863839 - create separate slavealloc pool for servo buildbot master/slave

Status: RESOLVED FIXED

(Release Engineering :: General, defect)

Comment 1

[[e-mail] (:bhearsum)]

Attachment: create a servo pool + master

As far as I can tell, this is all we need to create an exclusive pool for the servo master + slaves. I didn't put the datacentre in the name because it seems to me that doing so is a relic of the past, when we didn't do cross-datacentre master<->slave connections.

Once we know the fqdn of the master we can run this and move the slaves into this pool, I think?

I used the existing trust level ("core") and environment ("prod") because adding new ones won't make a difference AFAICT. To be honest, I'm not really sure why we have these when they seem to do the same thing, and are also duplicated in pools...

Comment 2

[Rail Aliiev [:rail]]

Comment on attachment 742491 [details]
create a servo pool + master

You also need an entry in slave_passwords with pooldid=52 and distroid=NULL (NULL if you don't plan to have different passwords for different platforms).

Comment 3

[[e-mail] (:bhearsum)]

Attachment: updated to add password

Comment 4

[Dustin J. Mitchell [:dustin] (he/him)]

You'll actually need a whole 'nother slavealloc instance.

The releng slavealloc instance is a webapp, and not built with puppetagain, so we should probably duplicate that on a new vhost, with different config.

Comment 5

[[e-mail] (:bhearsum)]

(In reply to Dustin J. Mitchell [:dustin] from comment #4)
> You'll actually need a whole 'nother slavealloc instance.

Per IRC from Friday, we're fine existing the existing one.

I just imported the patch w/ the real hostname + password into the production slavealloc db.

We still need to switch the mac slaves over to the servo pool, but I'll take care of that in bug 863278.