Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Answered
Hi People When I Try To Use Docker Agents They Fail. If I Run The Command Clearml-Agent Daemon --Gpus 0 --Queue Default --Foreground Inside A Docker Prepared With All The Requirements Installed. It Tells Me:

Hi people
When I try to use docker agents they fail.
If I run the command clearml-agent daemon --gpus 0 --queue default --foreground inside a docker prepared with all the requirements installed. It tells me:

Could not fetch URL 
: There was a problem confirming the ssl certificate: HTTP
SConnectionPool(host='pypi.org', port=443): Max retries exceeded with url: /simple/pip/ (Caused by SSLErro
r(SSLCertVerificationError(1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get l
ocal issuer certificate (_ssl.c:1007)'))) - skipping 

I don't know how to give my SSL certificate to the specific environment that the agents decided to use
I would like just to use the docker environment and no that cython thing
I would really like to use clearml in production, but it is becoming harder 🫠

  
  
Posted 2 months ago
Votes Newest

Answers 10


This is the last situation
I could create a docker that works

FROM pytorch/pytorch:2.2.1-cuda12.1-cudnn8-runtime

WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y \
    libgl1-mesa-glx \
    libglib2.0-0 \
    git \
    curl\
    && rm -rf /var/lib/apt/lists/*

# Install dependencies
COPY requirements.txt /tmp/requirements.txt
RUN pip install --no-cache-dir -r /tmp/requirements.txt

# Update CA certificates
COPY hme_root_CA.crt /usr/local/share/ca-certificates/hme_root_CA.crt
RUN update-ca-certificates

ENV SSL_CERT_FILE=/usr/local/share/ca-certificates/hme_root_CA.crt
ENV REQUESTS_CA_BUNDLE=/usr/local/share/ca-certificates/hme_root_CA.crt
ENV PIP_CONFIG_FILE=/root/conda.conf
ENV CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL=true
ENV CLEARML_AGENT_SKIP_PIP_VENV_INSTALL=true

RUN pip config set global.cert /usr/local/share/ca-certificates/hme_root_CA.crt
RUN pip install --trusted-host pypi.python.org --trusted-host files.pythonhosted.org --trusted-host pypi.org --upgrade pip
RUN conda config --set ssl_verify /usr/local/share/ca-certificates/hme_root_CA.crt

COPY conda.conf /root/conda.conf


# Configure ClearML
COPY clearml.conf /root/clearml.conf
RUN clearml-init

# Start ClearML worker
CMD [ "clearml-agent", "daemon", "--queue", "default", "--foreground" ]

But it just works if I use clearml-agent daemon --gpus 0 --queue default --foreground

If I try to use clearml-agent daemon --gpus 0 --queue default --foreground --docker (in the host)
When It opens the docker mentioned above, it always tries to install different apt packages, and tries to install some conda packages and gets stall

Do you know if is possible to avoid all the installations? And just run everything in the docker's environment?

  
  
Posted 2 months ago

so what was the solution/hack then ?

  
  
Posted 2 months ago

You are using CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL the wrong way

  
  
Posted 2 months ago

Hi @<1576381444509405184:profile|ManiacalLizard2> at the end we make it works
It has a lot of warning but is able to run the experiments hahaha
Thanks for your help

  
  
Posted 2 months ago

Are you running within a zero-trust environment like ZScaler ?
Feels like your issue is not ClearML itself, but issue with https/SSL and certificate from your zero-trust system

  
  
Posted 2 months ago

CLEARML_AGENT_SKIP_PIP_VENV_INSTALL need to be a path

  
  
Posted 2 months ago

@<1576381444509405184:profile|ManiacalLizard2>
I changhed that
But it continues trying to install the next packages:

