open-toolchain / commons Goto Github PK

IP_ADDR=$( ibmcloud ks workers --cluster ${CLUSTER_ID} | grep normal | head -n 1 | awk '{ print $2 }' )
failed on head -1
echo ${PIPESTATUS[0]} ${PIPESTATUS[1]} ${PIPESTATUS[2]} ${PIPESTATUS[3]}
0 141 0

Please update like this, using tail instead.
IP_ADDR=$( ibmcloud ks workers --cluster ${CLUSTER_ID} | grep normal | tail -n 1 | awk '{ print $2 }' )

I had an error with Check Dockerfile script https://raw.githubusercontent.com/open-toolchain/commons/master/scripts/check_dockerfile.sh

where its fails if I dont have an env defined for an env set during a RUN directive.

error:

ERROR: RUN contains undefined ARG or ENV variable on line 49

ERROR: ./Dockerfile failed.

Example of RUN directive below. env MOREHEADER_DIR is set during runtime and there's no ENV directive in the dockerfile for MOREHEADER_DIR

RUN CONFARGS=$(nginx -V 2>&1 | sed -n -e 's/^.*arguments: //p') \
	mkdir -p /usr/src && \
	tar -zxC /usr/src -f nginx.tar.gz && \
  tar -xzvf "headers-more.tar.gz" && \
  MOREHEADER_DIR="$(pwd)/headers-more-nginx-module-${MOREHEADER_VERSION}" && \
  cd /usr/src/nginx-$NGINX_VERSION && \
  ./configure --with-compat $CONFARGS --add-dynamic-module=${MOREHEADER_DIR} && \
  make && make install

check_and_deploy_helm3.sh does a helm lint before it upgrades helm to v3, which causes it to fail for any helm v3 Charts.

Linting Helm Chart
Error: 1 chart(s) linted, 1 chart(s) failed
==> Linting chart/drupal
[ERROR] Chart.yaml: apiVersion 'v2' is not valid. The value must be "v1"
[ERROR] templates/: apiVersion 'v2' is not valid. The value must be "v1"

Running a toolchain with BUILD stage that uses commons/scripts/build_image_buildkit.sh failes to import cache manifest from us.icr.io.
In my opinion, the ERROR is that it is unable to find an image with the tag :latest as the images are stored with a custom tag as below

# Minting image tag using format: BUILD_NUMBER-BRANCH-COMMIT_ID-TIMESTAMP
# e.g. 3-master-50da6912-20181123114435

Lines 42 and 43 in the script

I believe so because of the following snippet

user@cloudshell:~$ ibmcloud cr image-inspect us.icr.io/<container-registry>/<image>
Inspecting image 'us.icr.io/<container-registry>/<image>:latest'...

FAILED
The image was not found.
Correct the image name or digest reference, and try again.

Can one of the possible solutions be to add an additional tag :latest once the image is built?
so that subsequent toolchains can use cache manifest instead of building the image again from scratch.

P.S. I'm new to Container technology, please correct me if I'm missing something