fermyon / feedback Goto Github PK

Describe the bug
The Fermyon website and Github repos list https://discord.com/invite/AAFNfS7NGf as the Discord invite URL to use. However, it does not work; possibly expired?

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
N/A

To Reproduce
Visit URL.

Expected behavior
Valid invite code should deliver me to the Discord server's landing.

Screenshots

Additional context
Add any other context about the problem here.

In the configuration topic, https://developer.fermyon.com/spin/configuration - Request to add documentation for [component.build]. Though it is trivial, but just to get doc complete.

Pls. close this issue. It is mentioned in the link, https://developer.fermyon.com/spin/developing

Thanks.

A Fermyon Cloud user on Discord recently reported that they were receiving an SSL cert error when trying to connect to postgres RDS. To unblock the user, it was recommended that they turn off SSL validation as Fermyon Cloud doesn't support it at the time of creating this issue.

Request: add RDS certificates to Fermyon Cloud to support users who require SSL connections to run production workloads

Describe the bug
Cannot deploy basic "hello world" python app. Repo being deployed: spin-hello-py

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
spin 1.5.0 (ca08dd9 2023-09-13)
MacOS

To Reproduce
I am not sure this will be consistently reproducable. I can continue to get the same error with this spin app, but that does not mean it is a consistent error. At the very least, I would like to understand the error message. It is cryptic to a new spin user.

Expected behavior
Deploy spin-hello-py in the same manner that spin-hello-rs was deployed.

Screenshots
Pic of basic init:

spin watch worked fine & asset successfully served locally

spin build appears to work; spin deploy did not & error message is not clear to new spin user

...and rerunning spin build & spin deploy -- same failure message

Additional context
Add any other context about the problem here.

Hello,

Is it a possibility to consider to modify the Possibility to change RUST_LOG variable with sort of like SPIN_LOG for avoid misconfusions?

Hey 👋🏻

when I try to deploy a Spin application to Fermyon Cloud using spin deploy, I receive the following error message:

Uploading sample-for-the-issue version 0.1.0+r412cba...
Deploying...
Error: Unable to create app

Caused by:
    The key value service is temporarily unavailable.

Learn more at https://developer.fermyon.com/cloud/faq

For reference see the actual spin.toml:

spin_manifest_version = "1"
authors = ["Thorsten Hans <[email protected]>"]
description = ""
name = "sample-for-the-issue"
trigger = { type = "http", base = "/" }
version = "0.1.0"

[[component]]
id = "sample-for-the-issue"
source = "target/wasm32-wasi/release/sample_for_the_issue.wasm"
allowed_http_hosts = ["https://api.foo.com"]
[component.config]
some_config = "some_value"
[component.trigger]
route = "/bar"
[component.build]
command = "cargo build --target wasm32-wasi --release"
watch = ["src/**/*.rs", "Cargo.toml"]

Subsequent Deployment attempts

When I try to deploy the same app again using spin deploy, I receive the following output

Uploading sample-for-the-issue version 0.1.0+r5b15b3b0...
Deploying...
Error: No channel with app_id b65b8241-dade-44b5-a9b9-a5aa47cc5007 and name spin-deploy

Learn more at https://developer.fermyon.com/cloud/faq

Spin CLI Version

I tested with spin 1.2.1 and spin 1.1.0

im getting this error while building the goapp in spin .

i dont know the reason , but i tried on rust and it worked .

at metatable.ai we provide function deployments to our users. Users write rust code then we deploy it to Fermyon using our company account.

It'd be great to have logs for each deployed module via API, so we could show it to our users.

Hi there. I'm not sure how to run a Spin 2 app that uses variables in Fermyon Cloud. The new API spin_sdk::variables::get("var_name")? results in 500 error, with no logs. And the old API (spin_sdk::config::get()) is no longer available.

The new API works locally with Spin 2.0.0, but not in Fermyon Cloud, and it took a while to track it down, because it seems logs are no longer written when the workload panics.

Thanks for any help (totally get it's bleeding edge :-) )!

• Spin version (spin --version) spin 2.0.0 (e4bb235 2023-11-03)
• Installed plugin versions (spin plugins list --installed) cloud 0.5.1 [installed] js2wasm 0.6.1 [installed] py2wasm 0.3.2 [installed]

Describe the bug
Any TeaVM-based app currently fails immediately with a "Handler Timeout" error when run on Cloud.

