator
Run a Modern Retail Business with Neto Point of Sale
jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Thanks @Jupiter and @Yastiandrie.    I will save the log file and when I can reboot will see what that also shows.  Happy to provide any other information if required, this however I can confirm was definitely only seen after the firmware upgrade.  

 

Might be a few more quirks to iron out in the update (such as per @ausgumbie mentioned in his post).  The Mobile section does not display correctly using  Chrome Version 78.0.3904.97 (Official Build) (64-bit) on Windows 10. 

 

Chrome-Mobile.png

But does work on Edge and IE on Windows 10, and Safari on iPad.

 

IEEDGE-Mobile.png

 

Of course, after mentioning that no speed issues this morning, have found that this afternoon have dropped to 91/40.  Will see what occurs after rebooting later.

ausgumbie
Level 11: Detective

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi jsbrown724

 

Just came across your thread. Yes, we've had similar experiences (ref. my thread DJA0230TLS, Firmware Upgrade 18.1.c.0443-950-RA, & No 5GHz), although our technology is HFC, not FTTN.

 

My focal issue there was the Band Steering, but we also experienced a number of dropouts (maybe 3-4 in a row) during Friday night (8th November).

 

I didn't attribute those to the firmware upgrade. We'd been expecting some scheduled outages, and the Telstra outage webpage identified an outage in our street that night. Also, there are possibly disruptions due to bush-fire damage to Telstra infrastructure.

 

But, we've had another this evening (Tue. 12th November) which isn't declared on the outages webpage, so now I'm wondering.

 

I'll have a look back through the logs I've downloaded since last Friday, and post anything that might be useful for passing on.

 

Cheers

 

ausgumbie

Level 24: Supreme Being
Level 24: Supreme Being

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

@jsbrown724 I have only had the update for one day but have not noticed any drop outs and the Mobile backup displays Okay in Chrome running on a Windows 10 64 bit laptop.

 

 I have noticed the Internet Traffic Graph (Basic > Services > Traffic Monitor on DSL connection only displays data for the previous 10 minutes, starting at zero and peaking at the data used for the total period since modem was rebooted. Also the Total send and receive data is missing from graph.


  1. D9C6DDC9-7FC9-4710-BF90-1A539C22EAA4.jpeg
ausgumbie
Level 11: Detective

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi again

 

Just following up on my promise to post log extracts I thought might be helpful to @Yastiandrie's contacts.

 

(** NOTE: Mods, I try to ensure I don't include any personal identifiers in the extracts. But if you see something you recognize as such, would you mind deleting it for me? Thanks)

 

OK - Friday 8th November and Tuesday 12th were the two days we noticed dropouts, so I'm looking at logs from that period.

 

(1) Log: To Fri 08-11-2019-10:38:02

The really notable thing here was a "jump" in date from Mon 21-10-2019-20:30:24 to Fri 08-11-2019-05:05:17 Smiley Surprised.

 

Mon Oct 21 20:30:22 2019 user.info root: Mmpbx stopped, mmpbx.state=NAstopped
Mon Oct 21 20:30:23 2019 user.notice firmware-upgrade-ext: All component firmware is correct.
Mon Oct 21 20:30:24 2019 user.info root: Mmpbx starting mmpbx.state=STARTING
Mon Oct 21 20:30:24 2019 daemon.err modprobe: dsphal is already loaded
Fri Nov 8 05:05:17 2019 kern.notice kernel: [ 140.648000] [SLICSLAC] slicslac_init: Starting SLIC/SLAC driver.
Fri Nov 8 05:05:17 2019 kern.debug kernel: [ 140.649000] [SLICSLAC] slicProvInit: Allocated memory = 644
Fri Nov 8 05:05:17 2019 kern.warn kernel: [ 140.770000] [MMTONEGEN] :E: allocateBrcmHwTimer:448 - failed to allocate a brcm HW timer
Fri Nov 8 05:05:17 2019 kern.warn kernel: [ 140.770000] 
Fri Nov 8 05:05:17 2019 kern.warn kernel: [ 141.160000] [ERROR bcmlog] log_proc_write,358: Invalid Parameters 'ept' '0'
Fri Nov 8 05:05:17 2019 kern.warn kernel: [ 141.160000] 
Fri Nov 8 05:05:17 2019 user.notice cwmpd: Voice not up yet ... delaying startup
Fri Nov 8 05:05:19 2019 user.info root: the 2 values of max contact are equal
Fri Nov 8 05:05:19 2019 daemon.notice mobiled: Network registration state changed to: registered

 

I don't know that this actually represents a jump of 17 days - every time our old F@st5355 reset, the first batch of lines started in January 2013. At some point, the clock kicked in and events were recorded at the correct time.


