passing system date as variable to Kubernetes CronJob - kubernetes

Question

I found this Pass date command as parameter to kubernetes cronjob which is similar, but did not solve my problem.
I'm trying to backup etcd using a cronjob, but etcd image doesn't have "date" command.
apiVersion: batch/v1beta1
kind: CronJob
metadata:
name: backup
namespace: kube-system
spec:
concurrencyPolicy: Allow
failedJobsHistoryLimit: 1
jobTemplate:
spec:
template:
spec:
containers:
- args:
- -c
- etcdctl --endpoints=https://127.0.0.1:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt
--cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key
snapshot save /backup/etcd-snapshot-$(DATE_CURR).db
command:
- /bin/sh
env:
- name: ETCDCTL_API
value: "3"
- name: DATE_CURR
#value: $(date --date= +"%Y-%m-%d_%H:%M:%S_%Z")
value: $(date +"%Y-%m-%d_%H:%M:%S_%Z")
image: k8s.gcr.io/etcd:3.4.13-0
imagePullPolicy: IfNotPresent
name: backup
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
volumeMounts:
- mountPath: /etc/kubernetes/pki/etcd
name: etcd-certs
readOnly: true
- mountPath: /backup
name: backup
- args:
- -c
- find /backup -type f -mtime +30 -exec rm -f {} \;
command:
- /bin/sh
env:
- name: ETCDCTL_API
value: "3"
image: k8s.gcr.io/etcd:3.4.13-0
imagePullPolicy: IfNotPresent
name: cleanup
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
volumeMounts:
- mountPath: /backup
name: backup
dnsPolicy: ClusterFirst
hostNetwork: true
nodeName: homelab-a
restartPolicy: OnFailure
schedulerName: default-scheduler
securityContext: {}
terminationGracePeriodSeconds: 30
volumes:
- hostPath:
path: /etc/kubernetes/pki/etcd
type: DirectoryOrCreate
name: etcd-certs
- hostPath:
path: /opt/etcd_backups
type: DirectoryOrCreate
name: backup
schedule: 0 */6 * * *
successfulJobsHistoryLimit: 3
suspend: false
when this runs, it produces a file named "etcd-snapshot-.db", no date. If I can catch the logs, it says that "date" is not a known command. Sure enough, when I am able to exec into the pod while it's running, date does not work in the etcd image. How can I pass the date from the system as a variable so that it just uses the text and not try to actually invoke the "date" command?
Edit: Thanks to #Andrew, here is the solution:
apiVersion: batch/v1beta1
kind: CronJob
metadata:
name: backup
namespace: kube-system
spec:
concurrencyPolicy: Allow
failedJobsHistoryLimit: 1
jobTemplate:
spec:
template:
spec:
containers:
- args:
- -c
- etcdctl --endpoints=https://127.0.0.1:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt
--cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key
snapshot save /backup/etcd-snapshot-$(echo `printf "%(%Y-%m-%d_%H:%M:%S_%Z)T\n"`).db
command:
- /bin/sh
env:
- name: ETCDCTL_API
value: "3"
image: k8s.gcr.io/etcd:3.4.13-0
imagePullPolicy: IfNotPresent
name: backup
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
volumeMounts:
- mountPath: /etc/kubernetes/pki/etcd
name: etcd-certs
readOnly: true
- mountPath: /backup
name: backup
- args:
- -c
- find /backup -type f -mtime +30 -exec rm -f {} \;
command:
- /bin/sh
env:
- name: ETCDCTL_API
value: "3"
image: k8s.gcr.io/etcd:3.4.13-0
imagePullPolicy: IfNotPresent
name: cleanup
resources: {}
terminationMessagePath: /dev/termination-log
terminationMessagePolicy: File
volumeMounts:
- mountPath: /backup
name: backup
dnsPolicy: ClusterFirst
hostNetwork: true
nodeName: homelab-a
restartPolicy: OnFailure
schedulerName: default-scheduler
securityContext: {}
terminationGracePeriodSeconds: 30
volumes:
- hostPath:
path: /etc/kubernetes/pki/etcd
type: DirectoryOrCreate
name: etcd-certs
- hostPath:
path: /opt/etcd_backups
type: DirectoryOrCreate
name: backup
schedule: 0 */6 * * *
successfulJobsHistoryLimit: 3
suspend: false

Answer 1

You can use printf like this:
printf "%(%Y-%m-%d_%H:%M:%S_%Z)T\n"
Use man strftime to get conversion specification sequences.
Just tried it inside etcd container in kubernetes 1.19,
etcdctl --endpoints=https://127.0.0.1:2379 --cacert=/etc/kubernetes/pki/etcd/ca.crt --cert=/etc/kubernetes/pki/etcd/healthcheck-client.crt --key=/etc/kubernetes/pki/etcd/healthcheck-client.key snapshot save /tmp/etcd-snapshot-$(printf "%(%Y-%m-%d_%H:%M:%S_%Z)T\n").db
{"level":"info","ts":1603748998.7475638,"caller":"snapshot/v3_snapshot.go:119","msg":"created temporary db file","path":"/tmp/etcd-snapshot-2020-10-26_21:49:58_UTC.db.part"}
{"level":"info","ts":"2020-10-26T21:49:58.760Z","caller":"clientv3/maintenance.go:200","msg":"opened snapshot stream; downloading"}
{"level":"info","ts":1603748998.7616487,"caller":"snapshot/v3_snapshot.go:127","msg":"fetching snapshot","endpoint":"https://127.0.0.1:2379"}
{"level":"info","ts":"2020-10-26T21:49:58.889Z","caller":"clientv3/maintenance.go:208","msg":"completed snapshot read; closing"}
{"level":"info","ts":1603748998.9136698,"caller":"snapshot/v3_snapshot.go:142","msg":"fetched snapshot","endpoint":"https://127.0.0.1:2379","size":"5.9 MB","took":0.165411663}
{"level":"info","ts":1603748998.914325,"caller":"snapshot/v3_snapshot.go:152","msg":"saved","path":"/tmp/etcd-snapshot-2020-10-26_21:49:58_UTC.db"}
Snapshot saved at /tmp/etcd-snapshot-2020-10-26_21:49:58_UTC.db
Edit: A reproducible example using jobs, without passing any env vars:
apiVersion: batch/v1
kind: Job
metadata:
name: printf
spec:
template:
spec:
containers:
- name: printf
image: k8s.gcr.io/etcd:3.4.9-1
command: ["/bin/sh"]
args:
- -c
- echo `printf "%(%Y-%m-%d_%H:%M:%S_%Z)T\n"`
restartPolicy: Never
backoffLimit: 4
# kubectl create -f job.yaml
job.batch/printf created
# kubectl logs printf-lfcdh
2020-10-27_10:03:59_UTC
Notice the use of backticks `` instead of $()