Nodes don't boot from PXE after environment reset

[Max Lvov]

When I reset environment and all my 12 nodes going to reboot simultaneous only few nodes boot from PXE (in my case every time it's 2 same nodes) Only if I reboot them partially or one by one it works well.
Nodes are hardware supermicro servers.
MOS 6.1

[Maciej Kwiek (maciej-iai)]

Please provide diagnostic snapshot.

[Max Lvov (usrleon)]

Diagnostic snapshot from what moment? Right after reset?

[Andrey Danin (gcon-monolake)]

the reset time + 5 minutes. Or any snapshot but mention when the last reset occurred.

[Max Lvov (usrleon)]

Snapshot taken after reset and after all nodes was brought back online by rebooting them by 2-3 at once. Logs should content enough traces of PXE activity.
https://drive.google.com/file/d/0BxK3hv8QpqTSNTNJay1iM3JGQVE/view?usp=sharing

[Max Lvov (usrleon)]

This is what I see every time in ipmi

[Vitaly Sedelnik (vsedelnik)]

Won't Fix for 6.1-updates because of Medium importance. Moved to 8.0 and assigned to Linux team.

[Alexei Sheplyakov (asheplyakov)]

Please attach dnsmasq logs (dockerctl shell cobbler cat /var/log/dnsmasq.log)

[Alexei Sheplyakov (asheplyakov)]

The attached dnsmasq log contains no records regarding DHCP requests. This looks somewhat strange.
Usually dnsmasq logs the requests including the ones it was unable to process properly.

[Andrey Maximov (maximov)]

guys, please provide details why do you think mos-linux team can help with this bug. what component doesn't work.

[Dina Belova (dbelova)]

It's in incomplete for a more than a month. No more details were provided. Moving to invalid.