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

question about ti cc1352p-2 #2162

Closed
flower87lucky opened this issue Oct 19, 2019 · 115 comments
Closed

question about ti cc1352p-2 #2162

flower87lucky opened this issue Oct 19, 2019 · 115 comments

Comments

@flower87lucky
Copy link

)
zigbee2mqtt:info 2019-10-19T12:50:02: Starting zigbee-herdsman...
zigbee2mqtt:error 2019-10-19T12:50:09: Error while starting zigbee-herdsman
zigbee2mqtt:error 2019-10-19T12:50:09: Failed to start zigbee
zigbee2mqtt:error 2019-10-19T12:50:09: Exiting...
zigbee2mqtt:error 2019-10-19T12:50:09: Error: SRSP - SYS - version after 6000ms
at Timeout.object.timer.setTimeout [as _onTimeout] (/app/node_modules/zigbee-herdsman/dist/utils/waitress.js:44:24)
at ontimeout (timers.js:436:11)
at tryOnTimeout (timers.js:300:5)
at listOnTimeout (timers.js:263:5)
at Timer.processTimers (timers.js:223:10)
npm
ERR! code
ELIFECYCLE
npm
ERR! errno 1
npm
ERR! zigbee2mqtt@1.6.0 start: node index.js
npm
ERR! Exit status 1
npm ERR!
npm ERR! Failed at the zigbee2mqtt@1.6.0 start script.
npm ERR! This is probably not a problem with npm. There is likely additional logging output above.
npm ERR! A complete log of this run can be found in:
npm
ERR! /root/.npm/_logs/2019-10-19T12_50_09_649Z-debug.log
2019-10-19T20:50:09: PM2 log: App [npm:0] exited with code [1] via signal [SIGINT]
2019-10-19T20:50:09: PM2 log: App [npm:0] starting in -fork mode-
2019-10-19T20:50:09: PM2 log: App

@flower87lucky
Copy link
Author

serial:
/dev/ttyACM1
/dev/ttyACM0
/dev/serial/by-id/usb-Texas_Instruments_XDS110__02.03.00.18__Embed_with_CMSIS-DAP_L43000GM-if03
/dev/ttyAMA0
/dev/serial/by-id/usb-Texas_Instruments_XDS110__02.03.00.18__Embed_with_CMSIS-DAP_L43000GM-if00
input:

@flower87lucky
Copy link
Author

"serial": {
"port": "/dev/ttyACM0"
},
"advanced": {
"pan_id": 6758,
"channel": 15,

@hdo
Copy link

hdo commented Oct 19, 2019

I got the same error message with my 2652R board :-(
I also tried to change panid and channel in my configuration.

https://pastebin.com/KuNRvwnG

@bertran1
Copy link

bertran1 commented Oct 20, 2019

@Koenkk
Copy link
Owner

Koenkk commented Oct 20, 2019

Can you try with

"serial": {
"port": null,
},

@flower87lucky
Copy link
Author

Can you try with

"serial": {
"port": null,
},

Thanks.null is not allowed in Zigbee2mqtt Hass.io Add-on .I used Zigbee2mqtt Hass.io Add-on,maybe it may not support at present. I try to install ZigBee 2mqtt directly

@flower87lucky
Copy link
Author

Can you try with

"serial": {
"port": null,
},

I'm worried if I flash the firmware in the wrong way?
After cc1352p-2 flashed the firmware, I press reset key and it has no effect.

@Koenkk
Copy link
Owner

Koenkk commented Oct 21, 2019

@hdo
Copy link

hdo commented Oct 21, 2019

@Koenkk

I still have the problem that i need to soft reset first (via button press) to start the firmware.

Do you know how to disable this behaviour?

Little change request:

Let the green LED light up when firmware has loaded for quick check ;-)

@flower87lucky
Copy link
Author

Image 2

pi@raspberrypi:~ $ ls -l /dev/serial/by-id
total 0
lrwxrwxrwx 1 root root 13 Oct 21 14:23 usb-Texas_Instruments_XDS110__02.03.00.18__Embed_with_CMSIS-DAP_L43000GM-if00 -> ../../ttyACM0
lrwxrwxrwx 1 root root 13 Oct 21 14:23 usb-Texas_Instruments_XDS110__02.03.00.18__Embed_with_CMSIS-DAP_L43000GM-if03 -> ../../ttyACM1
pi@raspberrypi:~ $ test -w /dev/ttyACM0 && echo success || echo failure
success
pi@raspberrypi:~ $ test -w /dev/ttyACM1 && echo success || echo failure
success
pi@raspberrypi:~ $ 

@flower87lucky
Copy link
Author

DE2C8A3E876723407674E0E21AA6EAD3
When I connect to raspiberry pi 4b+ ,only the green led on.

