0

we have 5 Linux rhel machines , we noticed that all machines was hang with the same time

from messages file we can se the following

Dec 29 19:54:25 localhost kernel: pci 0000:ff:12.4: BAR 4: failed to assign [mem size 0x00000040]
Dec 29 19:54:25 localhost kernel: pci 0000:ff:12.0: BAR 1: failed to assign [mem size 0x00000010]

is this kernel messages are maybe the reason for linux crash/hang?

according to redhat , they recommended to update the kernel version ( https://access.redhat.com/solutions/2772311 )

but they not mention anything about linux crash/hang

King David
  • 549
  • 6
  • 20

1 Answers1

1

Not every problem is a crash, hang, or security issue. Or, what you found might not be related to the root cause of your problem.

Per that Red Hat KB 2772311, you are affected if:

  • Red Hat Enterprise Linux 7
  • Intel Broadwell-EP CPU
  • kernel version is older than 3.10.0-514.el7

That log message means the kernel warned about PCI base address registers out of specification. Red Hat notes that Intel has noticed no functional problems.

Update your kernel, plus any other important updates. Unknown if it will fix your hang problem. However, updates definitely have security and functional fixes, including this minor thing.

John Mahowald
  • 32,050
  • 2
  • 19
  • 34