Bug exists in 201911 branch where unit conversion for docker stats is incorrect. Both MiB/GiB to byes conversion is incorrect Example: admin@str-s6000-acs-10:/usr/bin$ docker stats --no-stream -a CONTAINER ID NAME CPU % MEM USAGE / LIMIT MEM % NET I/O BLOCK I/O PIDS e958c81d27a8 mgmt-framework 0.00% 0B / 0B 0.00% 0B / 0B 0B / 0B 0 9b6b7b4361d5 telemetry 3.13% 86.31MiB / 7.785GiB 1.08% 0B / 0B 0B / 106kB 30 e7fee0b617fe snmp 70.28% 57.03MiB / 7.785GiB 0.72% 0B / 0B 0B / 102kB 9 admin@str-s6000-acs-10:/usr/bin$ redis-cli -n 6 hgetall "DOCKER_STATS|e7fee0b617fe" "MEM%" "0.72" "MEM_LIMIT_BYTES" "8359080099840" "NAME" "snmp" "NET_OUT_BYTES" "0" "MEM_BYTES" "5980028928" "BLOCK_OUT_BYTES" "102000" "NET_IN_BYTES" "0" "BLOCK_IN_BYTES" "0" "PIDS" "9" "CPU%" "5.96" |
||
---|---|---|
.. | ||
apt | ||
bash | ||
caclmgrd | ||
config-setup | ||
constants | ||
corefile_uploader | ||
cron.d | ||
ebtables | ||
environment | ||
fstrim | ||
hostcfgd | ||
hostname | ||
interfaces | ||
kubernetes | ||
logrotate | ||
misc | ||
monit | ||
ntp | ||
platform | ||
procdockerstatsd | ||
process-reboot-cause | ||
rsyslog | ||
snmp | ||
sudoers | ||
sysctl | ||
syslog | ||
systemd | ||
topology | ||
updategraph | ||
warmboot-finalizer | ||
watchdog-control |