Bug 1810031 - [BAREMETAL] ocs-operator goes in CrashLoopBackOff State if storageclass is not selected while deploying ocs-storagecluster from UI
Summary: [BAREMETAL] ocs-operator goes in CrashLoopBackOff State if storageclass is no...
Keywords:
Status: CLOSED DUPLICATE of bug 1811003
Alias: None
Product: Red Hat OpenShift Container Storage
Classification: Red Hat Storage
Component: management-console
Version: 4.3
Hardware: Unspecified
OS: Unspecified
high
high
Target Milestone: ---
: ---
Assignee: Bipul Adhikari
QA Contact: Elad
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2020-03-04 12:54 UTC by Pratik Surve
Modified: 2020-03-20 07:58 UTC (History)
8 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2020-03-20 07:58:30 UTC
Embargoed:

Attachments (Terms of Use)

Comment 3 Nishanth Thomas 2020-03-05 12:41:48 UTC 
prsurve, Does this happens only is baremetal case?
Comment 6 Nishanth Thomas 2020-03-06 12:51:40 UTC 
Thanks, prsurve

This feature is being worked on and targeted for 4.5
Moving out.
Comment 7 Raz Tamir 2020-03-06 18:34:00 UTC 
Hi Nishanth,

I understand we can't do changes in the UI for 4.3 at this point but this bug is too severe and have to be fixed in 4.3. 

Maybe we can set a default SC in case nothing was selected from the UI? 

(moving back to 4.3 until we'll have conclusion)
Comment 8 Sahina Bose 2020-03-09 10:30:11 UTC 
(In reply to Raz Tamir from comment #7)
> Hi Nishanth,
> 
> I understand we can't do changes in the UI for 4.3 at this point but this
> bug is too severe and have to be fixed in 4.3. 
> 
> Maybe we can set a default SC in case nothing was selected from the UI? 
> 
> (moving back to 4.3 until we'll have conclusion)

On Baremetal environments - OCS cluster can only be created via the documented steps - i.e manually creating a cluster yaml from CLI. Creating from UI is not supported.
What's the expectation here?
Comment 9 Raz Tamir 2020-03-10 09:02:35 UTC 
Hi Sahina,

The documented steps are for the device discovery to happen (LSO) and the deployment should be from the UI like any other platform - that's the expectation.
allowing customers (intentionally) to deploy from CLI was considered in 4.2 as a NO GO from Dev side so I don't see any reason to change this decision now
Comment 11 Sahina Bose 2020-03-10 10:28:18 UTC 
(In reply to Sahina Bose from comment #10)
> (In reply to Raz Tamir from comment #9)
> > Hi Sahina,
> > 
> > The documented steps are for the device discovery to happen (LSO) and the
> > deployment should be from the UI like any other platform - that's the
> > expectation.
> > allowing customers (intentionally) to deploy from CLI was considered in 4.2
> > as a NO GO from Dev side so I don't see any reason to change this decision
> > now
> 
> For baremetal deployment - deploying from CLI is the only option that Dev
> has acked and worked on.
> Not sure where the disconnect is coming from, but the UI is a 4.5 item at
> this point.

To clarify, the documented steps are not just for the device discovery - we do have the manual steps to change MONs to use hostPath, provide the size of OSD based on attached device sizes as documented in https://issues.redhat.com/browse/RHSTOR-824
Comment 12 Raz Tamir 2020-03-10 12:25:19 UTC 
Thanks for clarification Sahina,

Moving back to 4.5
Comment 13 Nishanth Thomas 2020-03-11 12:32:14 UTC 
Cancelling the needinfo on me
Comment 14 Afreen 2020-03-20 07:58:30 UTC 
Possible duplicate of https://bugzilla.redhat.com/show_bug.cgi?id=1811003
The UI is same for any platform.

*** This bug has been marked as a duplicate of bug 1811003 ***
Note You need to log in before you can comment on or make changes to this bug.