Frequency of the bug
Consistent repro

Set-up information
Please share the version of Spin used when the bug occurred. You can find the version of Spin by running spin -V.
spin 0.6.0 (6b0af21 2022-11-28) (but the exact Spin version does not seem to matter)

To Reproduce
Steps to reproduce the behavior:

1. Clone https://github.com/dicej/spin-teavm-example
2. Follow the instructions to build it
3. Deploy it to Cloud using spin deploy
4. Hit the app URL using curl or a browser
5. Observe the app returns a 504 status with the text "Handler Timeout"

Expected behavior
The app should return an 200 status.

This feature would be useful for developers as they build backend jobs that need to run on a regular interval and integrate with other services such as M365

Note there is a cron trigger plugin for local Spin applications, but it is not supported on Fermyon Cloud.

Documentation topic (what would you like covered)?

Outbound IP-Address:
We have a scenario in which we need to call an external API from our service. This API is behind a firewall, which whitelists certain IPs which should have access to it.

Will it be possible to configure a static IP address, or will there be a static IP range?
Will there be additional costs, if so, how is the pricing model for that?

I was having a look at other serverless providers (Azure Functions, Cloudflare Workers), and both only allow static IPs in their (quite expensive) premium plans.

What documentation did you review and why was it not sufficient?

I had a look at several pages in the docs, for example: https://developer.fermyon.com/spin/configuration
But I couldn't find any information about outbound IPs.

Thanks in advance!

A discord user recently shared feedback that having shell support for running queries to debug their NoOps DB would be helpful for their use case.

Hi there,
I tested the new Domains feature, it is works. But something need improved.

1. How to setup a https custom domain? I found that the custom domain is http, it is not a safe one.
2. Can't see the custom domain in dashboard after add a new one, can‘t edit or remove it.

Thank you.

Hi,

When i see the logs at https://cloud.fermyon.com/ for specific app,
I believe it would be beneficial to display the logs in the latest first order, if possible.

Thank you very much!

Describe the bug
A clear and concise description of what the bug is.

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
Please share the version of Spin used when the bug occurred. You can find the version of Spin by running spin -V.

To Reproduce
Steps to reproduce the behavior:

1. Go to '...'
2. Click on '....'
3. Scroll down to '....'
4. See error

Expected behavior
A clear and concise description of what you expected to happen.

Screenshots
If applicable, add screenshots to help explain your problem.
(Remember to remove any personally-identifiable information.)

Additional context
Add any other context about the problem here.

Describe the bug
I did deploy with success a spin app in fermyon cloud, now I am using a subdomain within a fermyon app and it did work great, the problem is when I access it via http:// scheme it does not redirect to https:// scheme, it does work though when using the given fermyon domain for that app.

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
spin 1.4.1 (e0bd911 2023-07-12)

To Reproduce
Steps to reproduce the behavior:

1. Go to 'http://mysubdomain.mydomain.com' (with or without / at the end)
2. Does not load, it times out connection to https://mysubdomain.mydomain.com:1443
3. Go to 'https://mysubdomain.mydomain.com' (with or without / at the end)
4. Does work

Expected behavior
To have an automated redirect to https:// scheme at is happens with the given fermyon domain for the same app.

Screenshots
N/A

Additional context
Questions: is it a known issue? is there a well known solution? shall I implement an http listener and do a redirect on my own? thanks!

Describe the bug
Following the instructions in this video works, but if you name your project "hello-rust-2" instead of "hello-world", spin up fails.

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information

$ spin -V
spin 2.0.0 (e4bb235 2023-11-03)

To Reproduce
Steps to reproduce the behavior:

1. Follow the instructions in the above video but name your new project hello-rust-2 instead of hello-world.
2. Run spin build then spin up.

Expected behavior
It should create a localhost:3000 server and serve your new Spin app.

Screenshots

Additional context
Here is the error from spin up:

$ spin up
Error: TOML parse error at line 11, column 13
   |
11 | component = "hello-rust-2"
   |             ^^^^^^^^^^^^^^
data did not match any variant of untagged enum ComponentSpec

Describe the bug
CSS on https://cloud.fermyon.com/onboarding cuts off deploy step on Pixel 6 Android Chrome 106.0.5249.126.

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
Have a Pixel 6, load Android emulator for appropriate screen size, or use Chrome dev tools with appropriate screen size restriction (on any desktop platform).

To Reproduce
Steps to reproduce the behavior:

1. Go to https://cloud.fermyon.com/onboarding without any apps
2. Scroll to bottom of page

Expected behavior
Full page is visible.

Screenshots
Note the lack of ../spin deploy instruction at the bottom.

Should look as it does on desktop site

Note that this is even worse when using tab groups

Additional context
N/A

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

Describe the bug
Code blocks not showing properly? Website developer documentation formatting does not match what's in the Github repo for the Markdown.md file, specifically the "Accessing The Head (Front Matter) and the Body" section at the end of the document.