@Koenkk
Copy link
Owner

Koenkk commented Oct 21, 2019

@fredrikgk chip revision C is an older revision, right? Is it binary compatible with the E revision?

@flower87lucky
Copy link
Author

@fredrikgk chip revision C is an older revision, right? Is it binary compatible with the E revision?

@Koenkk How to fix binary to compatible,thanks.

@fredrikgk
Copy link

@flower87lucky , @Koenkk , IC revision C is deprecated and not supported. SW is not binary compatible between rev. C and rev. E.

The only alternative is to order a new LaunchPad.

@hdo
Copy link

hdo commented Oct 21, 2019

@fredrikgk

BTW Do you have an idea why i need to soft reset (via button press) each time after USB plug to start the firmware?

@flower87lucky
Copy link
Author

@fredrikgk

BTW Do you have an idea why i need to soft reset (via button press) each time after USB plug to start the firmware?

I agree you.

@Koenkk
Copy link
Owner

Koenkk commented Oct 21, 2019

@fredrikgk thanks, I will close this.

@Koenkk Koenkk closed this as completed Oct 21, 2019
@james-fry
Copy link

@fredrikgk

BTW Do you have an idea why i need to soft reset (via button press) each time after USB plug to start the firmware?

I got caught out with the need for soft reset too.
(and thanks for your assistance with this in the z2m issue @hdo)

I agree with your suggestion that either/both of the following would be good:

  • No need for soft reset on plug in of device
  • One of the LEDs on the MCU side will be used to indicate status or "success"

@Koenkk are either of these doable? should I raise an issue on the zstack fw repo?

@hdo
Copy link

hdo commented Oct 28, 2019

@james-fry

This is not a simple task.
I compiled a custom firmware version with LED support and that is also not a good indicator for this problem. I learned the firmware is loading correctly. However the culprit is the onbard debugger of the launchpad. On my Intel NUC i need the soft reset but not on my raspberry pi. There seems to be an issue with the usb handshake or something like this. I don't have any issues using the UART so that's why i think it's the usb part. The UART have other issues though (sending leading ZERO in some circumstances).

@james-fry
Copy link

Thanks for the info @hdo
I am also on intel so maybe that is why I also need the reset, but its not a big deal when you know you need to do this.

@fredrikgk
Copy link

@hdo Do you have to press the reset button also when the USB cable has been disconnected for a long time?

@fredrikgk
Copy link

I compiled a custom firmware version with LED support and that is also not a good indicator for this problem. I learned the firmware is loading correctly. However the culprit is the onbard debugger of the launchpad. On my Intel NUC i need the soft reset but not on my raspberry pi. There seems to be an issue with the usb handshake or something like this. I don't have any issues using the UART so that's why i think it's the usb part. The UART have other issues though (sending leading ZERO in some circumstances).

This is strange, as the reset button only resets the CC1352P, not the debugger.

How is your LaunchPad configured (jumpers) when you are "using UART"?

@hdo
Copy link

hdo commented Oct 29, 2019

@hdo Do you have to press the reset button also when the USB cable has been disconnected for a long time?

I have to soft reset each time i plug in the usb power cable.

This is strange, as the reset button only resets the CC1352P, not the debugger.

I not sure about that.

How is your LaunchPad configured (jumpers) when you are "using UART"?

I removed the TX/RX jumpers.

It could also work without removing those jumpers but i noticed some preceeding '0'-byte which disturbs the communication, at least in my environment.

I'm currently working on a zigbee-herdsmann modification to use my tcp2serial adapter without socat (not working with docker) and this (0-byte) happens with the tcp2serial adapter.

Also see: 0-byte

@a-bailey
Copy link
Contributor

I seem to have a B revision of the launchpad board is this supposed to work?

pi@raspberrypi:~ $ test -w /dev/ttyACM0 && echo success || echo failure
failure
pi@raspberrypi:~ $ test -w /dev/ttyACM1 && echo success || echo failure
failure

getting failure on both write tests.

@dariuszqrek
Copy link

I have board revision B, chip revision E and have same error - zigbee2mqtt won't start.
what could be a problem ?

