Servers stuck on PXE boot

Users
1

Hello, I have several devices on my cluster that are stuck on a PXE boot loop. We currently run a cluster with one region controller and four rack controllers. Some work just fine, but some fail to PXE boot.

We’ve been unable to debug this and re-checked settings several times.

rackd.log: https://gist.github.com/ebsalberto/283c04a29cd5466030ee27b51c51fd12
regiond.log: https://gist.github.com/ebsalberto/65b6024d9ee3e6ff43addca9f3573fc6

Tried getting the logs from /var/log/maas/rsyslog/186S025502NODE1 but there are no recent logs from that device, which is weird.

Thanks.

2

I have some servers that will not pxe boot the default bionic kernel. I start with a HWE kernel and switch during phase 1.

TPM can block booting also.

Just 2 ideas

3

also, check the boot order in BIOS. I had some machines with PXE enabled on extra interfaces that cause me no end of grief. And some that had the local SSD configured before PXE, so rescue did not work (or installs if I did not erase the disk).