Bug 1910070 - KubeVirt VMs, used for masters, are created with too short termination grace period which leads to FS corruption
Summary: KubeVirt VMs, used for masters, are created with too short termination grace ...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: Cloud Compute
Version: 4.7
Hardware: Unspecified
OS: Unspecified
urgent
urgent
Target Milestone: ---
: 4.7.0
Assignee: Nir Argaman
QA Contact: Chen Yosef
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-12-22 14:37 UTC by Chen Yosef
Modified: 2021-01-10 08:14 UTC (History)
1 user (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2021-01-10 08:14:54 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Github openshift installer pull 4511 0 None open Bug 1910070: Set termination_grace_period_seconds to 10 minutes 2021-01-06 05:28:13 UTC

Description Chen Yosef 2020-12-22 14:37:43 UTC
Description of problem:
The installer creates the master VMs without specifying termination grace period, which means using the default 30 seconds. When restarting the VM, the graceful shutdown takes more than that, so after 30 seconds the VM is killed during the shutdown process. This leads in some case (timing related) to XFS metadata corruption that can be recovered only manually. 

Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1. Install a tenant cluster
2. Restart one of the VMs (virtctl restart <vmi>) 
3. Repeat step 2 until the VM won't boot anymore.

Actual results:


Expected results:


Additional info:


Note You need to log in before you can comment on or make changes to this bug.