appwrite: π Bug Report: Functions not being executed
π Reproduction steps
Using Appwrite v:0.13.2.302 in docker And running Appwrite cli v0.15.0
- Create function with (node-16.0 runtime):
appwrite init function
- Deploy function without editing it
appwrite deploy function
- Open the created function in the function tab of the Appwrite UI
- Click on Execute Now without any custom data
- Go to Logs tab
- Click on the Errors button at the right of the executionβs list item
Then you get the below described error.
π Expected behavior
The function should run
π Actual Behavior
I get this error : " An internal curl error has occurred within the executor! Error Msg: Could not resolve host: 622bb6f3725e4422610e-622d226c1a20e10fc330 "
When taking a closer look the host is comprised of 2 parts :
- 622bb6f3725e4422610e: The project id
- 622d226c1a20e10fc330: The function deployment id
Both of the parameters seemed to coincide with the ones, visible on the UI.
π² Appwrite version
Version 0.13.x
π» Operating system
MacOS
𧱠Your Environment
I use Docker
π Have you spent some time to check if this issue has been raised before?
- I checked and didnβt find similar issue
π’ Have you read the Code of Conduct?
- I have read the Code of Conduct
About this issue
- Original URL
- State: closed
- Created 2 years ago
- Reactions: 3
- Comments: 56 (18 by maintainers)
Iβd be very happy to join to Discord, but the problem is that it is no searchable by Google; otherwise Iβd find an answer already. Documenting it here so that anyone having the same issue with Appwrite + Java can easily find, instead of creating duplicate bug reports:
The issue still exists, Appwrite v.1.0.1,
Same here.
I am also facing the same issue since few months now. How to solve this error, it is such a critical bug? Can someone please help
For anyone who have this kind error as TC:
In
docker-compose.yml
declared networks doesnβt have specific names and when you up your containers from docker-compose it creates networks with template as"directory-name-of-compose-location"_"network-in-compose-file"
.So, in
.env
described network is hardcoded as_APP_EXECUTOR_RUNTIME_NETWORK=appwrite_runtimes
which only work if yourdocker-compose.yml
file located inappwrite
directory.To avoid that do the next:
docker-compose.yml
like this (be aware that they must be unique because of docker-host scope):"project_id"-"deployment_id"
. Check withdocker inspect "container name"
that inNetworks
section there is a network with your defined name.Same here.
Has anyone solved this yet? I am still facing same issue with python function. Also there is no clear documenttaion or youtube video about this. may ber the code is broker in appwrite.
having the same issue as the users above
Noticed that this issue has been closed, but Iβm still having this issue with manual deployment.
I followed the tutorial on how to setup a simple function and deploy it through appwrite, but it fails during execution.
I have setup appwrite through docker on macOS, with the latest version (
1.1.2
).I use this in CLI and it didnβt work. I basically restart docker every day to minimize the error. No other solution yet π
"$id": "......", "name": "orderMatchingAlgorithmFunction4", "runtime": "dart-2.17", "path": "functions/orderMatchingAlgorithmFunction4", "entrypoint": "lib/main.dart", "ignore": [".packages", ".dart_tool"], "execute": ["role:member"], "events": [], "schedule": "", "timeout": 15
Same. (1.3.1)
We found a solution for our case, hope works for you
appwrite functions createDeployment --functionId=YOUR_FID --entrypoint='src/index.php' --code="functions/Functionname" --activate=true
Previously we got
Internal Runtime error.
When we just copied@LeonardoMantovani, please open a GitHub Q&A discussion or join us on Discord.
thanks i recreate installation as default and adapt my others containers to it⦠the function has been executed but docker logs appwrite-executor said;
I tried using the CLI to deploy the same function instead of choosing the manual deployment (without changing anything else) and everything seems to work. There is probably something wrong with the manual deployment.