Frequency of the bug
Not verified. I.e. this behavior could show up on other pages.
Only noticed it on this page.

Set-up information
n/a - public website presentation issue

To Reproduce
https://developer.fermyon.com/bartholomew/markdown#accessing-the-head-front-matter-and-the-body

Expected behavior
https://github.com/fermyon/developer/blob/main/content/bartholomew/markdown.md

Screenshots
Chrome view:

Repo view:

Additional context
n/a

Some cipher suites with known weaknesses are still in use such as CBC variants vulnerable to the Lucky 13 attack. We should evaluate whether we want to support these, weighing the weaknesses against broader compatibility.

at https://cloud.fermyon.com/

I believe it would be beneficial to have git repository link at the spin app view in the cloud ui, if possible.

Tailing and/or downloading logs does not really work well in production. You want to be able to aggregate logs in external locations in an automated fashion.

CloudFlare has Logpush for this. Something similar is wanted on Fermyon.

It's a feature request.

I would like to request periodic database backups or database snapshots. It doesn't have to be fully automated, such as recovering from backups through a fancy UI, but I would like a method that I can request through Fermyon, and for Fermyon to perform database backups and recovery for any incident, disaster, or even in the event of a global catastrophe like WWIII.

I was following the tutorial here:

(https://developer.fermyon.com/cloud/develop)

After creating the hello_rust example, when I run:

spin build

I see lots of compile errors like:

error[E0425]: cannot find function, tuple struct or tuple variant `Ok` in this scope
    --> /Users/dtaylor/.cargo/registry/src/github.com-1ecc6299db9ec823/bytes-1.4.0/src/bytes_mut.rs:1242:13
     |
1242 |             Ok(())
     |             ^^ not found in this scope

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
Please share the version of Spin used when the bug occurred. You can find the version of Spin by running spin -V.

 spin -V
spin 0.7.1 (6cf7447 2022-12-19)
cargo -V
cargo 1.67.1 (8ecd4f20a 2023-01-10)
rustc -V
rustc 1.67.1 (d5a82bbd2 2023-02-07)
Mac OS X 12.6.3

To Reproduce
Steps to reproduce the behavior:

1. Follow the tutorial (https://developer.fermyon.com/cloud/develop)
2. spin new
3. cd hello_rust
4. spin build

Expected behavior
I expected not to get syntax errors when building

Screenshots
If applicable, add screenshots to help explain your problem.
(Remember to remove any personally-identifiable information.)

Additional context
Add any other context about the problem here.

Is your feature request related to a problem? Please describe.
It would be awesome if I could specify wildcard DNS records as ENV variables on Fermyon Cloud to support multiple, tenant-specific, hostnames. e.g. customer-1.yourapp.com and customer-2.yourapp.com.

Describe the solution you'd like
I'm assuming that you all are using Nomad to handle DNS atm, so I'm imagining to answer would be some type of flag for Nomad and the ability for us as users to specify our the variables for our wildcard DNS record subdomains.

Logging feature request from Discord ~

There has been requests for the ability to load ones own model for the Serverless AI feature. This could for instance be LoRA tuned version of Llama 2.

Bartholomew site doesn't start (following the quickstart guide)

I'm following https://developer.fermyon.com/bartholomew/quickstart and at https://developer.fermyon.com/bartholomew/quickstart#spin-up-your-site it doesn't work. See the error message below. Otherwise it was very smooth sailing, impressive!

Frequency of the bug
Please check one:

• Consistent repro
• Intermittent repro
• No repro

Set-up information
spin 0.8.0 (7086236 2023-01-31)

To Reproduce
Steps to reproduce the behavior:

user@NB057:/mnt/c/WINDOWS/system32/cloud-start/spinsite$ spin up --follow-all
    Preparing Wasm modules is taking a few seconds...
    
    Serving http://127.0.0.1:3000
    Available Routes:
      bartholomew: http://127.0.0.1:3000 (wildcard)
      fileserver: http://127.0.0.1:3000/static (wildcard)
    2023-02-07T19:36:04.361567Z ERROR spin_http: Error processing request: app "bartholomew-template" component "bartholomew" instantiation failed

Caused by:
    Insufficient resources: cannot create a memfd

The error message appears after following either link in the message.

Expected behavior

"When you navigate to http://localhost:3000, you should see the website running". No, instead the error above is generated.

Additional context
This is Ubuntu 20.04 running on WSL2 on W10.

Hello, it appears that the discord invitation link shown on various spots in the website has expired - also on the feedback page:
https://github.com/fermyon/feedback (see page bottom).