Release Notes

Cause:
Red Hat Enterprise Linux 7.3 kernel is still following ACPI-5.0 specification, while BIOS is already moved
to ACPI-6 specification. Because of this upgrade in version, FADT table is not recognized by Red Hat
Enterprise Linux 7.3 GA OS, which is still following ACPI-5.0 specification.
Workaround: There is no workaround that is currently available for this issue.
Red Hat Enterprise Linux 7.3 displays a mcelog error
message
Description: After successful installation and boot into Red Hat Enterprise Linux 7.3 GA in UEFI mode, the following
error message is displayed in /var/log/messages:
mcelog: Family 6 Model 9e CPU: only decoding architectural errors
Applies to: Red Hat Enterprise Linux 7.3
Cause:
The error is due to the mcelog package. This package does not support Kabylake processor. This issue will
be fixed with the upstream patch.
Workaround: There is no workaround that is currently available for this issue.
IPv6 UEFI PXE installation fails
Description:
IPv6 UEFI PXE installation fails. The system can get the DHCP address but will not display the boot menu
to select an OS but rather drops to the Linux grub prompt.
Applies to: Red Hat Enterprise Linux 7.3
Cause:
bootloader.efi is hardlinked grubx64.efi
Due to hard link grub is not able to load boolloader.efi file and failing to grub shell.
Workaround: There is no workaround that is currently available for this issue.
Intel x520 FCoE first boot fails with kernel crash
Description:
After installing Red Hat Enterprise Linux 7.2 on FCoE LUN using Intel-X520 Card, the first boot fails with
Kernel Crash.
Applies to: Red Hat Enterprise Linux 7.2
Cause: The OS fails to select an FCF due to which an FLOGI is not sent. The reason for failure to select an FCF is
that FCF selection is getting reset every time an unsolicited multicast advertisement arrives from switch
with sub two second interval.
Workaround: There is no workaround that is currently available for this issue.
irqbalance spam messages with banned CPU conflict
Description:
irqbalance is spamming messages every 10 seconds with snippet below. The HFI driver is setting affinity
and the Omni Fabric install is setting /etc/sysconfig/irqbalance to use exact.
IRQBALANCE_ARGS= --hintpolicy=exact
32 Known issues