I've been facing this error on GCP for several days, where I get the 504 error on cloud however on the local machine it's working perfectly.
I'm using FastAPI for backend here is the endpoint :
@router.post("/create-pharmacy-user-form", status_code=status.HTTP_201_CREATED)
async def create_form_user(
db: db_dependency, # Database dependency
user_email: str = Form(...),
user_password: str = Form(...),
user_first_name: str = Form(...),
user_last_name: str = Form(...),
user_phone_number: str = Form(...),
user_city: str = Form(...),
user_country: str = Form(...),
local_name: str = Form(...),
user_newsletter_subscription: bool = Form(...),
local_image: UploadFile = File(...),
local_ice: UploadFile = File(...),
local_patent: UploadFile = File(...)
):
user_interface = UserInterface(db)
try:
# transform base 64 to file
print("Endpoint called")
user_data = {
"user_email": user_email,
"user_password": user_password,
"user_first_name": user_first_name,
"user_last_name": user_last_name,
"user_phone_number": user_phone_number,
"user_city": user_city,
"user_country": user_country,
"local_name": local_name,
"local_image": local_image,
"local_ice": local_ice,
"local_patent": local_patent,
"user_newsletter_subscription": user_newsletter_subscription,
}
db_user = user_interface.create_pharmacy_user(user_data)
print("Uploading image")
# Call the image upload function (ensure this is atomic or does not affect DB consistency)
try :
post_images(db_user["local_id"], user_data["local_image"], "pharmacy_images")
print("Image uploaded")
# Upload documents
post_document(db_user["user_id"], user_data["local_ice"], "pharmacy_ices")
post_document(db_user["user_id"], user_data["local_patent"], "pharmacy_patents")
print("Documents uploaded")
except Exception as e:
print(e)
#posting media
return api_response(
message="User created successfully",
data=db_user,
status_code=201
)
except Exception as e:
return api_response(
message="User creation failed: " + str(e),
status_code=500
)
and here are the functions:
def post_images(image_id: uuid.UUID, image_file: UploadFile, path: str):
bucket_name = settings.BUCKET_NAME
service_account_json = settings.SERVICE_ACCOUNT_JSON
try:
# Open image directly from UploadFile's file object
print("Opening image file")
# Initialize GCS client
client = storage.Client.from_service_account_json(service_account_json)
bucket = client.bucket(bucket_name)
# Generate a unique blob name
blob_name = f"{path}/{image_id}.webp"
blob = bucket.blob(blob_name)
# Upload image to GCS
blob.upload_from_file(image_file.file, content_type=image_file.content_type)
print("Image uploaded successfully")
print(blob.public_url)
return {"image_id": str(image_id), "url": blob.public_url}
except Exception as e:
print("Error uploading image:", e)
raise HTTPException(status_code=500, detail="Image upload failed.")
def post_document(document_id: uuid.UUID, document_file: UploadFile, path: str):
bucket_name = settings.BUCKET_NAME
service_account_json = settings.SERVICE_ACCOUNT_JSON
try:
# Read the file asynchronously
print("Reading Document file")
# Guess the mime type of the file
mime_type = mimetypes.guess_type(document_file.filename)[0] or "application/octet-stream"
extension = mimetypes.guess_extension(mime_type) or ".bin"
# Create a file-like object in memory
# Generate a unique file name with the correct extension
blob_name = f"{path}/{document_id}{extension}"
# Initialize GCS client
client = storage.Client.from_service_account_json(service_account_json)
bucket = client.bucket(bucket_name)
blob = bucket.blob(blob_name)
# Upload the file to GCS asynchronously
blob.upload_from_file(document_file.file, content_type=mime_type)
print("Document uploaded")
print(blob.public_url)
return True
except Exception as e:
print("Error uploading document:", e)
raise HTTPException(status_code=500, detail="Document upload failed.")
The error I'm getting on GCP cloud run is 504 : upstream request timeout
I'll appreciate really your help, since I couldn't resolve the issue. I've tried mutliple variations of code, but none has worked.
I've been facing this error on GCP for several days, where I get the 504 error on cloud however on the local machine it's working perfectly.
I'm using FastAPI for backend here is the endpoint :
@router.post("/create-pharmacy-user-form", status_code=status.HTTP_201_CREATED)
async def create_form_user(
db: db_dependency, # Database dependency
user_email: str = Form(...),
user_password: str = Form(...),
user_first_name: str = Form(...),
user_last_name: str = Form(...),
user_phone_number: str = Form(...),
user_city: str = Form(...),
user_country: str = Form(...),
local_name: str = Form(...),
user_newsletter_subscription: bool = Form(...),
local_image: UploadFile = File(...),
local_ice: UploadFile = File(...),
local_patent: UploadFile = File(...)