(2) Log: Fri 8-11-2019-17:22:00 to Tue 12-11-2019-23:16:53

 

With our old F@st5355 logs, I could pick outages reasonably easily. But, even with the help of the manual, I'm afraid I'm pretty hopeless with the DJA0230TLS logs.

 

What I've done, first up, is analyzed a log from Fri 8th to Tue 12th November, and looked for some diagnostic lines in @cf4's manual (Version 1.8, Last updated 01102019, pages 38/9, "Event Log - Extract from log at beginning of link loss" etc).

 

Working with an Excel workbook, I found:

 

    1 x instances = daemon.notice netifd: Network device 'wl1_1' link is down
    1 x instances = daemon.notice netifd: Network device 'wl1_2' link is down
    6 x instances = daemon.notice netifd: Network device 'eth1' link is down
  16 x instances = daemon.notice netifd: Network device '464-wwan_6_4' link is down
  22 x instances = daemon.notice netifd: Network device 'eth5' link is down
  24 x instances = daemon.notice netifd: Interface 'wwan_6' has link connectivity loss
  82 x instances = daemon.notice netifd: Network device 'eth4' link is down
115 x instances = daemon.notice netifd: Interface 'wan6' has link connectivity loss
121 x instances = daemon.notice netifd: Interface 'wan' has link connectivity loss

 

If someone can identify for me which are most likely to be associated with an internet dropout, I can go back and look at the lines around them.

 

Many thanks

 

ausgumbie

 

Level 24: Supreme Being
Level 24: Supreme Being

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

 

"I don't know that this actually represents a jump of 17 days - every time our old F@st5355 reset, the first batch of lines started in January 2013. At some point, the clock kicked in and events were recorded at the correct time."

 

My log does the same thing, 21 of October could the default date written into the firmware and the correct date doesn't get displayed till the link is extablished and modem can synchronise with time server.

 

eth5 is the WAN port and as you have an HFC connection the "22 x instances = daemon.notice netifd: Network device 'eth5' link is down" would indicate when the NBN link was down.

 

"1 x instances = daemon.notice netifd: Network device 'wl1_1' link is down
1 x instances = daemon.notice netifd: Network device 'wl1_2' link is down"

 

Are your 2.4 GHz and 5 GHz WiFi.

 

"6 x instances = daemon.notice netifd: Network device 'eth1' link"

 

is down is LAN Port 2

 

"82 x instances = daemon.notice netifd: Network device 'eth4' link is down"

 

is LAN port 4

 

"16 x instances = daemon.notice netifd: Network device '464-wwan_6_4' link is down
24 x instances = daemon.notice netifd: Interface 'wwan_6' has link connectivity loss
115 x instances = daemon.notice netifd: Interface 'wan6' has link connectivity loss"

 

All thes refer to the refer to the 4G backup.

 

"121 x instances = daemon.notice netifd: Interface 'wan' has link connectivity loss"

 

I am not sure what this refers to probably both the 4G backup and the WAN port.

 

It was definitely easier to read and understand the F@st5355 log, it is also easire to read and understand the Arcadyan Gen 2 Smart Modems log.

 

The main problem with the Technicolor log is it gets reset at each reboot which makes it difficult to find faults that cause modem reboots. The Arcadyan and F@st5355 logs anly reset when modems were factory reset.

 

 

 

 

 

 

 

 

 

 

 

ausgumbie
Level 11: Detective

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi @cf4

 

As ever, my awe and thanks Smiley Happy !

 


@cf4 wrote:

My log does the same thing, 21 of October could the default date written into the firmware and the correct date doesn't get displayed till the link is established and modem can synchronise with time server.


On that point, I'm going to assume my earliest recent log records the beginning of a reset to the point of downloading the log.

 

For interest, I'll re-read the lines from the start of the log to the changeover and see what I make of it all.

 

I'll return to your other points after a re-read,

 

Cheers

 

ausgumbie

ausgumbie
Level 11: Detective

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi @cf4 

 

Back again.

 

  • eth5 is the WAN port and as you have an HFC connection the "22 x instances = daemon.notice netifd: Network device 'eth5' link is down" would indicate when the NBN link was down.

Aaaaaaah. That implies more outages than I'd thought we'd had Smiley Sad. I'll try to map these lines timewise and see if there's a pattern.

 

  • "1 x instances = daemon.notice netifd: Network device 'wl1_1' link is down
    1 x instances = daemon.notice netifd: Network device 'wl1_2' link is down"
  • Are your 2.4 GHz and 5 GHz WiFi.

Many thanks - again I'll look at the timing on those and see if anything matches up with my notes.

 

  • "6 x instances = daemon.notice netifd: Network device 'eth1' link is down" is LAN Port 2

Again, thanks - I can work from that.

