Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Speedup deployment and propagate SSH keys on client nodes #8

Merged
merged 1 commit into from
Nov 18, 2021
Merged

Speedup deployment and propagate SSH keys on client nodes #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion terraform/examples/io500/setup_io500.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,9 @@ source ./configure.sh
# Clean and configure DAOS servers
source ./clean.sh

# Copy SSH keys to client nodes
pdcp -w ^hosts -r .ssh ~

echo "Copy agent config files from server"
rm -f .ssh/known_hosts
scp ${DAOS_FIRST_SERVER}:/etc/daos/daos_agent.yml .
Expand Down Expand Up @@ -78,7 +81,7 @@ sed -i "s/^blockSize.*/blockSize = 1000000m/g" temp.ini
sed -i "s/^filePerProc.*/filePerProc = TRUE /g" temp.ini
sed -i "s/^nproc.*/nproc = $(( ${NUMBER_OF_CLIENTS_INSTANCES} * $(nproc --all) ))/g" temp.ini

# Run IO500 benchmark
echo "# Run IO500 benchmark"
mpirun --hostfile hosts -env I_MPI_OFI_PROVIDER="tcp;ofi_rxm" --bind-to socket -np $(( ${NUMBER_OF_CLIENTS_INSTANCES} * $(nproc --all) )) /usr/local/io500/io500 temp.ini

echo "Cleaning up after run ..."
Expand Down
17 changes: 11 additions & 6 deletions terraform/examples/io500/start.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,8 +26,10 @@ terraform plan -out=tfplan -input=false
terraform apply -input=false tfplan
popd

echo "# Wait for instances"
sleep 10
echo "# Wait for DAOS client instances"
gcloud compute instance-groups managed wait-until ${TF_VAR_client_template_name} --stable --zone ${TF_VAR_zone}
echo "# Wait for DAOS server instances"
gcloud compute instance-groups managed wait-until ${TF_VAR_server_template_name} --stable --zone ${TF_VAR_zone}

echo "# Add external IP to first client, so that it will be accessible over normal SSH"
gcloud compute instances add-access-config ${DAOS_FIRST_CLIENT} --zone ${TF_VAR_zone} && sleep 10
Expand All @@ -37,18 +39,21 @@ echo "##########################"
echo "# Configure SSH access #"
echo "##########################"
echo "# Prepare SSH key"
rm -f ./id_rsa* ; ssh-keygen -t rsa -b 4096 -C "root" -N '' -f id_rsa
rm -f ./id_rsa* ; ssh-keygen -t rsa -b 4096 -C "${SSH_USER}" -N '' -f id_rsa
echo "${SSH_USER}:$(cat id_rsa.pub)" > keys.txt

echo "# Configuring SSH on nodes"
for node in $ALL_NODES
do
echo "# Configuring SSH on ${node}"
# Disable OSLogin to be able to connect with SSH keys uploaded in next command
gcloud compute instances add-metadata ${node} --metadata enable-oslogin=FALSE
gcloud compute instances add-metadata ${node} --metadata enable-oslogin=FALSE && \
# Upload SSH key to instance, so that you could login to instance over SSH
gcloud compute instances add-metadata ${node} --metadata-from-file ssh-keys=keys.txt
gcloud compute instances add-metadata ${node} --metadata-from-file ssh-keys=keys.txt &
done

# Wait for SSH configuring tasks to finish
wait

echo "# Copy SSH key to first DAOS client"
scp -i id_rsa -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null \
id_rsa \
Expand Down