I am sorry that you are continuing to encounter issues! We have put together a cURL command that should work directly with the API - I have tested this myself so I am fairly confident this should work.
API_ADDRESS with your api address (e.g. api.lamp.digital)
ADMIN:PASSWORD with you admin credentials
lamp.accelerometer for each new appropriate spec name
Unfortunately, you will have to run this once each for each sensor spec you want to initialize, so it is clunkier than the other way. We will also work on a script to automate this process, but if there are specific sensors you need asap you should be able to use this command.
I hope that this is helpful - I will get that automated script to you as soon as possible but I know with the holiday approaching you may need this intermediate script now.
Thanks for this. Unfortunately I receive the same â503.unimplementedâ error using the provided curl command as I did when running the python script you provided earlier:
The digest matches whatâs on docker hub for this image:
$ docker images --digests bidmcdigitalpsychiatry/lamp-server
REPOSITORY TAG DIGEST IMAGE ID CREATED SIZE
bidmcdigitalpsychiatry/lamp-server 2021 sha256:d29f8a21b4ce8579b5d1538e5010755b7fb671f4e7f674de0e1c78d2ef56e334 e21ac6c5057e 7 months ago 281MB
Pulled the image again:
$ docker image pull bidmcdigitalpsychiatry/lamp-server:2021
2021: Pulling from bidmcdigitalpsychiatry/lamp-server
Digest: sha256:d29f8a21b4ce8579b5d1538e5010755b7fb671f4e7f674de0e1c78d2ef56e334
Status: Image is up to date for bidmcdigitalpsychiatry/lamp-server:2021
docker.io/bidmcdigitalpsychiatry/lamp-server:2021
$ docker image ls
REPOSITORY TAG IMAGE ID CREATED SIZE
traefik <none> daa4d30f224b 2 months ago 96.8MB
bidmcdigitalpsychiatry/lamp-server 2021 e21ac6c5057e 7 months ago 281MB
apache/couchdb 3.1.1 cabe93489203 8 months ago 189MB
nats 2.1.9-alpine3.12 7b1589771fc3 9 months ago 16.2MB
redis 6.0.8-alpine 8835e6aeca99 13 months ago 31.1MB
Updated the service with --force
$ docker service ls
ID NAME MODE REPLICAS IMAGE PORTS
8dcjztei74bn lamp_cache replicated 1/1 redis:6.0.8-alpine
ria1usfe8a3s lamp_database replicated 1/1 apache/couchdb:3.1.1
r3ywbi9mdkzh lamp_message_queue replicated 1/1 nats:2.1.9-alpine3.12
t4bibc1ko5x5 lamp_server replicated 1/1 bidmcdigitalpsychiatry/lamp-server:2021
z6b0dkjuifv5 router_traefik replicated 1/1 traefik:latest *:80->80/tcp, *:443->443/tcp
$ docker service update lamp_server --force
lamp_server
overall progress: 1 out of 1 tasks
1/1: running
verify: Service converged
That is very helpful - I think you have found the main issue here! The image from docker hub is actually a bit out of date and has not been updated recently - the change happened just before I arrived but for a few reasons we switched to hosting the image on github directly, at ghcr.io/bidmcdigitalpsychiatry/lamp-server:2021 (ghcr.io/bidmcdigitalpsychiatry/lamp-server:2021)
That explains both of the issues - why we werenât able to use this new feature and why attempting to update the image didnât lead anywhere. If you replace the image you are using for your lamp_server service with this new one, I think both of the scripts (cURL or Python) will start working.
@khackett , I hope this helps get us past the sensor issues you encountered. Please test when you can and let me know if things are working as expected now.
Hi @lido - it appears this worked! I am now seeing the ability to add all the individual sensors on the researcher/dashboard end. I will go ahead and add them all to our study, do some testing and see what data I can pull on the backend afterwards. Thanks to everyone on this thread for helping out â have a wonderful Thanksgiving!
Hi again @lido@LukeS@spatel - after about a week of pilot testing on a few phones, it appears that the following sensors are not being collected on either iPhone 12 Pro (software version 15.1.1) or Android (google pixel) devices: accelerometer.motion; bluetooth; calls; device_state; distance; flights; gps.contextual; gyroscope; magnetometer; segment; sleep; sms; wifi [these were all sensors that @lido configured per the screenshot above]. Interestingly, lamp.steps and lamp.telephony were collected on the iPhone but not the Android device. Is this expected/is there anything you would recommend to ensure we have at a minimum the calls/sms data collected across both types of devices? Iâm also puzzled why the lamp.telephony sensor is collecting data for the iPhone even though it wasnât technically configured/loaded.
Just as an FYI the following sensors are being successfully collected on the iPhone: accelerometer; analytics; gps; screen_state; steps; telephony
And on the Android: accelerometer; analytics; gps; screen_state
Thanks for letting us know about this. Some of the sensors that you mentioned that are not collecting are deprecated or require further configuration and not used actively, so it makes sense they may not have data. These include accelerometer.motion, device_state, flights, gps.contextual, distance and segment. I will look into why the remainder of your sensors are not being collected with @LukeS.
lamp.telephony should have data on calls/sms, but let me know if that is not the case. It is a bit unique and can be collected from the phone despite being enabled.
Lastly, there are some minor differences in sensor collection/quality between Android and iPhone devices since sometimes the Android build is behind the iPhone, but these should be fixed soon to account for having the same sensor collection. Let me know if this helps and feel free to reach out with further questions!
Thanks for the quick reply @spatel! Please keep us posted on why the remainder of the sensors are not being collected and on the timeline for Android having the same sensors (we will likely start recruiting just iPhone users to get around this for now).
To answer your question, it looks like lamp.telephony is just collecting phone calls (not sms/texts) per the attached screenshot.
Hi @spatel@LukeS - just following up on the missing text/sms data for iphones. Right now this appears to be the one missing piece for us before we start recruiting iPhone users. Neither lamp.telephony nor lamp.sms are generating text data even though the study is configured to include these sensors. Are there any other sensors we should be trying, or anything in the phone settings that need to be manually configured (other than allowing all permissions after downloading mindLAMP as directed in the docs)? Thanks again for your help on this!
We have checked on this and have found that iOS is not setup to support SMS logs and this issue may be occurring because though Android does support SMS logs, this may not be happening lately due to the OS version and the phone type.