@timdonovanuk
Copy link
Sponsor

How do you know what revision board you've got? Going to be a bit annoyed if I wasted $50 on a board that isn't supported when the z2m docs don't mention that only certain revs work :(

@dariuszqrek
Copy link

dariuszqrek commented Dec 23, 2019 via email

@presslab-us
Copy link
Contributor

In my case I could always unplug and replug the USB and it would be fine. Does that work for you?
I didn't use the reset button because it didn't work well in the case I made.
Since updating the XDS110 I have rebooted many times and there has been no problem.
Maybe you can try removing the jumpers for TCK and TMS?

@upwindanderl
Copy link

In my case I could always unplug and replug the USB and it would be fine. Does that work for you?
I didn't use the reset button because it didn't work well in the case I made.
Since updating the XDS110 I have rebooted many times and there has been no problem.
Maybe you can try removing the jumpers for TCK and TMS?

hmmmm sounds promising! 👍

The thing is the server is not nearby me and i'm only remote on it. As well plugging or unplugging is a total difficult phonecall and some remote guidance.
The only thing i know is that i updated the last FW yesterday which was automatically recommended with uniflash as i've been there and again today big trouble after rebooting the server.
Next time im there i'll check exactly the fw version and what you mean with tck and tms :-)

@upwindanderl
Copy link

In my case I could always unplug and replug the USB and it would be fine. Does that work for you?
I didn't use the reset button because it didn't work well in the case I made.
Since updating the XDS110 I have rebooted many times and there has been no problem.
Maybe you can try removing the jumpers for TCK and TMS?

hmmmm sounds promising! 👍

The thing is the server is not nearby me and i'm only remote on it. As well plugging or unplugging is a total difficult phonecall and some remote guidance.
The only thing i know is that i updated the last FW yesterday which was automatically recommended with uniflash as i've been there and again today big trouble after rebooting the server.
Next time im there i'll check exactly the fw version and what you mean with tck and tms :-)

So in the end it looks pretty solid!
Had lot reboots because of server problems and of course some more restarts of HA and i can't remember to push the button again.
Didn't changed anything on the jumpers, the only solution for me to click yes after i was asked for some firmware update of the UNIFLASH tool

@likellindil
Copy link

In my case I could always unplug and replug the USB and it would be fine. Does that work for you?
I didn't use the reset button because it didn't work well in the case I made.
Since updating the XDS110 I have rebooted many times and there has been no problem.
Maybe you can try removing the jumpers for TCK and TMS?

hmmmm sounds promising! 👍
The thing is the server is not nearby me and i'm only remote on it. As well plugging or unplugging is a total difficult phonecall and some remote guidance.
The only thing i know is that i updated the last FW yesterday which was automatically recommended with uniflash as i've been there and again today big trouble after rebooting the server.
Next time im there i'll check exactly the fw version and what you mean with tck and tms :-)

So in the end it looks pretty solid!
Had lot reboots because of server problems and of course some more restarts of HA and i can't remember to push the button again.
Didn't changed anything on the jumpers, the only solution for me to click yes after i was asked for some firmware update of the UNIFLASH tool

Hi there,

My HA & Z2M are built on my NUC and I met exactly the same issue.
May I know how you fixed this issue in the end?
Now I have to unplug and re-plug it every one or two days to keep it working.

@joshuakoh1
Copy link

Has this been fixed yet?

@timdonovanuk
Copy link
Sponsor

My personal experience, coming from a CC2531 to the CC1352p-2 is the CC1352p-2 has been a nightmare. Weekly I get devices drop off my network, that need repairing. If I stop and restart Z2M I quite often need to repress the button on the side of the CC1352P, which is not great as physical access to it is not easy (it's up in my loft).

I tried to flash the CC1352P today using the command line TI tools and it actually managed to take out my entire Proxmox instance!

@joshuakoh1
Copy link

joshuakoh1 commented Dec 15, 2020

@timdonovanuk Yup I'm getting really upset about the need to press the reset button, I just wish this issue was made more obvious in the supported sticks page. I would not have purchased this at all if I knew about it from the start. Open to suggestions about a better stick without this issue as well.

Separately, how did you manage to pass through the device to VM on proxmox? Passing the USB port doesn't create any /dev/ttyACM0 on the VM.

