The exec Singularity sub-command allows you to spawn an arbitrary command within your container image as if it were running directly on the host system. All standard IO, pipes, and file systems are accessible via the command being exec’ed within the container. Note that this exec is different from the Docker exec, as it does not require a container to be “running” before using it.

Examples

Printing the OS release inside the container:

$ singularity exec container.img cat /etc/os-release
PRETTY_NAME="Debian GNU/Linux 8 (jessie)"
NAME="Debian GNU/Linux"
VERSION_ID="8"
VERSION="8 (jessie)"
ID=debian
HOME_URL="http://www.debian.org/"
SUPPORT_URL="http://www.debian.org/support"
BUG_REPORT_URL="https://bugs.debian.org/"
$ 

Printing the OS release for a running instance:

Use the instance://<instance name> syntax like so:

$ singularity exec instance://my-instance cat /etc/os-release

Runtime Flags

If you are interested in containing an environment or filesystem locations, we highly recommend that you look at the singularity run help and our documentation on flags to better customize this command.

Special Characters

And properly passing along special characters to the program within the container.

$ singularity exec container.img echo -ne "hello\nworld\n\n"
hello
world
$ 

And a demonstration using pipes:

$ cat debian.def | singularity exec container.img grep 'MirrorURL'
MirrorURL "http://ftp.us.debian.org/debian/"
$ 

A Python example

Starting with the file hello.py in the current directory with the contents of:

#!/usr/bin/python

import sys
print("Hello World: The Python version is %s.%s.%s" % sys.version_info[:3])

Because our home directory is automatically bound into the container, and we are running this from our home directory, we can easily execute that script using the Python within the container:

$ singularity exec /tmp/Centos7-ompi.img /usr/bin/python hello.py 
Hello World: The Python version is 2.7.5

We can also pipe that script through the container and into the Python binary which exists inside the container using the following command:

$ cat hello.py | singularity exec /tmp/Centos7-ompi.img /usr/bin/python 
Hello World: The Python version is 2.7.5

For demonstration purposes, let’s also try to use the latest Python container which exists in DockerHub to run this script:

$ singularity exec docker://python:latest /usr/local/bin/python hello.py
library/python:latest
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:fbd06356349dd9fb6af91f98c398c0c5d05730a9996bbf88ff2f2067d59c70c4
Downloading layer: sha256:644eaeceac9ff6195008c1e20dd693346c35b0b65b9a90b3bcba18ea4bcef071
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:766692404ca72f4e31e248eb82f8eca6b2fcc15b22930ec50e3804cc3efe0aba
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:6a3d69edbe90ef916e1ecd8d197f056de873ed08bcfd55a1cd0b43588f3dbb9a
Downloading layer: sha256:ff18e19c2db42055e6f34323700737bde3c819b413997cddace2c1b7180d7efd
Downloading layer: sha256:7b9457ec39de00bc70af1c9631b9ae6ede5a3ab715e6492c0a2641868ec1deda
Downloading layer: sha256:a3ed95caeb02ffe68cdd9fd84406680ae93d633cb16422d00e8a7c22955b46d4
Downloading layer: sha256:6a5a5368e0c2d3e5909184fa28ddfd56072e7ff3ee9a945876f7eee5896ef5bb
Hello World: The Python version is 3.5.2

A GPU example

If your host system has an NVIDIA GPU card and a driver installed you can leverage the card with the --nv option. (This example requires a fairly recent version of the NVIDIA driver on the host system to run the latest version of TensorFlow.

$ git clone https://github.com/tensorflow/models.git
$ singularity exec --nv docker://tensorflow/tensorflow:latest-gpu \
    python ./models/tutorials/image/mnist/convolutional.py
Docker image path: index.docker.io/tensorflow/tensorflow:latest-gpu
Cache folder set to /home/david/.singularity/docker
[19/19] |===================================| 100.0%
Creating container runtime...
Extracting data/train-images-idx3-ubyte.gz
Extracting data/train-labels-idx1-ubyte.gz
Extracting data/t10k-images-idx3-ubyte.gz
Extracting data/t10k-labels-idx1-ubyte.gz
2017-08-18 20:33:59.677580: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
2017-08-18 20:33:59.677620: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2017-08-18 20:34:00.148531: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:893] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2017-08-18 20:34:00.148926: I tensorflow/core/common_runtime/gpu/gpu_device.cc:955] Found device 0 with properties:
name: GeForce GTX 760 (192-bit)
major: 3 minor: 0 memoryClockRate (GHz) 0.8885
pciBusID 0000:03:00.0
Total memory: 2.95GiB
Free memory: 2.92GiB
2017-08-18 20:34:00.148954: I tensorflow/core/common_runtime/gpu/gpu_device.cc:976] DMA: 0
2017-08-18 20:34:00.148965: I tensorflow/core/common_runtime/gpu/gpu_device.cc:986] 0:   Y
2017-08-18 20:34:00.148979: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1045] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce GTX 760 (192-bit), pci bus id: 0000:03:00.0)
Initialized!
Step 0 (epoch 0.00), 21.7 ms
Minibatch loss: 8.334, learning rate: 0.010000
Minibatch error: 85.9%
Validation error: 84.6%
Step 100 (epoch 0.12), 20.9 ms
Minibatch loss: 3.235, learning rate: 0.010000
Minibatch error: 4.7%
Validation error: 7.8%
Step 200 (epoch 0.23), 20.5 ms
Minibatch loss: 3.363, learning rate: 0.010000
Minibatch error: 9.4%
Validation error: 4.2%
[...snip...]
Step 8500 (epoch 9.89), 20.5 ms
Minibatch loss: 1.602, learning rate: 0.006302
Minibatch error: 0.0%
Validation error: 0.9%
Test error: 0.8%