Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Sep 03 2017 05:02
    @scottleibrand banned @Prosulpump_twitter
cameronrenwick
@cameronrenwick
@miki3421 can I jump in here? I was attempting to migrate from sandbox to atlas and had an internet crash. I'm now stuck not looping. I'm trying to learn if my rig needs to be pointed in a different direction or... my old NS site is now read only and I'm not sure what to do. Mlab said I made it to step # 10 in the migration guide... sorry for being so slow on this.. I'm usually low maintenance!
hi @cobalt027 ...I've just gone over your guide and it illustrates where I messed up...I'm going to retry... thanks for taking the time to do the guide!
cobalt027
@cobalt027
I'm happy to help if i can. Also, I did not write the guide. I used it myself and just reposted the link. Someone smarter than I wrote it. That person should stand up and take credit for all their hard work on that!
@solomonT1 from what app are you getting the error? I'm confused as to where your problem lies. perhaps a screenshot to illustrate?
nicoleloop
@nicoleloop
Hmm. I have no idea how to test connectivity. I tried. Downloaded mongo shell on my computer, extracted it, opened it... And a small window appeared briefly on the screen, then it disappeared and I never figured out what it was, where it went, or how to use it once I did happen to locate it. Am I the only noob who feels a little bit abandoned by the devs here? I have yet to even receive any notifications about Any Of This, from Anyone. Thank goodness I just happened to wander onto mlab yesterday...
Jon
@jonprogers
@nicoleloop I guess you're at section D step 7 of these instructions - https://docs.mlab.com/how-to-migrate-sandbox-heroku-addons-to-atlas/?fbclid=IwAR1hhkMok4XqFdu6SXi1uJMhjycO39nwNzrJal-0OZvMSAXseFqoZOPuwlI
If so mongo shell runs in a shell - windows command prompt or powershell, linux terminal etc.
I then ran the following command in the shell (windows powershell) .\mongosh.exe "mongodb+srv://<username>:<password>@<node>/<database>?retryWrites=true&w=majority" - the bit in quotes is the connection string provided in your migration process
you can then enter something like the following commands to check you have a connection
ent = db.getCollection("entries")
ent.countDocuments()
this should return the number of documents in your blood glucose collection and shows you have connectivity...
good luck...
cameronrenwick
@cameronrenwick
anyone have any advice? I did my migration but now get a NS boot error that looks like this
Screen Shot 2020-08-16 at 8.38.50 PM.png
cameronrenwick
@cameronrenwick
I'm convinced that my MONGO_CONNECTION string is incorrect. It should be formatted as: mongodb+srv://<username>:<passord>@<databasename>.sgneg.mongodb.net/<databasename>?retryWrites=true&w=majority correct?
Jon
@jonprogers
@cameronrenwick I think it should be .....//<username>:<password>@<CLUSTERNAME>.sgneg........
@cameronrenwick depending on the instructions you used did you do the connection test (section D step 7 in the ones I followed) this will prove your atlas connection string...
cameronrenwick
@cameronrenwick
thanks @jonprogers ... I did not do the connection test as I used the shortened instruction list... (hindsight is always 20-20) is there a way to confirm cluster name? It appears to be the same as my database name, which may be the problem but I'm just not sure...
cameronrenwick
@cameronrenwick
is there a way to go back to the connection test and running it (after you've migrated?)
Jon
@jonprogers
@cameronrenwick cluster name is displayed on the 'Clusters' screen, there was nothing I can remember to stop the cluster being named the same as the db so you could be fine, I just try never to give two things the same name as I'm easily confused!!
It's OK to do the connection test with mongo shell at any time - I've just rerun mine to check in exactly the same way as I did before.
You can get confirmation of your connection string by clicking the connect button just below the cluster name on the Clusters screen...
cameronrenwick
@cameronrenwick
I've updated the string using the clustername after the @ but still no joy.
Jon
@jonprogers
@cameronrenwick have you confirmed your entire connection string and run the connection test?
@cameronrenwick mine has a node of <5 random chars>.azure.mongodb.net - the one you posted is missing the azure but you may have selected a different hosting option to me...
kskandispersonal
@kskandispersonal
@cameronrenwick See my comment above re: MONGO_CONNECTION ... You might be missing quotes around the connection string in the the mongo command.
cameronrenwick
@cameronrenwick
hey everyone.. so I've got nightscout working now but just need to get CGM to NS. I'm using an xdrip and android. Do I need to change my base URL in xdrip or how does CGM get to NS now?
