NWS Stopped Providing day/night forecast - it reports unavailable

[kirbyschapman]

The problem

The NWS integration with METAR code kmsn stopped providing the day/night forecast yesterday. It reports unavailable. I deleted and reinstalled the integration, and the problem persisted. I then deleted the integration and reinstalled with a different METAR code (kord), and the problem still persisted.

What version of Home Assistant Core has the issue?

2023.12.1

What was the last working version of Home Assistant Core?

2023.12.1

What type of installation are you running?

Home Assistant OS

Integration causing the issue

NWS

Link to integration documentation on our website

https://www.home-assistant.io/integrations/nws/

Diagnostics information

hass_log_of_nws_debug_data.log

Example YAML snippet

NA

Anything in the logs that might be useful for us?

Debug data is above - that's the only thing in the logs

Additional information

It became unavailable at 1:56 12/22/2023 CST

[home-assistant[bot]]

Hey there @matthewflamm, @kamiyo, mind taking a look at this issue as it has been labeled with an integration (nws) you are listed as a code owner for? Thanks!

Code owner commands

Code owners of nws can trigger bot actions by commenting:

• @home-assistant close Closes the issue.
• @home-assistant rename Awesome new title Renames the issue.
• @home-assistant reopen Reopen the issue.
• @home-assistant unassign nws Removes the current integration label and assignees on the issue, add the integration domain after the command.
• @home-assistant add-label needs-more-information Add a label (needs-more-information, problem in dependency, problem in custom component) to the issue.
• @home-assistant remove-label needs-more-information Remove a label (needs-more-information, problem in dependency, problem in custom component) on the issue.

_{^{(message by CodeOwnersMention)}}

---

nws documentation nws source _{^{(message by IssueLinks)}}

[Peaster111]

Exact same issue here

[slickbackwood]

The issue is on the NWS side. There NGINX server that hosts the API is not properly responding. If you go to the NWS documentation on their API and manually use it to pull API calls through the browser. When you try and pull the JSON info for the different forecasts you get a NGINX error.

[Peaster111]

Thanks.. We'll all hang tight and hope NWS gets it up and running again. Appreicate the quick reply and looking into it.

[voretaq7]

FY’sI - This is a known outage, NWS plans to have it back up on January 1st.

cf. https://www.nco.ncep.noaa.gov/status/messages/

SENIOR DUTY METEOROLOGIST NWS ALERT ADMINISTRATIVE MESSAGE NWS NCEP CENTRAL OPERATIONS COLLEGE PARK MD 1837Z FRI DEC 22 2023

...API FORECAST GRIDPOINT OUTAGE UNTIL 1 JAN 2024...

Due to a known software issue, the API gridpoint forecast service will be unavailable until Monday 1 January 2024.

Kempisty/SDM/NCO/NCEP

If it’s still broken after that we can check back for an update.

[Peaster111]

That explains it for sure... hold tight until Jan 1 then. You sir are on top of it... Well done.

[ApricotFarm]

January 1st??? BUT SANTA NEEDS THE WEATHER FORECAST!!!!

[gumby2]

Your tax dollars not so hard at work☹️

[cbhiii]

I've seen this over the past year and from what I can tell it won't be fixed until a day or two has passed with "normal business hours". It must only be checked on during the day by someone (nice that they have a bulletin for this outage). I personally have also been pulling data from the AccuWeather integration as well and when (not if) the NWS servers act up I use a template to supplement the missing NWS data (see HA community if not sure how to do that).

[nwithan8]

I personally have also been pulling data from the AccuWeather integration as well and when (not if) the NWS servers act up I use a template to supplement the missing NWS data.

I was planning on doing that myself with OpenWeatherMap, especially with the recent changes to how often weather is polled in HA. Mind sharing your template with the fallback technique?

[cbhiii]

Here is a link to the post I just made about suggestions for going about using a backup weather service for those interested.

