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

[v3] Beta #123

Closed
digitaldan opened this issue Feb 10, 2019 · 133 comments
Closed

[v3] Beta #123

digitaldan opened this issue Feb 10, 2019 · 133 comments

Comments

@digitaldan
Copy link
Collaborator

digitaldan commented Feb 10, 2019

We now have the V3 version of the sill deployed in a "Beta Test" through Amazon. This will allow us to have a small group of users upgrade early to the new skill and provide feedback.

I want to use this issue to coordinate updates and communicate to users who have access to the beta.

When opening issues against the beta, please prefix "[v3-beta]" in the issue title. Include as much information about:

What you were saying or doing, include any item names or labels that could help in debugging
What the expected result was
What the actual result was
Exact time, including seconds if possible of the issue
What time zone you are in
What country you are in (to identify which global lambda you are using)

For our first wave of users I want to focus on those using V2 style tags to ensure we have backwards compatibility and that we don't break it for the thousands using the current skill.

Configuration and usage can be found in at https://github.com/openhab/openhab-alexa/blob/v3/USAGE.md

@SamuelBrucksch
Copy link

Hi, i want to test. How can i participate?

@codeaholics
Copy link

Yes please!

@PheiPheiPhei
Copy link

I'd love to participate. I'm in Germany, if that makes any difference.

@mgeramb
Copy link
Member

mgeramb commented Feb 10, 2019

I am also very intested in joining the beta test!

@Flole998
Copy link
Member

Same here, I'd also love to join!

@digitaldan
Copy link
Collaborator Author

digitaldan commented Feb 10, 2019

Thanks everyone!!! Can you send me your amazon/Alexa email (and github username in the email) to [email protected] ?

@digitaldan
Copy link
Collaborator Author

As I mentioned in the description, would be great if people joining the Beta would test their existing V2 configuration first, and if that works, then move to using the V3 syntax and new features.

@codeaholics
Copy link

I've installed the new skill and it found all my devices just fine. I particularly like that Alexa can now query the status (in fact, this is the thing I was most looking forward to with the V3 skill). Seems to work well. I'll keep playing with it for a while.

I haven't tried the new tagging yet.

@digitaldan
Copy link
Collaborator Author

@codeaholics so you were using the v2 item tags and everything is working as expected?

@codeaholics
Copy link

codeaholics commented Feb 10, 2019 via email

@codeaholics
Copy link

Having said that... I can see that lights are on or off in the Alexa app now, but when I ask her "is [...] on?" she says "[...] doesn't support that".

@Flole998
Copy link
Member

@codeaholics Same thing here, I assume that is because the skill doesn't have retrievable set and doesn't support reportState request (yet).

@digitaldan Should we report here or open new issues for all bugs/missing features we find?

@digitaldan
Copy link
Collaborator Author

Feel free to open a bug , I think we are advertise we support reporte state , so not sure why Alexa thinks we don’t

@Flole998
Copy link
Member

Also another thing that happened to me (I don't know if this is because of the beta version though): Alexa re-discovered all things and I had to delete them from the rooms and add them again in order to say "Alexa, light on", otherwise it would just respond that the skill is currently disabled. This might be because of the beta but I am not sure and this would definitely be annoying if this happens to all users.

@codeaholics
Copy link

Yes, I had that too

@digitaldan
Copy link
Collaborator Author

FYi, I am off skiing this week, so I may be erratic in responding, timing is not great i know :-)

@SamuelBrucksch
Copy link

Looks like i did not get an invitation... its the mail of my wife because that is connected to alexa. do you get a mail or is the skill active directly?

@digitaldan
Copy link
Collaborator Author

@SamuelBrucksch did you send me a email so I can add you to the list ? I can’t seem to find it. Send me your wife’s email ( and your github username so I know who you are to ) [email protected]

@SamuelBrucksch
Copy link

@SamuelBrucksch did you send me a email so I can add you to the list ? I can’t seem to find it. Send me your wife’s email ( and your github username so I know who you are to ) [email protected]

Looks like my mail was not sent... Will do it again.

