Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

idrac_interfaces.yml has wrong boot order #92

Closed
bengland2 opened this issue Mar 17, 2020 · 3 comments
Closed

idrac_interfaces.yml has wrong boot order #92

bengland2 opened this issue Mar 17, 2020 · 3 comments

Comments

@bengland2
Copy link

@bengland2 bengland2 commented Mar 17, 2020

Your System Details
alias cloud02

  • Python Version: 3
  • Operating System: RHEL8/CoreOS
  • Target System Type: (e.g. Dell, SuperMicro) Dell 740xd
  • IPMI / Out-of-band Firmware Version: (_e.g. iDRAC 8 2.60.60.60)

Describe the bug
A clear and concise description of what the bug is.

I tried to run Dustin Black's playbook and it fails (log here). In progress of troubleshooting, I realized that the "director" interface that I was supposed to use was "eno1", but I was using ens7f0. But this means I can't use the 25-GbE interface for provisioning network, and I can't use both 25-GbE NIC ports. @dustinblack thought this was wrong, that "director_740xd_interfaces" should have the Foreman interface "eno3" last, so that any of the other NIC ports could PXE boot, right Dustin? This would give us maximum flexibility. But what it does in director_740_interfaces record is make the Foreman interface 2nd, after eno1 interface. When PXE boot gets to the Foreman interface, Foreman will tell it to boot from local drive, so subsequent interfaces will not get used. so the only other interface that we can use for PXE boot is eno1, which is the 10-GbE interface in Alias. If my understanding of the PXE boot sequence is correct, can we please change this?

To Reproduce / What were you Doing?
Steps to reproduce the behavior:

  1. See Dustin Black's IPI directions here
  2. edit inventory/hosts as shown here
@QuantumPosix

This comment has been minimized.

Copy link
Contributor

@QuantumPosix QuantumPosix commented Mar 25, 2020

The suggestion is to put in the following line for director - this has no conflicts with anything else (all high speed interfaces first, then foreman, then hard disk):
director_740xd_interfaces: NIC.Integrated.1-1-1,NIC.Integrated.1-2-1,NIC.Slot.7-1-1,NIC.Slot.7-2-1,NIC.Integrated.1-3-1,HardDisk.List.1-1

@bengland2

This comment has been minimized.

Copy link
Author

@bengland2 bengland2 commented Mar 25, 2020

Nitpick: It's not really necessary IMO to include both NIC.Slot.7-1-1 and 7-2-1, since they are equivalent, either one of them would do (as long as we are consistent), and having just 1 of them reduces the time spent timing out PXE boot interfaces. The reasoning here is the same as why we don't include eno2=NIC.Integrated.1.2.1 in the list. but I don't want that to get in the way of progress. Because eliminating need for a boot order change would significantly reduce time and complexity of an OCP4 UPI install in Alias. Would nodes arrive at a reservation with -t director set? e-mail from @sadsfae (Will) earlier e-mail seemed to suggest this.

@sadsfae

This comment has been minimized.

Copy link
Member

@sadsfae sadsfae commented Mar 25, 2020

@QuantumPosix has resolved this with #94.

Additionally a new badfish docker image is built with these changes so you'll want to docker pull quads:latest or the podman equivalent.

@sadsfae sadsfae closed this Mar 25, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Linked pull requests

Successfully merging a pull request may close this issue.

None yet
3 participants
You can’t perform that action at this time.