Bug 1928426 - Assisted Installer fails with OCP 4.7 pre-GA provided build
Summary: Assisted Installer fails with OCP 4.7 pre-GA provided build
Keywords:
Status: CLOSED DUPLICATE of bug 1926345
Alias: None
Product: OpenShift Container Platform
Classification: Red Hat
Component: assisted-installer
Version: 4.7
Hardware: x86_64
OS: Linux
unspecified
unspecified
Target Milestone: ---
: ---
Assignee: Michael Filanov
QA Contact: Yuri Obshansky
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2021-02-13 23:25 UTC by JP Jung
Modified: 2021-02-16 19:14 UTC (History)
4 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2021-02-16 19:14:28 UTC
Target Upstream Version:
Embargoed:
odepaz: needinfo-


Attachments (Terms of Use)

Comment 1 Osher De Paz 2021-02-14 07:56:16 UTC
@jjung seems like the nodes have been messed up between master and worker roles:
lab2_worker_worker2
lab2_worker_master3
lab2_master_worker1
lab2_master_master2
lab2_bootstrap_master1

Maybe it's worth setting the roles manually and reinstalling

Comment 2 Osher De Paz 2021-02-14 14:01:40 UTC
@jjung in any case you're right that we should consider the node names when automatically assigning roles
right now we only do it based on resource considerations

Comment 3 Michael Filanov 2021-02-14 14:46:49 UTC
@odepaz regardless of the roles, do we need to update to a different version?

Comment 4 Osher De Paz 2021-02-14 15:27:01 UTC
@lgamliel ?

Comment 5 JP Jung 2021-02-14 20:40:06 UTC
I can redo an installation specifying the roles / server, but the allocation error is not a major issue, as the worker1 (acting as master) and master 3 (acting as worker) that were swapped have exactly the same hardware. So it is not failing because one of the masters is below minimum hardware specs (e.g. worker 2 is only 2 CPUs and 8GB RAM).

Also the default allocation on the same hardware with Assisted Install toward 4.6.8 does pick up the right nodes with the right names (but maybe it was just luck?).

I'll wait for instructions to keep debugging this install, or to wipe it and redo an install. Let me know.

Comment 6 Osher De Paz 2021-02-16 11:55:14 UTC
Hey, sorry I missed your message
Can you do 'oc adm must-gather' against the created cluster?
If not, it will help having the console operator logs

Comment 7 JP Jung 2021-02-16 15:10:53 UTC
Hi,

Created must-gather.tar.gz
Attachement is 139MB: https://1drv.ms/u/s!AoJ0Ren1qyzsgr9lAaMlhK1d57o4hw?e=YygaGL

Let me know if you require further information.
Regards,

JP Jung

Comment 8 Igal Tsoiref 2021-02-16 19:13:59 UTC
we opened this issue : https://bugzilla.redhat.com/show_bug.cgi?id=1926345

Comment 9 Igal Tsoiref 2021-02-16 19:14:28 UTC

*** This bug has been marked as a duplicate of bug 1926345 ***


Note You need to log in before you can comment on or make changes to this bug.