@mgeramb
Copy link
Member

mgeramb commented Feb 12, 2019

I'am running the beta version now since two days with the old tagging without any problems. It was not necessary to change anything in the configuration. The devices are remained assigned in the groups. I have Lighting and CurrentTemperature tags in use. Getting and setting values through the alexa app works fine, including to see state changed immediately in the alexa app if the device is open. Also getting the value and changing the value by voice works as expected.

@SamuelBrucksch
Copy link

SamuelBrucksch commented Feb 13, 2019

@SamuelBrucksch did you send me a email so I can add you to the list ? I can’t seem to find it. Send me your wife’s email ( and your github username so I know who you are to ) [email protected]

Hi, i got this message:

This message was created automatically by mail delivery software.

A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:

  ***@***
    (ultimately generated from [email protected])
    host gmail-smtp-in.l.google.com [108.177.15.27]
    SMTP error from remote mail server after pipelined end of data:
    550-5.7.1 [80.67.18.30      12] Our system has detected that this message is
    550-5.7.1 likely unsolicited mail. To reduce the amount of spam sent to Gmail,
    550-5.7.1 this message has been blocked. Please visit
    550-5.7.1  https://support.google.com/mail/?p=UnsolicitedMessageError
    550 5.7.1  for more information. z18si10159271wrs.315 - gsmtp
  ***@***.***
    (ultimately generated from [email protected])
    host aspmx.l.google.com [74.125.71.26]
    SMTP error from remote mail server after pipelined end of data:
    550-5.7.1 [80.67.18.30      12] Our system has detected that this message is
    550-5.7.1 likely unsolicited mail. To reduce the amount of spam sent to Gmail,
    550-5.7.1 this message has been blocked. Please visit
    550-5.7.1  https://support.google.com/mail/?p=UnsolicitedMessageError
    550 5.7.1  for more information. z77si2125991wmc.15 - gsmtp

Looks like your mail is somehow on the spamlist of my provider...Guess i need to try another mail adress then.

@digitaldan
Copy link
Collaborator Author

Hmm, google/gmail seems to be rejecting that, since my personal email is also google hosted, PM me on community.openhab.org , my username is digitaldan

@digitaldan
Copy link
Collaborator Author

So its been pretty quiet around here, not sure if thats a good thing or not :-) Let me know how the skill is working so far, if all is well then I will start advertising the beta on our community site.

@Flole998
Copy link
Member

Right now I am getting "The device doesn't respond" message when I am trying to do something using Alexa, on myopenhab.org it says Online though, not sure what is causing this.

@digitaldan
Copy link
Collaborator Author

@Flole998 can you give me the item name, time, timezone and location for one of the devices that is not responding?

@digitaldan
Copy link
Collaborator Author

@Flole998 wait, I deployed a bad config file, one sec !!!

@Flole998
Copy link
Member

Working now again, thanks!

@digitaldan
Copy link
Collaborator Author

Yep, sorry about that.

@Flole998
Copy link
Member

No problem, this is still beta so everyone knows that something like this can happen at any time :)

To get comment on this:

if all is well then I will start advertising the beta on our community site.

I think we should find out what's going on in #126 first, especially since this "spams" the karaf console (and by that also the log files) and produces high load on your servers aswell most likely (to pass through a bunch of requests per second).

@codeaholics
Copy link

Nope that's not correct, you need to enable it in the App and give it a voice code for unlocking.

Thanks for the clarification. It seems things have moved on since I last looked at lock support.

@dastrix80
Copy link

I never I did this but Alexa shows it as a lock correctly but no voice prompt.

@Naruk-TUBK
Copy link

Is it possible to participate at the v3 beta test? I currently have a mid 2018 version of the skill up and running but for me it is not possible to upgrade to the newest version (Cannot figure out the error). Would be really cool if i could use your official service and give direct feedback.

@digitaldan
Copy link
Collaborator Author

@Naruk-TUBK , see #123 (comment) , i just need some info from you.