@timdonovanuk
Copy link
Sponsor

I have to agree, based on my personal experience I don't think CC1352P should be in the list of recommended devices, but maybe it's just some of us. Today I'm experiencing more fun, some of my bulbs randomly dont work with Data request failed with error: 'No network route' (205) errors 🤦

@joshuakoh1 here are my proxmox settings, works fine for me:

image

Needs a hard reboot. Then I get these showing up:

~$ ls -l /dev/serial/by-id
total 0
lrwxrwxrwx 1 root root 13 Dec 14 15:57 usb-Texas_Instruments_XDS110__03.00.00.05__Embed_with_CMSIS-DAP_L43001PG-if00 -> ../../ttyACM0
lrwxrwxrwx 1 root root 13 Dec 14 15:57 usb-Texas_Instruments_XDS110__03.00.00.05__Embed_with_CMSIS-DAP_L43001PG-if03 -> ../../ttyACM1

@joshuakoh1
Copy link

joshuakoh1 commented Dec 15, 2020

@timdonovanuk What distro are you using btw? Mine was Ubuntu 20.04 and there's no /dev/serial at all
I ended up passing /dev/ttyACM0 directly from host as /dev/ttyS0 on VM.

@timdonovanuk
Copy link
Sponsor

@joshuakoh1 Debian 10. To be fair, I think if the adapter is not passed through successfully then I used to get no /dev/serial at all either. Glad you got it going!

@timdonovanuk
Copy link
Sponsor

I finally updated to the latest firmware on the adapter and it's been an unstable mess. Once a day a different bulb drops off the network, or a battery powered sensor or button has to be repaired. It's actually worse than the firmware I was using before (~6 months old). At this point I don't know whether to abandon z2m or abandon the CC1352.

@joshuakoh1
Copy link

Looking for alternatives to CC1352P-2. Really frustrated with the stupid reset button problem. Does zzh/slaesh's stick have this issue?

@timdonovanuk
Copy link
Sponsor

@joshuakoh1 FWIW throughout all of this, my logs have been showing a lot of "failed to ping ". Today I updated to the latest version 1.17 and also set availability_timeout: 0 (it used to be set to 60) and so far not a single error in my log. I found the advice about availability_timeout in another issue thread. For some people setting it to zero made z2m more stable, and for others setting it to 60 made it more stable so I don't know...but so far so good.

@silver0007
Copy link

I see that issue is closed but is it fixed anyhow?
I moved over a week ago from PRi to NUC8i3BEH which runs HassOS in VM and I struggle with same error as everyone above. My CC1352 just disconnects for no reason. Once a day, sometimes even 3 times a day... Fast option to bring back connection is to re-plug USB.
I updated firmware to the newest available for CC1532, but didn't help. Had no issues with it on RPi (worked perfect for almost a year), so clearly looks like NUC problem as discussed above.
Yesterday I changed availability_timeout to "60" from default which is "0" I believe but didn't help too. Got same error after 11 hrs of running. And in my case problem do not occur only after system reboot etc., it just keeps getting disconnected during regular runtime and re-plugging USB twice a day is really annoying... Did anyone found permanent solution?

@presslab-us
Copy link
Contributor

I had problems before where I would need to replug it after rebooting the VM. New firmware seemed to fix that problem.

Then I had problems where it would disconnect after about a week of running. That problem seems to be fixed now, my CC1352P-2 has been stable for a few months now, no problems at all. I believe the improvement was related to upgrading my qemu (now running QEMU emulator version 5.2.0 (Debian 1:5.2+dfsg-3)). I'm running on Debian Sid.

@noohi
Copy link
Sponsor

noohi commented Feb 11, 2021

Mine is also now much better, potentially fixed.

What I think helped the most is to update the debugger, basically run SmartRF Flash Programmer 2 and it should popup the recommended debugger upgrade - it's parallel to the actual app/firmware from Z2M.

@silver0007
Copy link

@noohi, so what you say is to install "SmartRF Flash Programmer 2" and then after plugging CC1352 it will recognize that upgrade is recommended and run it automatically if I confirm?
I'm asking as that sounds quite safe (safer at least than downloading firmware by myself somewhere from the web) and I want to make sure that I won't do anything that breaks my system at this point... Have maybe around 60 zigbee devices so in case of a need of re-pairing etc. it's just a nightmare.

@noohi
Copy link
Sponsor

noohi commented Feb 11, 2021

@silver0007 , yes, it will look like this, page 6 in the manual.

@silver0007
Copy link

Thanks! I’ll try if it helps.

@silver0007
Copy link

silver0007 commented Feb 16, 2021

Downloaded Flash Programmer 2 but unfortunately no pop up's about automatic upgrade.
("Success" message bar below is there because I clicked "connect" to the device)

image

Looks that CC1352 is recognized correctly. There's "Unknown" on the list probably for debugger. Not sure if there's anything that I should upload manually. Anyone did that? I'm afraid to experiment as I have so many hidden zigbee devices that I'm terrified that I'll need to re-pair them if something goes wrong...
On the other hand, re-plugging USB twice a day is also not the most convenient process ever...

