Crane
crane docs and recipes
1
2
3
4
5
6
7
8
9
10
11
12
|
function img-tags {
if [ "${1}" = "-h" ]; then
echo "Usage: imgtags cgr.dev/reponame/imagename]"
echo "List tags for an OCI-compliant image, omitting digests, using crane."
return
fi
if [ "${1}" = "" ]; then
echo "Image name required."
return
fi
crane ls ${1} --omit-digest-tags
}
|
Docker
Clean up all the things
The “nuke it from orbit” approach that removes all stopped containers, dangling images, build cache, and volumes. I use this one a lot.
1
2
3
4
5
6
7
8
9
10
11
|
function docker-cleanup {
if [ "${1}" = "-h" ]; then
echo "Usage: docker-cleanup"
echo "Remove all stopped containers and dangling images, build cache, volumes, and more."
return
fi
# remove all stopped containers, networks, images
docker system prune -af
# remove all unused volumes, not just anonymous ones
docker volume prune -af
}
|
1
2
3
4
5
6
7
8
9
10
11
12
|
function docker-size {
if [ "${1}" = "-h" ]; then
echo "Usage: docker-size [image]"
echo "Inspect the size of a Docker image and return it in human-readable format."
return
fi
if [ "${1}" = "" ]; then
echo "Image name required."
return
fi
docker inspect -f "{{ .Size }}" ${1} | numfmt --to=si
}
|
Grype
Summarize Grype results
This prints out a markdown table to copy/paste into another system.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
|
function grype-summary () {
if [ "${1}" = "-h" ]; then
echo "Usage: grype-summary [path]"
echo "Summarize vulnerabilities found by Grype."
return
fi
if [ "${1}" = "" ]; then
echo "Path or image name required."
return
fi
grype ${1} -o json --file grype.json -q
echo "Total = $(cat grype.json | jq '.matches[].vulnerability.severity' | uniq -c | awk '{sum += $1} END {print sum}')"
echo ""
echo "| Count | Severity |"
echo "|-------|-------------|"
printf "| %-5s | %-11s |\n" $(cat grype.json | jq '.matches[].vulnerability.severity' | grep -c 'Critical') "critical"
printf "| %-5s | %-11s |\n" $(cat grype.json | jq '.matches[].vulnerability.severity' | grep -c 'High') "high"
printf "| %-5s | %-11s |\n" $(cat grype.json | jq '.matches[].vulnerability.severity' | grep -c 'Medium') "medium"
printf "| %-5s | %-11s |\n" $(cat grype.json | jq '.matches[].vulnerability.severity' | grep -c 'Low') "low"
printf "| %-5s | %-11s |\n" $(cat grype.json | jq '.matches[].vulnerability.severity' | grep -c 'Negligible') "negligible"
printf "| %-5s | %-11s |\n" $(cat grype.json | jq '.matches[].vulnerability.severity' | grep -c 'Unknown') "unknown"
rm grype.json
}
|
Outputs something like this:
1
2
3
4
5
6
7
8
9
10
|
Total = 558
| Count | Severity |
|-------|-------------|
| 38 | critical |
| 170 | high |
| 298 | medium |
| 43 | low |
| 0 | negligible |
| 9 | unknown |
|
Multi-image Grype summarization
I wrote a Python script that takes a newline-delimited text file to summarize multiple Grype results into a pretty console table.
1
2
3
4
5
6
7
8
9
10
11
|
ᐅ ./grype-table.py test-list.txt
Analyzing nginx:1...
Analyzing registry.access.redhat.com/ubi9/ubi-init:9.3...
Analyzing cgr.dev/chainguard-private/nginx-fips:1...
+----------------------------------------------+----------+------+--------+-----+------------+---------+
| Image | Critical | High | Medium | Low | Negligible | Unknown |
+----------------------------------------------+----------+------+--------+-----+------------+---------+
| nginx:1 | 2 | 14 | 34 | 4 | 72 | 20 |
| registry.access.redhat.com/ubi9/ubi-init:9.3 | 0 | 1 | 32 | 143 | 0 | 4 |
| cgr.dev/chainguard-private/nginx-fips:1 | 0 | 0 | 0 | 0 | 0 | 0 |
+----------------------------------------------+----------+------+--------+-----+------------+---------+
|
To make life easier, it’s symlinked into /usr/local/bin.
Helm
helm docs
List unique images
1
2
3
4
5
6
7
8
9
10
11
12
|
function helm-ls-images {
if [ "${1}" = "-h" ]; then
echo "Usage: helm-images [chart]"
echo "List of unique images used in Helm chart."
return
fi
if [ "${1}" = "" ]; then
echo "Chart name required."
return
fi
helm template "${1}" | grep -oE 'image: .+' | cut -d' ' -f2 | sort | uniq | tr -d '"'
}
|
Kubectl
Export a specific context
1
|
kubectl config view --minify --flatten --raw --context=CONTEXTNAME
|
List all images running in a cluster
Does exactly what the title says it does. I usually redirect output into a text file, then run the Python script above on it for a quick table of vulnerability count by images in a cluster.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
|
function k-ls-images {
if [ "${1}" = "-h" ]; then
echo "Usage: k-ls-images"
echo "List all images running in a cluster, using the current context."
return
fi
if [ "${1}" = "-c" ]; then
echo "Returning count of pods running each image as well."
echo " ... any `uniq` arguments work here ... "
fi
kubectl get pods --all-namespaces \
-o jsonpath="{.items[*].spec['initContainers', 'containers'][*].image}" |\
tr -s '[[:space:]]' '\n' |\
sort |\
uniq ${1}
}
|