ArthurusDent
@ArthurusDent
@cameronrenwick Everything you need should be in here: https://androidaps.readthedocs.io/en/latest/EN/Configuration/xdrip.html
cameronrenwick
@cameronrenwick
thanks @ArthurusDent .. that's what I needed! I was using an old set of directions.... many thx!
nicoleloop
@nicoleloop
@jonprogers Thanks! That was very helpful! 😊
nicoleloop
@nicoleloop
The wonderful team at mlab made this specialized migration guide specifically for Nightscout users! I asked them if I could share it here so others would also benefit from the rather EXTENSIVE help that I required to get my site back up and running again! LOL.
https://docs.mlab.com/how-to-migrate-nightscout-sandbox-heroku-addons-to-atlas/
gitanO0
@gitanO0
hello all. I've got a new anroid phone and have installed xdrip+ on it and am trying to start an existing sensor/transmitter on the new phone, transitioning from/replacing the old phone. It seems I've done everything right... I've stopped the sensor on the old phone, waited for the system status to say "stopped"... then I disabled the collection on the old phone and turned off bluetooth. I imported my settings into the new phone xdrip+, reinserted the sensor and started a new sensor. Everything seems to be working, I'm getting data, etc... and it says "warming up", which is the big success indicator, however, I am seeing a button below the "warming up" message that says, "Start Source Setup Wizard". That's making me nervous. If I click that, nothing happens. How do I make that go away and just let the app/sensor "warm up" as normal?
ArthurusDent
@ArthurusDent
@gitanO0 Try doing a long press on the blood drop at the top. Check if the wizard is still active.
gitanO0
@gitanO0
@ArthurusDent I just checked... the warmup is done and that button has gone away... so it looks like all's good. Thanks again.
Nick Tiliakos
@ntiliakos_twitter
Apart from the migrating guide which is super useful is there a guide for setting up Nightscout with Atlas in the first place?
Amr Haggag
@aimohaggag
Hi , i have a problem with api secret when login to my nightscout site , it gives wrong api
i have moved to atlas, ir was working fine before
Guy Davies
@guydavies
@aimohaggag the API secret didn't (necessarily) get touched during the update/migration. Check again on your Heroku config site and maybe update the API secret to a new known value and retest.
claresloggett
@claresloggett
Thanks @nicoleloop !
claresloggett
@claresloggett
I wonder if anyone can help me troubleshoot... I have a Nightscout problem that seems to have occurred spontaneously: I am not seeing any BGL readings on the graph, and <my-url>/api/v1/entries.json returns []. I do see treatments (carbs etc) and I do still see data at <my-url>/api/v1/treatments.json . I'm also getting an xDrip error saying that it can't add values to the database. But I'm not sure what's causing the xDrip error, and it doesn't just seem to be a data upload problem - old BGL entries have vanished too
The xDrip+ uploader status error is "Unable to do REST API Upload: Internal Server Error url: <my-url>/api/v1/ marking record: failed"
A bit of googling found a similar error in a case where the mongodb ran out of space (I think?) but again this doesn't explain why old entries have disappeared
I should probably look at NightScout logs?... how do I find those?
Foxy7
@Foxy7
@claresloggett the Nightscout help has migrated over to discord. https://discord.gg/zg7CvCQ best to ask on there now.
claresloggett
@claresloggett
Thanks @Foxy7
Sorry, I had heard, forgot!
Foxy7
@Foxy7
:thumbsup:
Guy Davies
@guydavies
@sulkaharo , 14.0.3 running happily receiving data from AAPS 2.7.0-rc7 + DanaRSv3. I like some of the new report styling. Subtly different and definitely better. Thanks for everything.
Chris
@Tchibola_twitter
Hi, I am trying to setup a local Nightscout instance on a raspberry pi. I have installed node and npm and also installed cgm-remote-monitor. When I try to start with pm2 (env $(cat my.env) PORT=1337 pm2 start server.js), I am getting the following error: "env: ‘careportal’: No such file or directory". In my.env, I have the line "ENABLE=careportal%20openaps%20iob%20bwp%20cage%20basal%20pump%20bridge". Am I doing something wrong with the ENABLE variable definition ?
Guy Davies
@guydavies
I have just noticed that the Automation flows I use to automatically record walks and gym sessions are not uploading the data to NS. When I checked a previously working request (POST) to my NS, I get the response
{
"status": 401,
"message": "Unauthorized",
"description": "Invalid/Missing"
}
It doesn't really tell me what is Invalid/Missing but the API_KEY is correctly converted into a SHA1 string and it was working pre-14
@sulkaharo , any suggestions what I may have missed in the git logs. I couldn't find anything suggesting that the authentication had changed, looking back through the tagged releases
Guy Davies
@guydavies
It appears, from a little testing with Postman, that it works if I put a valid token on the end of the URL, but not any more if I put the API-SECRET converted to SHA1 into the HEADERS.
I can use that approach but it's interesting that it's changed. If I've missed it in a release note then apologies.
Guy Davies
@guydavies
Hmm, but interestingly, it only works if the token has admin rights. I've tried api:treatments:create, api:activities:create, both of which are system defaults, but only admin works.
Sulka Haro
@sulkaharo
Right, that should not have changed. Need to check what’s up
Guy Davies
@guydavies
Thanks @sulkaharo. Let me know if you need anything from me. Happy to help.