Kubelet CPU/MemoryAccounting: Difference between revisions
Jump to navigation
Jump to search
DrEdWilliams (talk | contribs) (Created page with "If you start seeing this type of entry in the system logs: Aug 10 21:33:50 telmar kubelet[704]: W0810 21:33:50.479159 704 container_manager_linux.go:818] CPUAccounting n...") |
DrEdWilliams (talk | contribs) mNo edit summary |
||
| Line 6: | Line 6: | ||
Then (for some reason), the systemd configuration has not enabled the CPU/Memory Accounting, and kubelet doesn't like that ... | Then (for some reason), the systemd configuration has not enabled the CPU/Memory Accounting, and kubelet doesn't like that ... | ||
On both Debian 10 and CentOS 7, the parameters are set in the '''/etc/systemd/system.conf''' -- uncomment the lines reffering to '''DefaultCPUAccounting''' and '''DefaultMemoryAccounting''' | On both Debian 10 and CentOS 7, the parameters are set in the '''/etc/systemd/system.conf''' -- uncomment the lines reffering to '''DefaultCPUAccounting''' and '''DefaultMemoryAccounting''', making sure they are set to 'yes'. Then reload the systemd configuration and restart kubelet: | ||
systemctl daemon-reload | systemctl daemon-reload | ||
systemctl restart kubelet | systemctl restart kubelet | ||
That will stop the messages. | That will stop the messages. | ||
Latest revision as of 01:59, 11 August 2019
If you start seeing this type of entry in the system logs:
Aug 10 21:33:50 telmar kubelet[704]: W0810 21:33:50.479159 704 container_manager_linux.go:818] CPUAccounting not enabled for pid: 704 Aug 10 21:33:50 telmar kubelet[704]: W0810 21:33:50.479179 704 container_manager_linux.go:821] MemoryAccounting not enabled for pid: 704
Then (for some reason), the systemd configuration has not enabled the CPU/Memory Accounting, and kubelet doesn't like that ...
On both Debian 10 and CentOS 7, the parameters are set in the /etc/systemd/system.conf -- uncomment the lines reffering to DefaultCPUAccounting and DefaultMemoryAccounting, making sure they are set to 'yes'. Then reload the systemd configuration and restart kubelet:
systemctl daemon-reload systemctl restart kubelet
That will stop the messages.