node-red-contrib-alexa-remote2
node-red-contrib-alexa-remote2 copied to clipboard
586837r
Error: no crsf found
since the last two versions i experience the following wired issue. Reproduceable:
- refresh cookie - (result: working between 1-3 days)
- it stops working
- initiate account - (no crsf found error)
- refresh cookie: (no crsf found error)
Workaround:
a) reinitiate and or refresh cookie a few time over again
b) restart node red
Environment: Raspberry 3 RaspMatic: 3.47.22.20191026 Node Red: 5.2.3 Alexa Remote 2: 3.9.0
I've just started experiencing this as well over the last couple of days. It didn't start after updating versions (which I did minutes after you released 3.9.0), but rather seems to have cropped up some time over the last two days.
I attempted the following code-less workarounds to no avail:
- Restarting Node-Red
- Reinitializing the account
- Setting the account to use a different file path
Not sure what's wrong here just yet, but still looking.
After some more experimenting, I was not able to circumvent this while using the "proxy" authentication mode. I was, however, able to work around it by using Cookie authentication and copying my cookie in from one of the requests I made while logged into https://alexa.amazon.com.
@zkniebel well with the cookie I experimented too. The problem with this on Raspmatic is, that the cookie never get refreshed proper. Meaning, after approx. 5 days I need a manual refresh and then, after a few more days i need to apply a new cookie manually. So the Proxy Auth was working literally better. That why it should be proxy Auth.
Maybe, I found with Raspmatic a tiny workaround working approx. 50%. after restarting node-red i get a DNS error for amazon.com for about 1-2 min, after this the initialization is working and the crsf is gone for the next 1-3 days. So is very unpredictive when the crsf error appears exactly again. @zkniebel could you test this with restarting node-red
I’m on Hassio, and confirmed that the restart-wait (gave it 30 minutes) workaround is not working for me. Odd.
Hi folks, I guess we Need to wait for @586837r providing us with an answer. I'm not fit enough reading the source Code anymore :(
Hey, there are a lot of open issues about reliability of the authentication. It's probably about the dependency "alexa-remote" or how i interact with it. Thing is i don't use these nodes anymore at the moment so i would appreciate if someone else could look into that.
Well, I would help, only my DevOps skills are a little outdated. Sure if someone wants to work together on it, please ping me.
If you want a workarround to refresh amazon auth cookie this are the steps.
- Look for a file called authFile. If you are using docker it is located at your home when you get into your instance.
- Remove it
- Restart NodeRed and make the proxy validation. It will copy a new fresh cookie from amazon in the authFile
Regards
I have not been able to get the cookie working for some time by using the appropriate URL of my raspberry pi. I tried again today and now get the csrf not found. @ljpaff will authFile be located on my PC (that I browse with) or the pi that runs node-red
@ljpaff will authFile be located on my PC (that I browse with) or the pi that runs node-red
On the pi
@ljpaff will authFile be located on my PC (that I browse with) or the pi that runs node-red
On the pi
could you give a detailed step by step for the linux/docker noobs and illiterate for how to find and delete the authfile please?
thanks in advance
I have searched the entire pi file structure (using filezilla) and cannot find authFile? I have found folders /home/pi/.node-red/node_modules/alexa-cookie2 and /home/pi/.node-red/node_modules/alexa-remote2.
Any clues please?
@grumpazoid same with me. I'm using PI3 with RaspMatic, even I told the node where to store the Auth-File but I never saw or found any. My programming skills are a little outdated, so I can't find the where to save function in the source code.
I just worked on it a little right now. Wired, I didn't restart node-red, neither I injected refresh nor Initialize - after typing in the proxy credentials, it work without crsf error. I will now observe what it behaves. I had a time, injecting refresh every 48 hrs.
Dear,
in my case you have to do a "docker exec -it <instance-name> /bin/bash" and them in the same directory making a "ls" command you see "authFile"
Then you only have to do a "docker restart <instance-nodered>"
Regards
i keep getting this as well and i can not find the authFile anywhere. i do know this is a hass.io that was installed on proxmox. there was a script that was ran. node red is installed and running on docker.
Solution:
Choose a path like /home/echo/ and create a empty file named cookie.txt in the echo folder. You end up with /home/echo/cookie.txt then set this as the File Path. It will complain about Empty Json but it will initialize and work.
@TheCookieMonster70 I was just about to try your solution. I just tried getting the cookie again by going to local address and logging in to Amazon and this time it worked !!! ?
I've now added the path as you suggested. Got cookie again. Still working. Thanks for this.
In path I did /config/alexa.txt This is on hassio though
On Sun, Mar 29, 2020, 4:36 PM ctyd [email protected] wrote:
the solution dont work for me
— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/586837r/node-red-contrib-alexa-remote2/issues/102#issuecomment-605705599, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAFWQFMRBCMJPH5MGHFCFITRJ65N5ANCNFSM4KVW5WMA .
what worked for me is just restarting node-red. lets see how long this helps.
After creating the cookie file I put the refresh to 5 days. 8 days later the nodes still say ready!!
mine will say ready sometimes but they surely are not doing what is commanded.
mine will say ready sometimes but they surely are not doing what is commanded.
It definitely seems to fail silently.