[drewkeller]

My NWS forecast stopped working at the same time I updated to HA on Raspberry Pi 5, so I thought it was something related to that and pinging their server too often. LOL. I'm guessing that whoever is going to have this done on "Jan 1" is not actually going to be working on New Years Day. So maybe it will go live on Jan 1 or it will not actually happen until "some day" after that.

[voretaq7]

It appears NWS has someone in the IT shop working for the holiday, forecasts are coming through again (though for some reason not showing on the card for me, but I'll go fight with that later).

[darthsebulba04]

That's interesting. I see the forecast, but nothing else. The current conditions (temp, humidity, etc) stopped working mid-last week as well as the forecast the week before that.

[robertgetch]

I think waiting for a business day or two to pass might work, it does seem to have partially recovered (future forecasts) but current conditions are 'unknown'. Interestingly I would have thought current conditions would have appeared first given I doubt those are all reviewed individually 24/7.

[voretaq7]

Well, reloading the integration sort-of fixed the card - forecasts are back but now current temp/humidity are unknown (I don't see them in the JSON data from KJFK or KFRG - the data there is stale from 2023-12-27).

So apparently something is still broken and so don't see that in the known issues from NWS. (ETA: I’m getting current METAR data from other NWS systems, so this is clearly something fried in the API, I can’t imagine they’re not aware of it by now but I’ve sent a report in to the NCO team.)

[NJBILLT]

Same issue noticed over the holiday but I figured someone was on vacation and patiently waited. Now, not only are day/night forecasts not updating, I don't even get the current conditions. Everything is unavailable. This is true for 2 different instances of HA running in 2 different houses requesting meta data from 2 different stations. KPHL and KPGD.

[ski522]

Seems to be working fine for me as of this post. It must have started working correctly this afternoon as I was still having some trouble this morning.

[robertgetch]

I can also confirm as of 12:23 PT it has started working again as expected.

[NicolasAmbrose]

Looks like everything is back for me as well, KIWA, 2:26 MST. Happy camper, thanks for all the work/updates.

[kirbyschapman]

Thanks, everyone - it's working for me, too, now.

[NJBILLT]

Somebody was definitely on vacation. Works for me too now.

[voretaq7]

Yeah, apparently the observations going out was a backend routing issue with the API servers, I got the "Shit. We fixed it." email around 4PM Eastern & everything seems to be running normally now.

[CaptTom9]

I'm still getting no current conditions here. The forecast has been working for a couple of days now.

Is it just me?

[grischard]

@CaptTom9 yes, just you. Try restarting?

[Joeslinky]

I'm still getting no current conditions here. The forecast has been working for a couple of days now.

Is it just me?

@CaptTom9 Not just you!! I have this same exact issue.

[CaptTom9]

Thanks @Joeslinky. Good to know it wasn't just me.

Here's what I found. I had a bunch of templates to extract various values from the attributes, like this: state: "{{ state_attr('weather.kpwm_hourly', 'temperature') }}"

For whatever reason, that stopped working when I updated to 2023.12, which happened to be right around the time the NWS outage began. After seeing here that it was supposed to be working again, I dug deeper. It turns out the values I'm after, like temperature, are now defined as sensor entities, as in: sensor.kpwm_temperature

I commented out my templates and started using the entities instead, and all is well now.

My guess is there was a breaking change at some point which I'd missed.

[issue-triage-workflows[bot]]

There hasn't been any activity on this issue recently. Due to the high number of incoming GitHub notifications, we have to clean some of the old issues, as many of them have already been resolved with the latest updates. Please make sure to update to the latest Home Assistant version and check if that solves the issue. Let us know if that works for you by adding a comment 👍 This issue has now been marked as stale and will be closed if no further activity occurs. Thank you for your contributions.

[NJBILLT]

This topic is safe to close. Resolved.

[MatthewFlamm]

Looks like this one was confirmed to be resolved by the author and several others, closing it.