Is there a way you can allow our account to make a copy of the AMI and store it privately?

Is it possible to use an IAM role rather than user credentials in the clearml.conf file?

Thanks. Where is the configuration stored on the server? Currently we have deployed an EC2 instance using the marketplace AMI - if this demo is successful we would be looking at splitting the environment into the different AWS services - logging to S3, use of secrets manager, elasticsearch, redis, mongoDB etc

I've looked through the documentation, but didn't initially spot anything that would help with doing this (granted I may have overlooked something)

Hi @<1523701205467926528:profile|AgitatedDove14> I tried this out, but I keep getting connection timeouts in the browser getting to the ELB. The instance is showing as inservice and passing the healthcheck. Is there any other configuration I need to do in the clearml.conf to make this work?

yes

Yep i've done all that, it didn't seem to work until I set the deploy key to write

thanks, i'll try that out

our setup currently consists of an EC2 instance for clearml-server and one for clearml-agent. We're not using a load balancer at the moment.

• What are the credentials that are referred to in the logs? Where do I get these?
• How do I ensure this container is run automatically if it keeps restarting using the docker-compose file?
• If I run sudo docker run -it allegroai/clearml-agent-services:latest manually, do I need to put this in a crontab? My knowledge of running docker containers is a bit limited.
• Does this service need to run if I already have a clearml-agent running on a separate instance?

thank you

Host key verification failed.
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.

Just by chance I set the SSH deploy keys to write access and now we're able to clone the repo. Why would the SSH key need write access to the repo to be able to clone?

Thanks. Although it's AWS related, the context was with an error we see within clearml "ValueError: Insufficient permissions for None "

have 2 listeners setup. LB 80 > instance 8080 and LB 443 > instance 8080

no, that's what i'm trying to do

ok i'll try that out thanks

Thank you very much 🙂 I don't think our Data team ever use this container so I will stop it for now and comment it from the compose file

that should be the case, we have default_output_uri: set to an s3 bucket

Hi @<1523701205467926528:profile|AgitatedDove14>
Yes the clearml-server AMI - we want to be able to back it up and encrypt it on our account

all sorted, I somehow missed the documentation about the mongodb migration

Some ideas, not all directly related to this:

• Passwords should be encrypted before being stored
• A mechanism on the server application to add/remove users, avoiding having to SSH on to the server would be nice
• Some level of permissions in the application would be nice - Admin/Owner/Viewer restrictions which would dictate would users can do and give finer control

or have I got this wrong, and it's the clearml-agent that needs to read/write to S3?

Hey @<1523701205467926528:profile|AgitatedDove14> I am helping Max to get this working. I ran the clearml-agent init and now have the correct entries in the clearml.conf file.

Created an ssh key from the agent, uploaded it to the git repo, but still getting this error:

Host key verification failed.
fatal: Could not read from remote repository.
Please make sure you have the correct access rights
and the repository exists.

If I manually ssh on to the agent, and run:
`git clon...

is there any documentation for connecting to an S3 bucket?

I added this to each of the containers

    logging:
      options:
	    max-file: 5
        max-size: 10m

yes those have all been created

After making the change yesterday to the docker-compose file, the server is completely unusable - this is all I see for the /dashboard screen

yep still referring to the S3 credentials, somewhat familiar with boto and IAM roles

yep, in most of them:

/opt/clearml/config
apiserver.conf
clearml.conf

/opt/clearml/data/elastic_7
/nodes

/opt/clearml/data/fileserver
<empty>

/opt/clearml/data/mongo/configdb
<empty>

/opt/clearml/data/mongo/db
collection/index files, /diagnostic.data, /journal etc

/opt/clearml/data/redis 
dump.rdb

/opt/clearml/logs
apiserver.log.x, filserver.log (0 bytes)

so am I right in thinking it's just the mount points that are missing?based on the output of df above