@codeaholics
Copy link

Hello. I've got an issue with my Alexa skill. I'm currently on the V3 beta, but I can't be sure if this is specifically a V3 problem.

Often, when I issue a voice command to Alexa, it can take 5+ seconds before the OH Cloud connector picks up the command.

To be specific:

  1. I've got a routine that turns a certain set of lights on and issues a voice response
  2. I've enabled DEBUG logging for org.openhab.io.openhabcloud
  3. log:tail
  4. I run the routine from the Alexa app using the "play routine" action, so as to avoid any delays introduce by voice processing
  5. Observe that it is normally 3+ seconds and often 5 and over before running the action and the first OH Cloud connector log entry

For example, the following are the log entries when I run the routine as close as I can manually manage to 19:24:15:

19:24:19.258 [DEBUG] [.io.openhabcloud.internal.CloudClient] - on(): request
19:24:19.260 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Got request 1046723
19:24:19.261 [DEBUG] [.io.openhabcloud.internal.CloudClient] - {"content-length":"2","host":"myopenhab.org","content-type":"text/plain","user-agent":"openhab-cloud/0.0.1"}
19:24:19.261 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Request method is POST
19:24:19.262 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Jetty set header content-length = 2
19:24:19.262 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Jetty set header host = myopenhab.org
19:24:19.263 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Jetty set header content-type = text/plain
19:24:19.263 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Jetty set header user-agent = openhab-cloud/0.0.1
19:24:19.267 [INFO ] [smarthome.event.ItemCommandEvent     ] - Item 'MusicRoomLampSwitch' received command ON
19:24:19.269 [DEBUG] [.io.openhabcloud.internal.CloudClient] - Jetty finished receiving response header
19:24:19.270 [INFO ] [arthome.event.ItemStatePredictedEvent] - MusicRoomLampSwitch predicted to become ON
...
[etc. Lots more entries for the cloud connector, ZWave, etc. for each of the devices included in the routine]

@dastrix80
Copy link

Hi Guys, im using V3 with V2 tags for some proxy items and discovery is failing.

Switch Motion_PartyMode "[MAP(motion.map):%s] Proxy to Disable/Enable Motion Sensing for all Motion Sensors" (gMotionSensors) [ "Lighting" ]

Switch gAlarm "[MAP(alarm.map):%s] Proxy for switching on/off the Alarm" [ "Lighting" ]

I see nothing obvious wrong with these items though. Any thoughts?

Thanks

@jsetton
Copy link
Collaborator

jsetton commented May 7, 2019

@dastrix80 the item label on your items you listed is incorrectly formatted. The mapping definition should be at the end. Under this format, the item label is basically empty which means the relevant items will be rejected during the Alexa discovery.

Additionally, I would recommend to simplify the label description as well since there are used as friendly name when calling devices via Alexa.

Switch Motion_PartyMode "Party Mode [MAP(motion.map):%s]" (gMotionSensors) [ "Lighting" ]

Switch gAlarm "Alarm [MAP(alarm.map):%s]" [ "Lighting" ]

@dastrix80
Copy link

Can you provide an example?? or link me to some doco with examples?

@dastrix80
Copy link

Thank you! ill give that a go

@jsetton
Copy link
Collaborator

jsetton commented May 7, 2019

@codeaholics thanks for providing all the details.

Since you mentioned you are calling a routine, is the delay still present for each device part of the routine after the first action is triggered on the OH side?

Do you get the same delay when interacting with a device directly or when you trigger the same routine twice in a row?

@codeaholics
Copy link

Once the cloud connector logs the incoming POST for the first device, it normally logs the ones for the other posts within a second - still slower than I’d expect. So this could be a delay with Alexa, or it could be MyOpenHAB holding a queue internally that only gets processed periodically. I don’t know the architecture details to offer more opinion.

I would imagine it would be possible to add logging to the skill lambda to track this delay down, but you’d need to be able to correlate the log entries with my actions and for me to do a test at an agreed time for it to be useful unless you can reproduce it yourself?

