mirror of
https://github.com/ClusterCockpit/cc-docker.git
synced 2025-07-24 21:56:21 +02:00
Introduce slurm cluster
This commit is contained in:
39
slurm/worker/Dockerfile
Normal file
39
slurm/worker/Dockerfile
Normal file
@@ -0,0 +1,39 @@
|
||||
FROM scidas/slurm.base:19.05.1
|
||||
MAINTAINER Michael J. Stealey <stealey@renci.org>
|
||||
|
||||
# install openmpi 3.0.1
|
||||
RUN yum -y install \
|
||||
gcc-c++ \
|
||||
gcc-gfortran \
|
||||
&& yum -y localinstall \
|
||||
/packages/openmpi-*.rpm
|
||||
|
||||
# install Lmod 7.7
|
||||
RUN yum -y install \
|
||||
lua-posix \
|
||||
lua \
|
||||
lua-filesystem \
|
||||
lua-devel \
|
||||
wget \
|
||||
bzip2 \
|
||||
expectk \
|
||||
make \
|
||||
&& wget https://sourceforge.net/projects/lmod/files/Lmod-7.7.tar.bz2 \
|
||||
&& tar -xjvf Lmod-7.7.tar.bz2
|
||||
WORKDIR /Lmod-7.7
|
||||
RUN ./configure --prefix=/opt/apps \
|
||||
&& make install \
|
||||
&& ln -s /opt/apps/lmod/lmod/init/profile /etc/profile.d/z00_lmod.sh \
|
||||
&& ln -s /opt/apps/lmod/lmod/init/cshrc /etc/profile.d/z00_lmod.csh
|
||||
|
||||
WORKDIR /home/worker
|
||||
|
||||
# clean up
|
||||
RUN rm -f /packages/slurm-*.rpm /packages/openmpi-*.rpm \
|
||||
&& yum clean all \
|
||||
&& rm -rf /var/cache/yum \
|
||||
&& rm -f /Lmod-7.7.tar.bz2
|
||||
|
||||
COPY docker-entrypoint.sh /docker-entrypoint.sh
|
||||
|
||||
ENTRYPOINT ["/usr/local/bin/tini", "--", "/docker-entrypoint.sh"]
|
13
slurm/worker/Makefile
Normal file
13
slurm/worker/Makefile
Normal file
@@ -0,0 +1,13 @@
|
||||
|
||||
SLURM_VERSION = 19.05.1
|
||||
IMAGE = scidas/slurm.worker
|
||||
|
||||
.PHONY: all build clean test
|
||||
|
||||
all: build
|
||||
|
||||
build:
|
||||
docker build -t $(IMAGE):$(SLURM_VERSION) .
|
||||
|
||||
clean:
|
||||
@[ -z $(docker images -q $(IMAGE):$(SLURM_VERSION)) ] || docker rmi $(IMAGE):$(SLURM_VERSION)
|
3
slurm/worker/README.md
Normal file
3
slurm/worker/README.md
Normal file
@@ -0,0 +1,3 @@
|
||||
# Slurm Worker
|
||||
|
||||
TODO
|
69
slurm/worker/docker-entrypoint.sh
Executable file
69
slurm/worker/docker-entrypoint.sh
Executable file
@@ -0,0 +1,69 @@
|
||||
#!/usr/bin/env bash
|
||||
set -e
|
||||
|
||||
# start sshd server
|
||||
_sshd_host() {
|
||||
if [ ! -d /var/run/sshd ]; then
|
||||
mkdir /var/run/sshd
|
||||
ssh-keygen -t rsa -f /etc/ssh/ssh_host_rsa_key -N ''
|
||||
fi
|
||||
/usr/sbin/sshd
|
||||
}
|
||||
|
||||
# start munge using existing key
|
||||
_munge_start_using_key() {
|
||||
echo -n "cheking for munge.key"
|
||||
while [ ! -f /.secret/munge.key ]; do
|
||||
echo -n "."
|
||||
sleep 1
|
||||
done
|
||||
echo ""
|
||||
cp /.secret/munge.key /etc/munge/munge.key
|
||||
chown -R munge: /etc/munge /var/lib/munge /var/log/munge /var/run/munge
|
||||
chmod 0700 /etc/munge
|
||||
chmod 0711 /var/lib/munge
|
||||
chmod 0700 /var/log/munge
|
||||
chmod 0755 /var/run/munge
|
||||
sudo -u munge /sbin/munged
|
||||
munge -n
|
||||
munge -n | unmunge
|
||||
remunge
|
||||
}
|
||||
|
||||
# wait for worker user in shared /home volume
|
||||
_wait_for_worker() {
|
||||
if [ ! -f /home/worker/.ssh/id_rsa.pub ]; then
|
||||
echo -n "cheking for id_rsa.pub"
|
||||
while [ ! -f /home/worker/.ssh/id_rsa.pub ]; do
|
||||
echo -n "."
|
||||
sleep 1
|
||||
done
|
||||
echo ""
|
||||
fi
|
||||
}
|
||||
|
||||
# run slurmd
|
||||
_slurmd() {
|
||||
if [ ! -f /.secret/slurm.conf ]; then
|
||||
echo -n "cheking for slurm.conf"
|
||||
while [ ! -f /.secret/slurm.conf ]; do
|
||||
echo -n "."
|
||||
sleep 1
|
||||
done
|
||||
echo ""
|
||||
fi
|
||||
mkdir -p /var/spool/slurm/d
|
||||
chown slurm: /var/spool/slurm/d
|
||||
cp /.secret/slurm.conf /etc/slurm/slurm.conf
|
||||
touch /var/log/slurmd.log
|
||||
chown slurm: /var/log/slurmd.log
|
||||
/usr/sbin/slurmd
|
||||
}
|
||||
|
||||
### main ###
|
||||
_sshd_host
|
||||
_munge_start_using_key
|
||||
_wait_for_worker
|
||||
_slurmd
|
||||
|
||||
tail -f /dev/null
|
Reference in New Issue
Block a user