-
Bug
-
Resolution: Duplicate
-
Normal
-
None
-
4.12.0
-
None
-
No
-
False
-
Description of problem:
MCD rollout (drain and reboot) takes a very long time, even with a high `maxUnavailable` value. Want to reduce the boot time by bypassing BOOT POST using "kexec-tools" package. Is it possible by using MCD in openshift 4.x version ? something we can achieve in RHEL7/8 by creating a systemd service: ~~~ # cat /etc/systemd/system/dummy.service [Unit] Description=Perform a kexec-boot [Service] Type=oneshot ExecStart=/usr/sbin/kexec -l /boot/vmlinuz-3.10.0-514.6.2.el7.x86_64 --initrd=/boot/initramfs-3.10.0-514.6.2.el7.x86_64.img --reuse-cmdline ExecStartPost=/usr/bin/systemctl kexec ~~~ 2) # systemctl daemon-reload # systemctl start dummy.service
Version-Release number of selected component (if applicable):
How reproducible:
Steps to Reproduce:
1. 2. 3.
Actual results:
MCD rollout (drain and reboot) takes a very long time, even with a high `maxUnavailable` value
Expected results:
Want to reduce the node boot time.
Additional info:
- duplicates
-
RFE-3067 Using kexec as alternative to full reboot when applying MachineConfig
-
- Backlog
-