yeah i tried.. but its needed access key & secret ley

okay cool.. i will try that

I could not find any documentation for this support

yes.. if from clearml side.. there is no option then its fine.. i want this option .. to play with public ip

i hope you are saying this method.. i tried but i dont see any changes in UI, how can i verify sir

okay understand, thanks

same node multiple GPU working but.. diffrent nodes GPU is not balancing the workload

i hope u understand my concern.. i want to install clearml server & cleaerml agen together.. same script..

@<1523704207914307584:profile|ObedientToad56> already i added

this will work but i need clearml_basic_token or secret key & accesss key

this localhost i need it as ip .. like example.. aws public ip.. how can i do that..?? now this is not accessible.. i mean jupyter ..

if u say clearml have good security.. i dont agree.. i dont see any username & password authentication.. its open server

am reading it

but missing part now.. queue.. this one even not there in doc..little worried now 🥶

but i need it.. from terminal.. how to create using terminal in clearml server. this one i needed.. how to create access key & secret key using terminal

user creation even it will work after containers is up right

@<1523701087100473344:profile|SuccessfulKoala55> @<1523701070390366208:profile|CostlyOstrich36> thanks for the help.. i created creds.. using this command

curl -s -u "antony:antony" 

now i need help to bypass clearml-init manual paste key its wont support .. any other bypass is available

The reason am asking.. while agen is spin up.. it will add queue also ..

i am saying in the UI, if u give all in UI .. then give to UI.. why u mixing all.. becz of this password auth issue.. i am trying to do all from backend

@<1523701087100473344:profile|SuccessfulKoala55> 1st time.. it is running from jupyter notebook.. synced with clearml default.. 2md time it will run agent..

clear my doubt.. actually i tried .. i created queue name called test. i attached 2 GPU from diffrent machine.. but workload is going only one GPU.. this is my problem now.. slurm is not happening.

is this enough??

@Jake H. 
#!/bin/bash

# Get the access key and secret key from the API response
response=$(curl -s -u "antony:antony" 

)
access_key=$(echo "$response" | jq -r '.data.credentials.access_key')
secret_key=$(echo "$response" | jq -r '.data.credentials.secret_key')

# Echo the access key and secret key
echo "Access Key: $access_key"
echo "Secret Key: $secret_key"

# Verify the presence of access key and secret key
if [[ -n "$access_key" && -n "$secret_key" ]]; then
    echo "Acc...

i tell u my issue.. i created the template to come up clearml server & clearml work worker.. so our ml enginner expecting no manual authentication. they want all automation. now your case all manual.. when we want make pipeline like jenkins..we need automation.. my expectation.. only i need access key & secret key .. how to create?? other steps i created for the template.. then i will approach to my team.. thn they start cleaml enterprises solution

or before i have to do?? my case this one also its not working

my concern.. is .. if you can explain how to create access key & secret key from backend then all good..

okay.. this clearml dynamic access creation how its work..

How.. it will support.. when i have 20G+20G in diffrent machine.. how can i run 40G workload

@<1523701087100473344:profile|SuccessfulKoala55> sorry for the delay msg. actually i tried to connect clearml agent to clearml master from backend shell script. usually we will create api token access key from UI right. i wanna do it fully in terminal