Page MenuHomePhabricator

an-druid1003 Bootlooping: able to reseat/power drain?
Closed, ResolvedPublic

Description

Hello DC Ops,

Per parent ticket, an-druid1003 was rebooted to apply OS updates. Unfortunately, the server has been bootlooping since it was rebooted.

The DRAC and BIOS firmware is up-to-date and I just updated the NIC firmware, but it doesn't seem to have made a difference. I've powered down the host so it doesn't bootloop endlessly.

Are you able to power drain the host and reseat the drives? We are in danger of running out of disk space on the an-druid cluster, recovering this could help mitigate that.

Event Timeline

bking removed bking as the assignee of this task.Fri, Oct 2, 8:02 PM
bking triaged this task as High priority.

Mentioned in SAL (#wikimedia-operations) [2026-10-02T20:54:34Z] <inflatador> bking@an-druid100[4-7] tune2fs -m0 /dev/mapper/vg0-srv T440033

Assiging to @Jclark-ctr per conversation in #wikimedia-dcops IRC.

Change #1351177 had a related patch set uploaded (by Btullis; author: Btullis):

[operations/puppet@production] install_server: Install grub on all disks in druid reuse recipes

https://gerrit.wikimedia.org/r/1351177

Change #1351177 merged by Btullis:

[operations/puppet@production] install_server: Install grub on all disks in druid reuse recipes

https://gerrit.wikimedia.org/r/1351177

I think that this is a preseeding issue, so something that we should be able to sort out.
My reckoning at the moment it that it is because grub is not installed to all of the hard drives. I'm trying out a patch now to see if I can fix it.

Yes, this is fixed now. an-druid1003 is reimaged and back in the cluster.
I'll reimage all of the other druid servers in the cluster, to make sure that they also have grub installed to all of the SSDs in the array.

Change #1351321 had a related patch set uploaded (by Btullis; author: Btullis):

[operations/puppet@production] install_server: Rebuild an-druid100[67] RAID10 across all 8 disks

https://gerrit.wikimedia.org/r/1351321

Change #1351322 had a related patch set uploaded (by Btullis; author: Btullis):

[operations/puppet@production] install_server: Reuse the 8 disk recipe for an-druid100[67]

https://gerrit.wikimedia.org/r/1351322

Change #1351321 merged by Btullis:

[operations/puppet@production] install_server: Rebuild an-druid100[67] RAID10 across all 8 disks

https://gerrit.wikimedia.org/r/1351321

Change #1351322 merged by Btullis:

[operations/puppet@production] install_server: Reuse the 8 disk recipe for an-druid100[67]

https://gerrit.wikimedia.org/r/1351322