@silver0007
Copy link

silver0007 commented Feb 26, 2021

@Koenkk, this issue is closed so there won't be any further investigations about the root cause and potential fix?
In my case CC1352P-2 still disconnects at least once a day even though I tried below:

  • disabled all of the energy saving features on NUC
  • flashed adapter to the newest firmware from zigbee2mqtt "Supported adapters" page (from Jan 2021)
  • pressed reset button on the board next to USB port (it seems to do nothing)
  • changed USB cable
  • played around with different settings for availability_timeout
  • checked on few different Z2M versions (now I'm on the newest stable: 1.17.1-4)

Effect always the same:
https://pastebin.pl/view/raw/cbb6140c
and error disappear basically just after I re-plug USB.

I'm thinking about changing adapter to CC2652R, but the question is: will I have to re-pair devices?

@Koenkk
Copy link
Owner

Koenkk commented Feb 26, 2021

@silver0007 I doubt if the issue can be solved from z2m, likely it's a combination of adapter/OS/computer. I don't know how we can debug this further.

I'm thinking about changing adapter to CC2652R, but the question is: will I have to re-pair devices?

No, https://www.zigbee2mqtt.io/information/FAQ.html#requires-repairing . I would recommend to go for a zzh or slaesh stick ( you will probably have the same issue with a cc2652 launchpad)

Except when switching between adapters with the following chips: CC2652, CC1352, CC253* (only when running Zigbee 3.0 firmware)

@silver0007
Copy link

Thanks for info! I'll try with Slaesh's CC2652RB and see if that helps.
Additionally I think it would be good to add some warning about CC1352 on recommended devices page:
https://www.zigbee2mqtt.io/information/supported_adapters.html
Now it's only mentioned in FAQ which everyone check when it's already too late.

@silver0007
Copy link

Just to summarize my story - I bough CC2652P stick to replace CC1352P launchpad. New adapter works without any issues for over 4 days and as said above, I didn't have to repair any devices.
Range on new stick is quite similar as on CC1352, means equally good as I've never had any range issues with CC1352 in 180m2 house.

@valepe
Copy link

valepe commented Mar 12, 2021

Just to summarize my story - I bough CC2652P stick to replace CC1352P launchpad. New adapter works without any issues for over 4 days and as said above, I didn't have to repair any devices.
Range on new stick is quite similar as on CC1352, means equally good as I've never had any range issues with CC1352 in 180m2 house.

Interesting. But which power level did you use with CC1352?

@noohi
Copy link
Sponsor

noohi commented Mar 13, 2021

Aa a side-note, which may be relevant to those with intermittent disconnects of USB on AMD systems:
Updated AGESA Coming for Intermittent USB Connectivity

@timdonovanuk
Copy link
Sponsor

My CC1352P-2 actually died the other night. I replaced it with an old CC2531 and with the normal firmware - it kept crashing. I then flashed it with the "source routing" (i.e. stable firmware used with big networks) and introduced a second CC2531 repeater. You can roll back from CC1352P-2 -> CC2531 without repairing devices (just had to delete coordinator_backup.json).

Since I've done this, my network has never been more stable. Maybe it was the introduction of the CC2531 repeater, but given the antenna size on the CC1352P-2 it shouldn't have been necessary. I will get a ZZH when I can, but I'm in no hurry to ever use the CC1352P-2 again (I appreciate technically the ZZH uses the same chip but lets see).

@jvrrr
Copy link

jvrrr commented Aug 25, 2021

I run this CC1352P-2 for more then a year and never had issues. I run zigbee2mqtt in container on Ubuntu on VMWare ESXI on a HPe Microserver 10. I still need to find a case for my CC1352...Maybe the case is giving issues in your situation)

The only time I had a issue was this:

I use Osram "smart+ plug" for a couple of lamps.
One day I decided not to use one plug anymore.
But there where some script still running in Domoticz that wanted to switch the Osram.
and I also was forget to remove te device in Zigbee2mqtt.

This situation did run well for about 20 hours and then finally the CC1352P crashes and I needed to reset it,
Day in day out, I needed to restart zigbee2mqtt and reset the CC1352P
After weeks I realisted I removed the Osram hard from the network, so I booted the Osram again and after that the CC1352p runs well for months now.

Maybe my story can help you out.

I'm realy happy with this CC1352P-2!

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