Bug 1595288

Raised bug 1598344, in 3.1 for issue observed in comment #4

Hi Erin,

Ignore my previous comment #9.

The doc text from comment #8, lgtm 

"The 'iSCSI Overview' Grafana page does not behave correctly When using the Red Hat Ceph Storage metrics Dashboard, the 'iSCSI Overview' Grafana page does not display any graphs or values as it is expected to"

Hi guys,

The issue seems to be that cephmetrics-ansible is expecting a [iscsis] section in ansible inventory file while the RHCS block device guide states this section should be called [iscsigws]. Adding a [iscsis] section is enough to have the prometheus-node-exporter package correctly installed on iSCSI targets and metrics recorded and exposed on iSCSI Grafana dashboards.

You can refer to these 2 commits from Zack:
https://github.com/ceph/cephmetrics/commit/a99fdea00f19bedb8093484330a34619d7e682dd
https://github.com/ceph/cephmetrics/commit/9b5dafd45d22769fb6c33a93df2d31d363d6c673

And the RHCS block device guide: 
https://access.redhat.com/documentation/en-us/red_hat_ceph_storage/3/html-single/block_device_guide/#iscsi-config-target-ansible

Regards,
Frédéric.

Hi Frederic,

great catch! We should probably fix that for 3.2 or document your workaround. Zack, what do you think?

Regards,
Boris

Zack would you please let us know the status of this?

*** Bug 1656053 has been marked as a duplicate of this bug. ***

*** Bug 1515153 has been marked as a duplicate of this bug. ***

Hello ,

Regarding case : #02337240

Customer is facing the same issue. May I please know the status of this bug. Or can we provide any workaround possibly?

Waiting for response!

Thanks

Regards,
Servesha

A workaround would be to have both [iscsis] as well as [iscsigws] sections in the ansible hosts file.

Hello Boris ,

Thank you for the reply. I will suggest this workaround to the customer.

Regards,
Servesha

Hello Boris,

I have a question. If we mention both [iscsis] as well as [iscsigws] in ansible hosts file, customer will need to run site.yml (or site-docker.yml) playbook. Playbook can be run using --limit option. So, the question is, when using --limit option what command will be right?

#ansible-playbook site.yml --limit  iscsigws  OR

#ansible-playbook site.yml --limit iscsis

Awaiting response!

Regards,
Servesha

Hello,

I would say this one: ansible-playbook site.yml --limit iscsis,iscsigws

Frédéric.

If you are talking site.yml as in ceph-ansible then that should support iscsigws. For the cephmetrics-ansible (playbook.yml), you should use iscsis. Alternatively, you can use the syntax as suggested in the above comment, that should be the safest option and can be used for both ceph-ansible as well as cephmetrics-ansible.

Hello Boris,

The customer was already aware of work around. And I asked if it is worked for him. But apparently it is not working for him. May I please know if we have some other way to provide to the customer? 

Thanks,

Best Regards,
Servesha

Hey Servesha,

they only need to re-deploy cephmetrics after they have both iscsis and iscsigws in their ansible/hosts file. Re-deploying cephmetrics is generally less heavy than doing a full ceph deployment. It would be nice if they could re-run the full cephmetrics-ansible deployment scripts (without any --limit options). If that won't work then I would suggest purging the old cephmetrics installation and then doing a full re-deployment.

The deployment playbook should be located at

/usr/share/cephmetrics-ansible/playbook.yml

The purge playbook should be located at

/usr/share/cephmetrics-ansible/purge.yml

Regards,
Boris

Hello Boris,

I will check with the customer regarding purging the old cephmetrics installation. Thank you for your reply.

Best Regards,
Servesha

Updating the QA Contact to a Hemant. Hemant will be rerouting them to the appropriate QE Associate. 

Regards,
Giri

Updating the QA Contact to a Hemant. Hemant will be rerouting them to the appropriate QE Associate. 

Regards,
Giri