| avobridge | <alvaro.soto> I hope this channel gets more movement when OpenInfra Summit starts hehehe | 16:51 |
|---|---|---|
| prometheanfire | I'm working on a test install for proof of concept now, once security approves the oan vlan to access the ipmi vlan | 17:38 |
| avobridge | <alvaro.soto> it's always the network hahaha | 17:40 |
| avobridge | <alvaro.soto> hey, anybody working with raspberry pi? | 17:40 |
| prometheanfire | not with starlingx, also hard to get them :P | 17:42 |
| avobridge | <alvaro.soto> =( | 17:43 |
| prometheanfire | if running a pxe network / vlan, does starlingx run a dhcp server with the right options/etc? | 19:02 |
| khyr0n | any network configuration to used as pxe boot tagged or untagged should run as native | 19:18 |
| khyr0n | and, remember that disable negotiation with the switch or pxe boot timeout could end first than the negotiation with the switch | 19:19 |
| khyr0n | 's/that/to/g' | 19:20 |
| ildikov | hi all! | 20:11 |
| ildikov | Great to hear you're experimenting with the platform :) | 20:11 |
| ildikov | I don't know the answer to the PXE question from the top of my head, I'll see if I can figure it out | 20:12 |
| ildikov | OutBackDingo: do you know the answer to the above PXE question? | 20:12 |
| prometheanfire | ildikov: khyr0n looks like we are going to be taging the management network (with pxe functions on the management network) to the controller nodes and having it be the native vlan for the compute nodes | 20:14 |
| khyr0n | yup, that sounds the best option for it... I've use it on openstack / ceph / kubernetes production deployments, and change the native vlan id it's always the best way to pxe boot nodes | 20:16 |
| ildikov | prometheanfire: +1 | 20:34 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!