ValidationError:Stack [CDKToolKit] doesn't not exist - aws-cdk

When follow this Github case:amazon-textract-serverless-large-scale-document-processing to execute cdk bootstrap, an error message shows when "CDKToolkit:creating CloudFormation changset..." as "ValidationError: Stack [CDKToolkit] does not exist."
Used to refer to this Stack Overflow case to update policy for DescribeStack, DescribeChangeset and Createchangeset already.
{
"Version": "2012-10-17",
"Statement": [
{
"Sid": "VisualEditor0",
"Effect": "Allow",
"Action": [
"cloudformation:CreateChangeSet",
"cloudformation:DescribeChangeSet",
"cloudformation:DescribeStacks"
],
"Resource": "*"
}
]
}
Would like to know what need to do more to proceed.

Checking AWS CloudFormation console and learned the status of "CDK Toolkit" stack suspended by "In review process". Learned from this Stack Overflow case so deleted that stack and run cdk bootstrap in command line again then it worked.

Assuming you are testing in a test AWS account, can you trying giving the user account admin rights just to test and make sure it is not a permissions issues. I see you are giving account cloudformation permissions, but it does a lot more than that. It needs to create IAM roles, and create many additional resources.

Related

Electron AutoUpdater Issue

I have an issue with updating Electron app from a private repo.
I'm using https://github.com/electron-react-boilerplate/electron-react-boilerplate for web.
I've tried Hazel + Vercel approach(https://github.com/vercel/hazel) but i receive Cannot find Squirrel error.
I have target nsis and still getting Squirrel error.
"win": {
"icon": "./assets/android-chrome-512x512.png",
"target": [
"nsis"
]
},
What is the best way of doing autoupdates with private repo so it could scale and minimize bandwidth?
I'm asking because Hazel, Nuts and Nucleus are not maintained anymore and i'm having some hard time with Hazel.

iOS: universal link cannot pass the test on apple website

I upload apple-app-site-association(AASA) file on my website and trying pass validation test on Apple validator tool,
but It return me error:
http://example.com is returning 503. Please check your url and try again.
{
"applinks": {
"apps": [],
"details": [
{
"appID": "XXXXXXXXXX.com.XXXXXX.iosapp",
"paths": [ "*" ]
}
]
}
}
This file is uploaded on root folder and also .well-knows/ folders
like this -> https://www.example.com/.well-known/apple-app-site-association
503 means that your web site was not available or server returned error to test tool at the time of check. It may be due to some servers or connection errors at the time of checking. But out of the blue, you also may want to check:
does your website handles HTTPS ? AASA file needs to be accessible via HTTPS—without any redirects
is your .well-known folder accessible for web server? To test enter whole link to it in browser
check if you don't have any .htaccess or IP filtering rules that may prevent test tool access the file, even though you can access the file correctly from your location

Do 'orientation' and 'display' properties of manifest.json only work when service worker is registered?

I'm building my first progressive web app. I haven't got as far as registering a service worker just yet, but I have created a manifest.json file, which looks like this:
{
"short_name": "My PWA",
"name": "My Progressive Web Application",
"start_url": "/",
"background_color": "#222",
"display": "standalone",
"scope": "/",
"theme_color": "#222",
"orientation": "portrait",
"icons": [
{
"src": "/assets/images/logo-square-192.png",
"type": "image/png",
"sizes": "192x192"
},
{
"src": "/assets/images/logo-square-512.png",
"type": "image/png",
"sizes": "512x512"
}
]
}
I've loaded this onto a test server (not running over HTTPS currently, so I'm aware that a service worker won't work, hence I haven't got that far yet) and have looked at the site on Chrome latest on a new Nokia 6.1 running Android. I notice that the 'display' property doesn't seem to be doing what it should, and I can also turn the device to landscape and browse the site in that orientation. So it seems that the manifest.json file isn't doing anything.
I can't find any documentation stating whether the manifest.json file only becomes 'active' when a service worker is installed and/or via HTTPS so I wondered if anyone here could shed any light on this?
Many thanks.
So a bit of experimentation later... It seems the manifest.json rules only become active once the site has been added to the home-screen. (I don't have a service worker to facilitate this yet, but it can be done in the browser settings). So on Chrome/Android at least the manifest settings don't need HTTPS enabled nor a service worker.

Can iOS universal app link support multiple apps using the same domain?

Background:
Building a vanilla app for multiple clients.
Same code base with different bundle ids, i.e.:
com.company.client1
com.company.client2
Want to support all client builds with the same universal app link, i.e.:
company.com/app/path
Tried to add this to the 'apple-app-site-association' file
'apple-app-site-association' file:
{"applinks": {"apps": [],"details": [
{"paths": ["/app/*"],"appID": "XXXXXXXXXX.com.company.client1"},
{"paths": ["/app/*"],"appID": "XXXXXXXXXX.com.company.client2"}]}
Is this a limitation from apple?
This is possible. There is no limitation from Apple on Universal Links for multiple apps on the same domain.
It appears your apple-app-site-association is malformed. You need it to look like this:
{
"applinks": {
"apps": [ ],
"details": [
{
"appID": "XXXXXXXXXX.com.company.client1",
"paths": [
"/app/*"
]
},
{
"appID": "XXXXXXXXXX.com.company.client2",
"paths": [
"/app/*"
]
}
]
}
}
Note the order of the appID and paths keys, and the final closing }.
You will also run into issues with this setup if more than one app is installed, since they're all registering for the same paths. You may want to consider adding a unique ID to each one, such as /app/client1/*.
Another important note is that Universal Links don't work in many situations so this is not a complete deep linking solution (despite Apple's wishful claims to the contrary). If you want a simpler deep linking approach that will easily handle a multi-app requirement like this, take a look at Branch.io (full disclosure: I'm on the Branch team).

Permissions required to run inside Compute Engine

I have a Google Cloud Powershell script for managing snapshots that works flawlessly from my workstation but will not run inside the VM due to permissions errors like this:
PS>TerminatingError(Get-GceDisk): "Google.Apis.Requests.RequestError
Insufficient Permission [403] Errors [ Message[Insufficient
Permission] Location[ - ] Reason[insufficientPermissions]
Domain[global] ] " Get-GceDisk : Google.Apis.Requests.RequestError
Insufficient Permission [403] Errors [
Message[Insufficient Permission] Location[ - ] Reason[insufficientPermissions] Domain[global] ]
We have attempted to add my same permissions to the service account without success.
We have attempted to run the script with my Google account from the VM without success.
I think this may possibly have something to do with the Cloud API access scopes, but am having difficulty researching this online.
Can someone point me in the right direction?
Cloud API access scopes was the solution, but changing them requires a new VM. This was easiest to accomplish by drilling into a recent snapshot and clicking 'new instance' from there. When making the new instance, select the appropriate level of Cloud API access. Then it all works fine.

Resources