Description of problem: The RPMs of ipxe images doesn't have the new NICs It is necessary to upgrade it to include them and the customers can deploy new compute nodes. Version-Release number of selected component (if applicable): ipxe-bootimgs-20170123-1.git4e85b27.el7_4.1.noarch ipxe-roms-qemu-20170123-1.git4e85b27.el7_4.1.noarch How reproducible: NIC card that it is not include # ethtool -i eno7 driver: tg3 version: 3.137 firmware-version: 5719-v1.43 NCSI v1.2.12.0 expansion-rom-version: bus-info: 0000:af:00.2 supports-statistics: yes supports-test: yes supports-eeprom-access: yes supports-register-dump: yes supports-priv-flags: no Steps to Reproduce: 1. 2. 3. Actual results: Customers are using latest ipxe.efi from http://boot.ipxe.org/ipxe.efi. Then we were able to deploy the compute node Expected results: Update the ipxe-bootimgs-20170123-1.git4e85b27.el7_4.1.noarch ipxe-roms-qemu-20170123-1.git4e85b27.el7_4.1.noarch with the new NICs cards Additional info:
ipxe is delivered by RHEL. OpenStack does not control what is included. Moving this to RHEL
tg3? We've had support for tg3 for over a decade. This isn't about missing a driver. My guess is the system vendor changed the device id on the NIC to be vendor specific, updated the pci id file, but didn't send a patch to ipxe.org. Can you provide: 1) The pci device and vendor id of the NIC in question 2) The console log of ipxe as it boots up
NetXtreme BCM5719 Gigabit Ethernet PCIe vendor/device id : 14e4:1657 You can see the driver is supported for above nic in below link. http://ipxe.org/appnote/hardware_drivers
Can we get an update about this BZ please
Within our version of the ipxe package we have this entry: PCI_ROM(0x14e4, 0x1657, "14e4-1657", "14e4-1657", 0) so we obviously match on that vendor and device tuple. Please provide me a more accurate description of what exactly is going on here, as I requested. If you just want to stand on the assertion that you need the latest version of ipxe instead, I already have a scheduled update bug open for that and we can close/dup this one.
Neil what is the other dup of this one? Let me check and I think I can answer to you. The case attached i mine ;) Thanks!
closing on request from the repoter as a dup of my update bug *** This bug has been marked as a duplicate of bug 1597210 ***