Can you add your full configurations again?
btw @<1590514584836378624:profile|AmiableSeaturtle81> , can you try to specify the host without http*
and try to set the port to 443? like s3.my _host:443
(or even without the port)
@<1523701070390366208:profile|CostlyOstrich36> Any news on this? We are currently stuck without this fix, cant finish up clearml setup
also, when uploading artifacts, I see where they are stored on the s3 bucket, but I cant find where the debug images are stored at
@<1590514584836378624:profile|AmiableSeaturtle81> ok, I think that your credentials from clearml.conf are actually working now. let's not change them.
Now let's try this simple code:
from clearml import Task
import numpy as np
if __name__ == "__main__":
task = Task.init(task_name="test4", project_name="test4", output_uri="
")
image = np.random.randint(0, 256, size=(500, 1000, 3), dtype=np.uint8)
task.upload_artifact("image", image)
You should change the task_name
and project_name
from test
just in case some object has been created previously
The problem is that clearml.conf s3 config doesnt support empty region field, even empty strings crashes it
@<1523703436166565888:profile|DeterminedCrab71> Thanks for responding
It was unclear to me that I need to set 443 also everywhere in clearml.conf
Setting s3 host urls with 443 in clearml.conf and also in web UI made it work
Im now almost at the finish line. The last thing that would be great is to fix archived task deletion.
For some reason i have error of missing S3 keys in clearml docker compose logs, the folder / files are not deleted in S3 bucket.
You can see how storage_credentials.conf looks like for me (first image). The same as for client clearml.conf (with port as you suggested)
I have the storage_credentials.conf mounted inside of async_delete as a volume
I have also confirmed that mounting works and i have the storage_credentials.conf inside of async_delete container config folder.
Maybe im misconfiguring someting?
@<1523701070390366208:profile|CostlyOstrich36> Hello, im still unable to understand how to fix this
@<1523701070390366208:profile|CostlyOstrich36> Still unable to understand what im doing wrong.
We have self hosted S3 Ceph storage server
Setting my config like this breaks task.init
Adding bucket in clearml.conf causes the same error: clearml.storage - ERROR - Failed uploading: Could not connect to the endpoint URL: " None "
I think that the problem is with missing region definition. You need to set region in the config file.
But it looks like that for the existing version it will not work since there still appears to be a bug related to this. The hotfix is already on the way from my understanding
So, in short, you need to set the region in the config file + wait for the hotfix that is pending for 1.14
@<1523701070390366208:profile|CostlyOstrich36> Hello John, we are still unable to use clearml with our self hosted s3 CEPH instances, is there any update on the hotfix for 1.14?
what about this script (replace with your creds, comment out creds in clearml.conf
for now)
from clearml import Task
from clearml.storage.helper import StorageHelper
task = Task.init("test", "test")
task.setup_aws_upload(
bucket="bucket1",
host="localhost:9000",
key="",
secret="",
profile=None,
secure=True
)
helper = StorageHelper.get("
")
Hey, i see that 1.14.2 dropped
I tried it but the issue is still there, maybe the hotfix is in next patch?
Here is the setup so you can reproduce it (we dont have region field)
clearml.conf:s3 {
use_credentials_chain: false
credentials: [
{
host: "
s3.somehost.com "
key: "XXXXXXXXXXXXXXXXXXXX"
secret: "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX"
bucket: "rnd-dev"
},
]
}
test.py
task: clearml.Task = clearml.Task.init(
project_name="project",
task_name="task",
output_uri="
None ",
)
2024-02-08 11:23:52,150 - clearml.storage - ERROR - Failed creating storage object
None Reason: Missing key and secret for S3 storage access (
None )