Bug 1397720

Summary: Can't import the exist ceph cluster via storage console web UI
Product: Red Hat Storage Console Reporter: liuwei <wliu>
Component: Ceph IntegrationAssignee: Nishanth Thomas <nthomas>
Status: CLOSED EOL QA Contact: sds-qe-bugs
Severity: high Docs Contact:
Priority: high    
Version: 2CC: ceph-eng-bugs, cylopez, fabian, gmeno, mhackett, mkarnik, nthomas, sankarshan, stuartjames, tcole
Target Milestone: ---Flags: gmeno: needinfo? (nthomas)
Target Release: 3   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
the error picture none

Description liuwei 2016-11-23 09:07:50 UTC
Created attachment 1223075 [details]
the error picture

Description of problem:
According to the "Red Hat Storage Console 2.0 Quick Start Guide ", the customer can't import the exist ceph cluster .   The error is below:

Failed to retrive cluster information from the selected host 'rhcs hostname'. Please select a monitor host and try again.

For the explore debug messages, the error is below:

Failed to load resource: the server responded with a status of 500 (Internal Server Error) 

"Error getting import cluster details"

The guide link is below:

https://access.redhat.com/documentation/en/red-hat-storage-console/2.0/single/quick-start-guide

Version-Release number of selected component (if applicable):

OS version: RHEl7.2/7.3

ceph rpm version:

1 ceph storage console side:

rpm -qa |grep ceph-
ceph-ansible-1.0.5-45.el7scon.noarch
ceph-installer-1.0.15-2.el7scon.noarch
rhscon-ceph-0.0.43-1.el7scon.x86_64

2 ceph storage monitor side:

rpm -qa |grep ceph-
ceph-common-10.2.2-41.el7cp.x86_64
ceph-test-10.2.2-41.el7cp.x86_64
ceph-ansible-1.0.5-34.el7scon.noarch
ceph-osd-10.2.2-41.el7cp.x86_64
ceph-selinux-10.2.2-41.el7cp.x86_64
ceph-mon-10.2.2-41.el7cp.x86_64
ceph-base-10.2.2-41.el7cp.x86_64
rpm -qa |grep calamari-
calamari-server-1.4.9-1.el7cp.x86_64
How reproducible:

Follow the guide, 100% reproduced.

Steps to Reproduce:
1.
2.
3.

Actual results:

import the exist ceph cluster information failure

Expected results:

import it successfully 

Additional info:

Comment 7 Tupper Cole 2017-01-25 09:19:18 UTC
I'm sitting on site with a customer, hitting the same issue. The inability to perform the *primary task of RHSC* is a pretty big blocker, no? I would expect some kind of discussion or workaround in the BZ notes. As is, I'm staring at a customer who would like to use RHSC and Ceph, but can't. I hope this doesn't push him to the competing product he is evaluating, which has a working GUI management system.

Comment 10 Cyril Lopez 2017-03-27 09:07:01 UTC
+1 

please share the password ceph-console is expecting for calamari-light user :




Mar 27 05:05:59 ceph1-mon calamari-lite: 172.20.1.16 - - [2017-03-27 04:05:59] "GET /api/v2/cluster?format=json HTTP/1.1" 403 242 0.206186
Mar 27 05:05:59 ceph1-mon calamari-lite: 172.20.1.16 - - [2017-03-27 04:05:59] "GET /api/v2/auth/login/ HTTP/1.1" 200 552 0.029658
Mar 27 05:05:59 ceph1-mon calamari-lite: 172.20.1.16 - - [2017-03-27 04:05:59] "POST /api/v2/auth/login/ HTTP/1.1" 401 462 0.118966
Mar 27 05:05:59 ceph1-mon calamari-lite: 172.20.1.16 - - [2017-03-27 04:05:59] "GET /api/v2/cluster?format=json HTTP/1.1" 403 242 0.003048

Comment 12 Fabian Dammekens 2017-03-31 08:41:45 UTC
It seems the default 'admin' user is not created, create it using calamari-ctl add_user and you should be able to continue the import process. For the record, the password should also be 'admin'.

Comment 13 Stuart James 2017-05-25 15:06:18 UTC
To fix this do the following on all the monitor nodes

yum install calamari-server
calamari-ctl clear --yes-i-am-sure
calamari-ctl initialize --admin-username admin --admin-password admin --admin-email RTFM


From my testing you must then select the monitor node that has Calamari-server installed otherwise the cluster import fails, this brings me to the question what if this monitor node(with calamari-server) fails how does RHSCON handle this scenario, i am assuming it relies on the calamari-server for its associated cluster metrics, assuming you did actually install calamari-server on all monitor nodes does RHSCON know to utilize the others in the event the one you associated with during import fails?

https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/2/html-single/installation_guide_for_red_hat_enterprise_linux/#calamari_server_installation

Comment 14 Stuart James 2017-05-25 20:34:57 UTC
Ive tested this further and there is single point of failure here, the Monitor node you select (you are restricted to only select one when you import) becomes a single point of failure within the RHSCON interface(it is used exclusively for Calamari server), this does not relate to original ticket i will raise a new one.


i have created bug 1455693

Comment 17 Shubhendu Tripathi 2018-11-19 05:43:51 UTC
This product is EOL now