 

  • "82 x instances = daemon.notice netifd: Network device 'eth4' link is down" is LAN port 4

Ditto.

 

  • "16 x instances = daemon.notice netifd: Network device '464-wwan_6_4' link is down
    24 x instances = daemon.notice netifd: Interface 'wwan_6' has link connectivity loss
    115 x instances = daemon.notice netifd: Interface 'wan6' has link connectivity loss"
  • All thes refer to the refer to the 4G backup.

Hmm. I might leave this one till last.

 

  • "121 x instances = daemon.notice netifd: Interface 'wan' has link connectivity loss"
  • I am not sure what this refers to probably both the 4G backup and the WAN port.

I'll compare this with my 4G observations. Maybe some connections.

 

  • The main problem with the Technicolor log is it gets reset at each reboot which makes it difficult to find faults that cause modem reboots. The Arcadyan and F@st5355 logs only reset when modems were factory reset.

Smiley Sad

 

Thanks again, @cf4 

 

ausgumbie

jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi @ausgumbie , @cf4 

 

Did a soft restart of the modem yesterday evening and all speeds returned to normal.  Interestingly the log viewer still had the same information as posted above, yet if I did an export the full correct logging was recorded and in the export file.   Now for whatever reason just taking a look and the log viewer display is now operating as we would expect, plus I can find none of the previous errors.  Didn’t change a thing!

 

Yesterday’s restart was the first reboot/restart since the device upgraded to the new firmware.

 

Whilst no dropout yet, the error counts are higher than they previously used to be so will keep an eye on the log and check when one does occur. 

jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Checking this morning and on closer inspection  it looks as if the Log Viewer is not showing the latest entries on screen but Ones that are more recent.  Last entry is 13/11 at around 7pm.  Export function has everything including current, so log entries are being recorded.

Level 24: Supreme Being
Level 24: Supreme Being

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

@jsbrown724 the wifi doctor error spam should be resolved Smiley Wink

I am a Premium Services Support Consultant (Telstra Platinum, Telstra Smart Home, Locator/Device Locator and Neto POS). However, I am not an official representative on CrowdSupport.

IT Helpdesk and Technicial Support by Telstra Platinum
Smart Home Automation & Monitoring from Telstra
Helping Australians Find the Things that Matter Most with Telstra Locator
Run a Modern Retail Business with Neto Point of Sale
Level 24: Supreme Being
Level 24: Supreme Being

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

My log viewer has the same viewer. All the records are in the exported file bu last record idisplayed in log veiwer is for 18:51 on 13th. The Technicolor Gen 2 modems have the same problem with the log viewer.

jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Thanks @Yastiandrie.

 

Has anything been changed with the VDSL aspects of this.  Since receiving this upgrade I am definitely seeing frequent drop outs and also getting line rate changes.  Now it could just be coincidence but this has been extremely solid for nearly 24 months.

 

The Log file contains entries such as:

Fri Nov 15 07:03:18 2019 kern.crit kernel: [205492.294000] Line 0: Rate Change, us=40599, ds=102994
Fri Nov 15 07:03:18 2019 kern.warn kernel: [205492.294000] bcmxtmcfg: XTM LinkInfo. P-0, S-UP(1), Service =PTM(2)
Fri Nov 15 07:03:18 2019 kern.warn kernel: [205492.294000] bcmxtmcfg: ReconfSAR P-0, T-2
Fri Nov 15 07:03:18 2019 kern.crit kernel: [205492.680000] Line 0: Rate Change, us=40599, ds=101831
Fri Nov 15 07:03:19 2019 kern.crit kernel: [205493.065000] Line 0: Rate Change, us=40599, ds=100761
Fri Nov 15 07:03:19 2019 kern.crit kernel: [205493.451000] Line 0: Rate Change, us=40599, ds=99751
Fri Nov 15 07:03:20 2019 kern.crit kernel: [205493.836000] Line 0: Rate Change, us=40599, ds=98803
Fri Nov 15 07:03:20 2019 kern.crit kernel: [205494.222000] Line 0: Rate Change, us=40599, ds=97885
Fri Nov 15 07:03:20 2019 kern.crit kernel: [205494.607000] Line 0: Rate Change, us=40599, ds=96998
Fri Nov 15 07:03:21 2019 kern.crit kernel: [205494.993000] Line 0: Rate Change, us=40599, ds=96172
Fri Nov 15 07:03:21 2019 kern.crit kernel: [205495.378000] Line 0: Rate Change, us=40599, ds=95377
Fri Nov 15 07:03:22 2019 kern.crit kernel: [205495.764000] Line 0: Rate Change, us=40599, ds=94551
Fri Nov 15 07:03:22 2019 kern.crit kernel: [205496.149000] Line 0: Rate Change, us=40599, ds=93786
Fri Nov 15 07:03:22 2019 kern.warn kernel: [205496.359000] bcmxtmcfg: XTM LinkInfo. P-0, S-UP(1), Service =PTM(2)
Fri Nov 15 07:03:22 2019 kern.warn kernel: [205496.359000] bcmxtmcfg: ReconfSAR P-0, T-2
Fri Nov 15 07:03:22 2019 kern.crit kernel: [205496.535000] Line 0: Rate Change, us=40599, ds=93022
Fri Nov 15 07:03:23 2019 kern.crit kernel: [205496.920000] Line 0: Rate Change, us=40599, ds=92318
Fri Nov 15 07:03:23 2019 kern.crit kernel: [205497.306000] Line 0: Rate Change, us=40599, ds=91645
Fri Nov 15 07:03:23 2019 kern.crit kernel: [205497.691000] Line 0: Rate Change, us=40599, ds=91186
Fri Nov 15 07:03:24 2019 kern.crit kernel: [205498.334000] Line 0: Rate Change, us=40599, ds=90850
Fri Nov 15 07:03:24 2019 kern.crit kernel: [205498.719000] Line 0: Rate Change, us=40599, ds=90636
Fri Nov 15 07:03:25 2019 kern.crit kernel: [205499.105000] Line 0: Rate Change, us=40599, ds=90483
Fri Nov 15 07:03:25 2019 kern.crit kernel: [205499.491000] Line 0: Rate Change, us=40599, ds=90299
Fri Nov 15 07:03:26 2019 kern.crit kernel: [205499.876000] Line 0: Rate Change, us=40599, ds=90146
Fri Nov 15 07:03:26 2019 kern.crit kernel: [205500.261000] Line 0: Rate Change, us=40599, ds=90116
Fri Nov 15 07:03:26 2019 kern.warn kernel: [205500.421000] bcmxtmcfg: XTM LinkInfo. P-0, S-UP(1), Service =PTM(2)
Fri Nov 15 07:03:26 2019 kern.warn kernel: [205500.421000] bcmxtmcfg: ReconfSAR P-0, T-2

 

When I have rebooted the Modem previously it goes back to the expected speeds of 109/40 with line sync max speeds of 119/50.  The one thing that appears to have changed is the noise margin reported where it used to be 10db/8db it is now around 12db/5db.

 

@cf4 any thoughts also appreciated.

Level 24: Supreme Being
Level 24: Supreme Being

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

There are no speed change events in my modem's log viewer, nor have I experienced any drop outs. The noise margin of my link is about 7 down and 6 up. For error free performance the modem and Node try to keep the noise margin to a minimum of 6. Something might have happened to your line about the same time as firmware update causing an intermittent fault which would explain the lower noise margin, the dropouts and changes in speed. The DSL Diagnostics will give a good indication of any problems with the NBN link. (Advanced > Broadband > DSL Diagnostics). The image below is an example of a normal connection.

 

frontier_Link_Diagnostic.JPG

jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Thanks @cf4 

 

We had a widespread power failure on Friday evening, and after the power cycle of the modem and the node, the service is a little more "normal" than what it was.  Still have more errors than I would have had previously, but so far no reduction in speed.

Status191117.PNG

 

 

 

jsbrown724
Level 3: Gumshoe

Re: Increasing dropouts since Firmware Upgrade 18.1.c.0443-950-RA

Hi @Yastiandrie, thought it might be worth mentioning that the error log is now full of the following:

 

Tue Nov 26 07:35:40 2019 user.warn wifi-doctor-agent[27205]: [exec] request https://device-auth-ap.wifi-doctor.org/oauth/token - 0:35=Error
Tue Nov 26 07:35:40 2019 user.warn wifi-doctor-agent[27205]: [exec] got unexpected curl reply code 35 http_rc = 0
Tue Nov 26 07:35:40 2019 user.info wifi-doctor-agent[27205]: [auth] restart curl request delayed due to unexpected error (delay = 90 s) (1)
Tue Nov 26 07:35:40 2019 user.info wifi-doctor-agent[27205]: [curl] delayed (90) curl request to 'https://device-auth-ap.wifi-doctor.org/oauth/token'

 

Telstra Smart Modem

Plug in and connect in minutes. Smart.

Find out more
Find out more about the Telstra 24x7 App and direct debit ×

Avoid queueing up and never worry about late fees againSet up direct debit
Manage your services and take advantage of offers while you’re out and about with the Telstra 24x7® App Find out more

Need a hand or want to share your expertise?
Register for CrowdSupport and get involved

Register now