Remote host

Date and time

timedatectl list-timezones
sudo timedatectl set-timezone <timezone>

Scheduled tasks

systemctl list-timers

journalctl

journalctl --since 08:55:00 --no-pager | less

sysstat

Install sysstat

sudo apt install sysstat
sed -i 's/ENABLED="false"/ENABLED="true"/' /etc/default/sysstat
systemctl restart sysstat sysstat-collect.timer sysstat-summary.timer

Then inspect:

sar -r -q -f /var/log/sysstat/{sa_for_that_day} -s 17:50:01 -e 18:01:01

Unexpected pod restarts

Get restart count per pod:

kubectl -n $namespace get pods \
  -o=jsonpath='{range .items[*]}{range .status.containerStatuses[*]}{.name}{" "}{.restartCount}{"\n"}{end}{end}' \
  | grep -v charm

or:

kubectl -n cos-lite-load-test get pods -o=jsonpath='{"{"}{range .items[*]}{"\""}{.metadata.name}{"\":{"}{range .status.containerStatuses[*]}{"\""}{.name}{"\":"}{.restartCount}{","}{end}{"},"}{end}{"}"}' | sed 's/,},/},/g' | sed 's/},}/}}/g' | jq

Check for OOM kills:

journalctl --no-pager -kqg 'killed process' -o verbose --output-fields=MESSAGE

Check for node pressure eviction

journalctl | grep eviction

Pod hits resource limit

This is useful to see if resource limits prevent scheduling a pod:

kubectl get pod grafana-0 -o=jsonpath='{.status}' -n test-bundle-zdsv

Certs

View

echo | openssl s_client -showcerts -servername charmhub.io -connect charmhub.io:443 | openssl x509 -text -noout

Compare

diff -y <(openssl x509 -in a.crt -text -noout) <(openssl x509 -in b.crt -text -noout)