วีดีโอ

there are free credits if I remember but it only cost a few cents

Thanks for the support next one will be better going into more video essay direction

@NemesisT23 you need to make new folder and without temp files

@@NemesisT23runner docker in temp is folder is your blocker

@@lesliewubbel9157 I don't quite get you. Do you mean I should place my files in a folder which doesn't contain the temp folder?

​@@NemesisT23 Yes that is what I mean because when you run docker command you are taking all the files within that folder and pushing it to google cloud. Docker cannot push files that are being used or are currently open i.e. \Local\\Temp\\tmpr4m_xe5o\\file.tgz is file being used by the OS dynamically so you cannot push it. You have to redo it in a new folder lets say under documents

Hey clemence, Thank you for the offer but at this time I cannot spend money on Thumbnail Artist, I will do the thumbnails DIY style when I get more time, but I am going keep pushing it.

No problem. I appreciate your transparency. Wishing you good luck. Thankyou

Hi Sahal, Hmm that is new error I havent seen amongst other commenters. I think Google updated their policy recently? This could mean budget for the google cloud container if out of tries i.e. you have to pay google to spin up the container, it used to be like 75 cents but I don't know the problem cloud so often is pricing is confusing for customers and policy changes occur every few months to a year. Sorry I wish I could help. Warm regards, Leslie Wubbel

@lesliewubbel9157 turns out i didn't have the permissions for storage access or anything setup. Got it fixed with a little help from gemini. Now I'm facing a new error where the iam policy creation failed. Will look into it tomorrow and update. Overall your tutorial helped a lot thanks. I'm quite new to Cloud and your video helped to demystify most of the steps!

No problem!

can you print out error logs they should tell you what is wrong?

i am getting this error too , did u find any solution to this , i checked the logs but they are showing nothing

no problem! I will be honest I don't know I would just double check the files again and the blog post. its hard debug the problem in the youtube comments

Thank you for the kind hearted message I hope you find success in your journey.

Thanks! Please checkout my blog I would love feedback!

@@lesliewubbel9157 I liked your website. A recommendation would be to get the current user’s device theme and set the background color based on that

Thanks I will have to look into that

you mean change the domain name in url?

No problem

That is my wife lmao

It could be a number of reasons you could check the logs first as a clue

Truthfully I don't know, I have not been in that situation, but I would look into microservices, docker, and proxy engine nginx.org/en/ to handle high server loads it should work but there are ways to load balance. I would google load balancing

Could dump the logs here or in a paste bin? Others are getting thus error I havnt experience it but it would be a great help

Yes, this message you might get if your Dokerfile has a syntax error. In my case I write RUN exexc gunicorn instead of RUN excec and produced exactly this error message. Correcting it resolved the issue immediately.

@@planetscore Thanks for info!

could you print out your requirements.txt for me here?

@@lesliewubbel9157 yeah! beautifulsoup4==4.12.2 blinker==1.7.0 certifi==2023.7.22 charset-normalizer==3.3.2 click==8.1.7 colorama==0.4.6 comtypes==1.2.0 dnspython==2.6.1 flask==3.0.0 Flask-PyMongo==2.3.0 idna==3.4 importlib-metadata==6.8.0 itsdangerous==2.1.2 Jinja2==3.1.2 MarkupSafe==2.1.3 MouseInfo==0.1.3 Pillow==10.1.0 PyAutoGUI==0.9.54 PyGetWindow==0.0.9 pyjokes==0.6.0 pymongo==4.7.2 PyMsgBox==1.0.9 pyperclip==1.8.2 pypiwin32==223 PyRect==0.2.0 PyScreeze==0.1.30 python-dotenv==1.0.1 pyttsx3==2.90 pytweening==1.0.7 pywhatkit==5.4 pywin32==306; platform_system == "Windows" requests==2.31.0 soupsieve==2.5 SpeechRecognition==3.10.0 urllib3==2.0.7 werkzeug==3.0.1 wikipedia==1.4.0 zipp==3.17.0 When I was searching for solution on stackoverflow, GPT etc they told me to comment out the line pywin32==306 or write pywin32==306; platform_system == "Windows" since it is not compatible with linux (thats what gpt said). I did all of that but still nothing worked.

@@lesliewubbel9157 Yeah beautifulsoup4==4.12.2 blinker==1.7.0 certifi==2023.7.22 charset-normalizer==3.3.2 click==8.1.7 colorama==0.4.6 comtypes==1.2.0 dnspython==2.6.1 flask==3.0.0 Flask-PyMongo==2.3.0 idna==3.4 importlib-metadata==6.8.0 itsdangerous==2.1.2 Jinja2==3.1.2 MarkupSafe==2.1.3 MouseInfo==0.1.3 Pillow==10.1.0 PyAutoGUI==0.9.54 PyGetWindow==0.0.9 pyjokes==0.6.0 pymongo==4.7.2 PyMsgBox==1.0.9 pyperclip==1.8.2 pypiwin32==223 PyRect==0.2.0 PyScreeze==0.1.30 python-dotenv==1.0.1 pyttsx3==2.90 pytweening==1.0.7 pywhatkit==5.4 pywin32==306; platform_system == "Windows" requests==2.31.0 soupsieve==2.5 SpeechRecognition==3.10.0 urllib3==2.0.7 werkzeug==3.0.1 wikipedia==1.4.0 zipp==3.17.0 when I searched the error on stackoverflow and gpt they said it might be because pywin is not compatible with linux and mac. and server we're using from GCP is linux hence specify it in .txt so i did that. Also i tried to comment out that line but nothing worked.

