-
Notifications
You must be signed in to change notification settings - Fork 45
[11/n] [sled-agent] move around ZoneImageFileSource logic #8235
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[11/n] [sled-agent] move around ZoneImageFileSource logic #8235
Conversation
Created using spr 1.3.6-beta.1
Created using spr 1.3.6-beta.1 [skip ci]
Created using spr 1.3.6-beta.1
Created using spr 1.3.6-beta.1 [skip ci]
Created using spr 1.3.6-beta.1 [skip ci]
Created using spr 1.3.6-beta.1 [skip ci]
@@ -2003,7 +2004,7 @@ impl InstanceRunner { | |||
.with_log(self.log.clone()) | |||
.with_underlay_vnic_allocator(&self.vnic_allocator) | |||
.with_zone_root_path(root) | |||
.with_zone_image_paths(&["/opt/oxide".into()]) | |||
.with_file_source(&ramdisk_file_source("propolis-server")) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could ramdisk_file_source
take a ZoneType
or ZoneKind
instead of a bare str?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah no because propolis-server isn't an Omicron zone (what ZoneType
and ZoneKind
represent). Will add a comment here.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ah right. That's unfortunate.
…t error (#8237) For mupdate overrides, in order to be safe, we must know that the data stored in the JSON is consistent with the images stored on disk. We currently apply this logic to install dataset hashes, and will apply it to artifact hashes in the future. Questions: * [x] ~~Should we apply this logic to more critical zones like the switch zone? They're always going to be part of the ramdisk maybe?~~ We unconditionally succeed for RAMdisk zones, and in particular the switch zone, now. * [ ] How does this interact with config-reconciler work? On error, that would permanently be blocked it seems to me. Note: we decide errors per-zone now. * [ ] We ask for the boot zpool to be passed in but also cache it as part of constructing the info... how do we reconcile the relative presence or absence of this info? Depends on #8235.
ZoneImageFileSource
to illumos-utils and use it in the zone image builder.Depends on: