Bug 2223987

Summary: CPU Freeze on systems with just one cpu core after latest 1.8 update
Product: [Fedora] Fedora Reporter: customercare
Component: java-1.8.0-openjdkAssignee: jiri vanek <jvanek>
Status: NEW --- QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: unspecified    
Version: 37CC: ahughes, dbhole, jerboaa, jvanek
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: ---
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description customercare 2023-07-19 13:04:52 UTC
2023-07-18T06:01:31+0200 SUBDEBUG Upgrade: java-1.8.0-openjdk-1:1.8.0.372.b07-6.fc37.x86_64
2023-07-18T06:01:32+0200 SUBDEBUG Upgraded: java-1.8.0-openjdk-1:1.8.0.362.b09-2.fc37.x86_64

The systems that had those freezes ( 2 atm ) have in common:

Xen VM
1 cpu core
> 2 GB ram
active java apps
same CPU Freeze message after rebooting with 100% cpu load while system is stuck
freezes started today ~24 after the upgrade
system had a 90+-1 day uptime before, and YEARS in this vm setup before those 90 Days ( which was the upgrade date of the fedora os )

as soon, as we changed the vm's to 2 cpu cores, freezing was gone.

no hints in journal and/or /v/l/messages ( which sometimes survives hard reboots better than the system journal).

Possibilities:

a) it has nothing to do with java/update at all
b) it has to do with the java 1.8 update, as java apps are running constantly.
c) someone messes with the system via network ( kernel dos? )

Please keep this bugreport open for tracking. IF something changes, we get more infos, we will add those here.

Comment 1 customercare 2023-08-10 16:11:43 UTC
After three weeks of java update in action, servers with a lot of java activity just freeze or crash. 

i.e. today at 15:05 tomcat (java was / 2 cores) just froze without any error message and 3 hours later the entire kernel crashed with a cpu freeze in a java process. 

I THINK, as this happend already some years ago, the actual java code messes with the memory management and the kernel ends up in a bad state trying to work throu the mem-structures.

- More memory will delay the process
- more cpu cores compensate the freeze and give opportunities to fix the issue by the kernel itself.