):
user_interface = UserInterface(db)
try:
# transform base 64 to file
print("Endpoint called")
user_data = {
"user_email": user_email,
"user_password": user_password,
"user_first_name": user_first_name,
"user_last_name": user_last_name,
"user_phone_number": user_phone_number,
"user_city": user_city,
"user_country": user_country,
"local_name": local_name,
"local_image": local_image,
"local_ice": local_ice,
"local_patent": local_patent,
"user_newsletter_subscription": user_newsletter_subscription,
}
db_user = user_interface.create_pharmacy_user(user_data)
print("Uploading image")
# Call the image upload function (ensure this is atomic or does not affect DB consistency)
try :
post_images(db_user["local_id"], user_data["local_image"], "pharmacy_images")
print("Image uploaded")
# Upload documents
post_document(db_user["user_id"], user_data["local_ice"], "pharmacy_ices")
post_document(db_user["user_id"], user_data["local_patent"], "pharmacy_patents")
print("Documents uploaded")
except Exception as e:
print(e)
#posting media
return api_response(
message="User created successfully",
data=db_user,
status_code=201
)
except Exception as e:
return api_response(
message="User creation failed: " + str(e),
status_code=500
)
and here are the functions:
def post_images(image_id: uuid.UUID, image_file: UploadFile, path: str):
bucket_name = settings.BUCKET_NAME
service_account_json = settings.SERVICE_ACCOUNT_JSON
try:
# Open image directly from UploadFile's file object
print("Opening image file")
# Initialize GCS client
client = storage.Client.from_service_account_json(service_account_json)
bucket = client.bucket(bucket_name)
# Generate a unique blob name
blob_name = f"{path}/{image_id}.webp"
blob = bucket.blob(blob_name)
# Upload image to GCS
blob.upload_from_file(image_file.file, content_type=image_file.content_type)
print("Image uploaded successfully")
print(blob.public_url)
return {"image_id": str(image_id), "url": blob.public_url}
except Exception as e:
print("Error uploading image:", e)
raise HTTPException(status_code=500, detail="Image upload failed.")
def post_document(document_id: uuid.UUID, document_file: UploadFile, path: str):
bucket_name = settings.BUCKET_NAME
service_account_json = settings.SERVICE_ACCOUNT_JSON
try:
# Read the file asynchronously
print("Reading Document file")
# Guess the mime type of the file
mime_type = mimetypes.guess_type(document_file.filename)[0] or "application/octet-stream"
extension = mimetypes.guess_extension(mime_type) or ".bin"
# Create a file-like object in memory
# Generate a unique file name with the correct extension
blob_name = f"{path}/{document_id}{extension}"
# Initialize GCS client
client = storage.Client.from_service_account_json(service_account_json)
bucket = client.bucket(bucket_name)
blob = bucket.blob(blob_name)
# Upload the file to GCS asynchronously
blob.upload_from_file(document_file.file, content_type=mime_type)
print("Document uploaded")
print(blob.public_url)
return True
except Exception as e:
print("Error uploading document:", e)
raise HTTPException(status_code=500, detail="Document upload failed.")
The error I'm getting on GCP cloud run is 504 : upstream request timeout
I'll appreciate really your help, since I couldn't resolve the issue. I've tried mutliple variations of code, but none has worked.
Share Improve this question edited 2 days ago Chris 34.3k9 gold badges99 silver badges234 bronze badges asked 2 days ago RedBlue197RedBlue197 211 silver badge1 bronze badge New contributor RedBlue197 is a new contributor to this site. Take care in asking for clarification, commenting, and answering. Check out our Code of Conduct. 2 |1 Answer
Reset to default 0The error you've encountered indicates the timeout of the upstream container. It means that the request to the server is taking too long to process, causing the timeout. One possible cause is due to large file uploads. It is also possible that Cloud Run might not have enough memory and CPU for file processing which is why it works well locally. You might need to configure it and set memory limits.
As mentioned by Chris, it might be helpful to change the default timeout value. Aside from that, you can implement logging and tracing so you'll be able to know exactly what process is taking too much time. You can read this documentation about logging and viewing logs in Cloud Run.
timeoutvalue, e.g.,blob.upload_from_file(file.file, timeout=180). Please have a look at the links above for more details and options. – Chris Commented 2 days ago