@codeaholics
Copy link

To answer your other questions:

If I interact with the device in, say, Paper UI then it’s near instantaneous because it’s Z-Wave, so all processed locally.

If you’re asking about interactions via Alexa, which I assume you are, I’d need to check tomorrow. Ditto the question about doing the routine twice in quick succession.

@jsetton
Copy link
Collaborator

jsetton commented May 7, 2019

This could be a delay with Alexa, or it could be MyOpenHAB

That's what I was trying to determine. One thing for sure is that, especially with the beta skill where I will assume there isn't a lot of requests compared to the official one, the Lambda function has to be cold started if no request came in for some time. This certainly adds some delay but 5 seconds seem a bit excessive to me based on my own usage.

I would imagine it would be possible to add logging to the skill lambda to track this delay down

Your requests were logged on the Lambda side but @digitaldan would need to take a look as I don't have access to them.

@digitaldan
Copy link
Collaborator Author

All we exceeded the 90 day period for beta testing. I opened it back up, but i will need everyone's email address. You should have received an email from amazon about this, please send me a email to the address in the amazon message and i will add you back.

This is also a wakeup call that we need to get the skill published!

@Flole998
Copy link
Member

I haven't had any issues recently and as we maintain compatibility to the old version the new skill could be published even though the documentation might need some more improvements. Maybe instead of starting a new Beta it would be better to directly rollout the new version.

@digitaldan
Copy link
Collaborator Author

I agree its very nearly ready to publish, but the submission process for a smart home skill can take weeks, so i'll enable the beta in the mean time.

@digitaldan
Copy link
Collaborator Author

Sorry I have been traveling a bunch and just got back into town, and coming back to our project. I think its time to move this into production!

@jsetton I want to merge our v3 branch into master and prepare for a release. Are there any outstanding issues that need to be taken care of ? I can't remember since the last time we chatted.

@jsetton
Copy link
Collaborator

jsetton commented May 24, 2019

@digitaldan there is actually one issue #181 that @fmeies just opened and I am currently looking into it. Other than that, I think we should be good to go on the production release.

@digitaldan
Copy link
Collaborator Author

Awesome, I will wait for that to be resolved and then merge the v3 code into master. I'll start working on the release mechanics today and this weekend.

@scott-vincent
Copy link

Any news when v3 might go live. I'm hoping it's soon because my v3 beta access seems to be broken as my v3 items no longer work. My guess is I've reverted back to the live version somehow. Thanks

@scott-vincent
Copy link

Does anyone know how I switch back to the beta skill in the Alexa app. I'm pretty sure I'm just picking up the live version of the skill atm. I know it's linked to Gmail address so not sure why I can't get beta anymore.

@DerEros
Copy link

DerEros commented Jun 25, 2019

Any news on the release? It's been a month since the last lifesign. I've now set up v2 skill, but I'd love to use the new features. If v3 needs more testing, I'd sign up for beta as well.

@Flole998
Copy link
Member

@jsetton @digitaldan
Just wanted to make you aware that we are getting close to 90 days since start of the Beta again. Assuming that it's like last time (automatically ends after 90 days), we either need another round of beta, extend this one (contact Amazon Dev Support) or go live soon.

@jsetton
Copy link
Collaborator

jsetton commented Aug 14, 2019

Thanks to all the users that participated to this beta testing. The v3 skill is now live!

@jsetton jsetton closed this as completed Aug 14, 2019
@codeaholics
Copy link

codeaholics commented Aug 15, 2019 via email

@jsetton
Copy link
Collaborator

jsetton commented Aug 15, 2019

@codeaholics That's a good point. According to this post, you may need to re-enable the live skill and disable the beta one if it is still enabled on your account. I think the difference between the two is that the beta skills would show under the "dev skills" tab.

@digitaldan when you get a chance, would you be able to confirm if the beta testing ended? It seems that you may have to manually end it.

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