@@lesliewubbel9157 i did reply but that comment is disapearing idk how!!

@vaishnavikulkarni663 you have to refresh youtube I think youtube algo is checking comments I had the same issue

I learned how to use it over time it's ok

Sure np, in terminal that is default behavior and in vscode I use github copilot. Als the code is from the blog post linked in the description

Check the logs? In Google cloud dashboard?

What does the network tab say on the browser?

Let me get back to you tomorrow

@@lesliewubbel9157 thank you for the update

​@@martinmalic Hi martin sorry for that late response I have been super busy lately I am getting my MBA right now its best to just set up a zoom call? lesliemwubbel.com/11-consulting/ 1. Best guess without seeing your code is that the template files were not copied from your local machine to the cloud instance when you deploy I would double check the the dockerignore file to see if html files are being ignore on deployment SSH into the Cloud instance to 2. Then I would check your endpoint @app.route('/upload', methods=['POST']) if your @app.route is not set to @app.route('/' ) that is your homepage on URL So I need to see code? is the template render function being called on the endpoint? Try to use paste bin? to share code with me Images of folder structure helps too Lot of different things could trigger this? - Check Server Logs - Verify Dependencies - File Permissions - Web Server Configuration - Application Errors - Network Configuration -Resource Limits

@@lesliewubbel9157 hi Leslie thanks for getting back. I've managed to get my app working by uploading it to pythonanywhere instead!

@@lesliewubbel9157 Leslie, many thanks for your response and congrats on your MBA. At the moment I have managed to host the web-app on pythonanywhere which seems to do the job just fine for my needs. Let's not waste your time at the moment, I appreciate you taking the time however. Perhaps in the future. Martin (I think I responded already, but I don't see it so here's another try)

Thank!

It shouldn't be too different just managing DNS and domain name becomes a waste of time that is why I perfer using Netlify or vercel

What is your frontend?

just react front end , wanted to learn more on how to do both front end and backend into gcp

@@jz319 I think I can figure it out for react front-end, how about a video?

That willl be great !! THANK YOU !!

Sure I can make a video on that

@@lesliewubbel9157 thank you so much

It is indeed, I answered my own question so it's all good haha

yes!

No problem

Can you double check the docker file? the last line should be CMD exec gunicorn --bind :$PORT --workers 1 --threads 8 --timeout 0 app:app also check check the yaml file was that create it should match runtime: python39 entrypoint: gunicorn -b :$PORT app:app if not I will have go through it again and make sure its all correct on my tutorial

@@lesliewubbel9157 Actually, I think I got it to work already. For some reason my docker does not recognize the environment variable (PORT) and also some of my api key (being used in the backend file that I had). So I went ahead and manually add the key (secret key) on the google cloud directly, and it works. Thanks for such a swift response. Appreciated!

@andrewchou1720 was it .environment variable I think it's missing from my blog post if that is true

I got the same issue. "...The user-provided container failed to start and listen on the port defined provided by the PORT=8080 environment variable." I have confirm: app.yaml -> entrypoint: gunicorn -b :$PORT app:app Dockerfile -> CMD exec gunicorn --bind :$PORT --workers 1 --threads 8 --timeout 0 app:app I tried changing .env to PORT=5000, but it still saying, PORT=8080 on gcloud deploy...

@PaulThach1 did you get to work? I can come Back and take look

Option 1. You have to ssh into the cloud run service container there should be some information in the google cloud dashboard heres a link I hope that helps. Then you edit the .env file but would have to restart the application for it to pick the .env file changes cloud.google.com/compute/docs/ssh-in-browser Option 2. is redeployment which is the more realistic thing DevOps does when we have to connect entire other software to your platform or new databases you will need to redeploy and do the whole build process again anyways. This is the unfortunate reality of the DevOps, they do build custom tools you build off the self to manage and automate .env files but this really out of scope for this video locally you can just edit the .env file and see the changes update locally to test stuff out first however I know you are talking about the deployment specifically here

yes because DICOM files are 3D images and their is information for (x,y,z) you can check the code here pastebin.com/MXkpnAmq or ask chatgpt for help

@@lesliewubbel9157 I will check that out. Thanks

What is your code? Can you share it? In paste bin

@@lesliewubbel9157 my mistake I pasted wrong code from that programmer's kaggle site. Thanks it works, but I have my own ct scans and I get an error "AttributeError: 'FileDataset' object has no attribute 'ImagePositionPatient'. Have to look for that issue

I inserted slices = [dcmread(path + '/' + s) for s in os.listdir(path)] print("after slices") if slices[0].ImagePositionPatient[2] not in slices[0]: print("not in slices") return slices # list of DICOM and it worked

Thanks

Yup at lesliemwubbel.com schedule a zoom call and hopefully I won't have class

Yes it's from google's docs this a default dockerfile used to make deployments its in the blog post the hyperlink should be there if its not just tell me and I will update it

Glad it helped!

Yes this happens sometimes Restarted the VScode?

Try plug in called Google Cloud Code see if that helps

did you resolve it?

@@lesliewubbel9157 no but I found a solution other than Google cloud. If I come to need Google cloud again I Will certainly do that. Thanks a lot

@@lesliewubbel9157 I had this same issue and restarting VS Code fixed the issue

No problem, I have been there done that before with the same issue its the concept is not explained in depth its important. On deployment is always issue too

Thanks man