The following additional packages will be installed:                                                      
  binutils binutils-common binutils-x86-64-linux-gnu build-essential cpp cpp-9                            
  dpkg-dev fakeroot file g++ g++-9 gcc gcc-9 gcc-9-base git-man krb5-locales                              
  less libalgorithm-diff-perl libalgorithm-diff-xs-perl                                                   
  libalgorithm-merge-perl libasan5 libatomic1 libbinutils libbrotli1 libbsd0                              
  libc-dev-bin libc6 libc6-dev libcbor0.6 libcc1-0 libcrypt-dev libctf-nobfd0                             
  libctf0 libcurl3-gnutls libdpkg-perl libedit2 liberror-perl libexpat1                                   
  libexpat1-dev libfakeroot libfido2-1 libfile-fcntllock-perl libgcc-9-dev                                
  libgdbm-compat4 libgdbm6 libgomp1 libgssapi-krb5-2 libisl22 libitm1                                     
  libk5crypto3 libkeyutils1 libkrb5-3 libkrb5support0 liblocale-gettext-perl                              
  liblsan0 libmagic-mgc libmagic1 libmpc3 libmpdec2 libmpfr6 libnghttp2-14
  libperl5.30 libpsl5 libpython3-dev libpython3-stdlib libpython3.8
  libpython3.8-dev libpython3.8-minimal libpython3.8-stdlib libquadmath0
  librtmp1 libssh-4 libstdc++-9-dev libtsan0 libubsan1 libx11-6 libx11-data
  libxau6 libxcb1 libxdmcp6 libxext6 libxmuu1 linux-libc-dev make manpages
  manpages-dev mime-support netbase openssh-client patch perl perl-base
  perl-modules-5.30 publicsuffix python-pip-whl python3 python3-dev
  python3-distutils python3-lib2to3 python3-minimal python3-pkg-resources
  python3-setuptools python3-wheel python3.8 python3.8-dev python3.8-minimal
  xauth xz-utils zlib1g-dev

(this installation has some errors or warnings in the middle)
Then it tries to install with pip, but it gets stall after it finish and doesn't continue any more

  Using cached zipp-3.20.1-py3-none-any.whl (9.0 kB)
Installing collected packages: attrs, six, orderedmultidict, furl, urllib3, rpds-py, referencing, zipp, importlib-resources, jsonschema-specifications, pkgutil-resolve-name, jsonschema, python-dateutil, psutil, PyYAML, pyjwt, pyparsing, certifi, charset-normalizer, idna, requests, filelock, platformdirs, distlib, virtualenv, pathlib2, clearml-agent

Using the next docker file for the worker

FROM pytorch/pytorch:2.2.1-cuda12.1-cudnn8-runtime

WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y \
    libgl1-mesa-glx \
    libglib2.0-0 \
    git \
    curl\
    && rm -rf /var/lib/apt/lists/*

# Install dependencies
COPY requirements.txt /tmp/requirements.txt
RUN pip install --no-cache-dir -r /tmp/requirements.txt

# Update CA certificates
COPY hme_root_CA.crt /usr/local/share/ca-certificates/hme_root_CA.crt
RUN update-ca-certificates

ENV SSL_CERT_FILE=/usr/local/share/ca-certificates/hme_root_CA.crt
ENV REQUESTS_CA_BUNDLE=/usr/local/share/ca-certificates/hme_root_CA.crt
ENV PIP_CONFIG_FILE=/root/conda.conf
ENV CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL=1

RUN pip config set global.cert /usr/local/share/ca-certificates/hme_root_CA.crt
RUN pip install --trusted-host pypi.python.org --trusted-host files.pythonhosted.org --trusted-host pypi.org --upgrade pip
RUN conda config --set ssl_verify /usr/local/share/ca-certificates/hme_root_CA.crt
COPY conda.conf /root/conda.conf



# Configure ClearML
COPY clearml.conf /root/clearml.conf
RUN clearml-init

# Start ClearML worker
CMD [ "clearml-agent", "daemon", "--queue", "default", "--foreground" ]
  
  
Posted 2 months ago

It could be something related with conda in the docker?

  
  
Posted 2 months ago

while the other may need to be 1 instead of true

  
  
Posted 2 months ago

eg: None

  
  
Posted 2 months ago
146 Views
10 Answers
2 months ago
2 months ago
Tags