Description of problem: manila configuration fails when using manila-backend-vnx.yaml template. does not create manila user, database in mariadb. also the template is missing manila::backend::dellemc_vnx::share_backend_name parameter and sets manila::backend::dellemc_vnx::emc_share_backend incorrectly causing additional errors. May 01 13:53:47 wcmsc5-l-rh-ocld-0 puppet-user[396288]: (/Stage[main]/Manila::Db::Sync/Exec[manila-manage db_sync]) Failed to call refresh: manila-manage db sync returned 1 instead of one of [0] May 01 13:53:47 wcmsc5-l-rh-ocld-0 puppet-user[396288]: (/Stage[main]/Manila::Db::Sync/Exec[manila-manage db_sync]) manila-manage db sync returned 1 instead of one of [0] Version-Release number of selected component (if applicable): openstack-tripleo-heat-templates-7.0.3-22.el7ost.noarch How reproducible: Every time it is deployed Steps to Reproduce: 1. Deploy overcloud using supplied templates. 2. 3. Actual results: Manila fails to deploy as described above. Expected results: Manila deploys properly. Additional info: I've attached the customers manila deployment configuration. I've just reproduced this in my lab environment and Mark Jones was also able to reproduce this issue by attempting to deploy manila in his lab environment.
Can you attach the manila.conf file from the deployment pl
And possibly the environment file used for deployment pl
I don't see the attachment. May be it is marked private
(In reply to Rajini Karthik from comment #4) > I don't see the attachment. May be it is marked private Hi Rajini, The attachment is marked private due to containing customer data. I'm working on sanitizing the files you've requested, I will attach them this afternoon.
I confirmed the two issues described in the first paragraph of the description, and discussed them with Rajini. She will fix malformed emc_shared_backend hiera value, and I will propose a change to puppet-tripleo that will eliminate the need to set the shared_backend_name. Both of those issues are easy to work around, which I see you have done using a custom template. The hiera values can also be set using ExtraConfig, like this: parameter_defaults: ExtraConfig: manila::backend::dellemc_vnx::emc_share_backend: vnx manila::backend::dellemc_vnx::share_backend_name: tripleo_manila_vnx But that doesn't explain the deployment problem, which I was not able to reproduce. I would start by focusing on the "manila-manage db_sync" error. But here are a couple of other points to consider. openstack-tripleo-heat-templates-7.0.3-22.el7ost.noarch shipped in 12z1, and a newer version is available in 12z2. I don't know if that's a factor, but it would be good to test using the latest version(s). I also notice the manila-environment.yaml file in the BZ attachment is setting ManilaVNXNasServer with an IPv6 address, which is not supported in OSP-12 (but is planned for OSP-13). You might try deploying with an IPv4 address, even a bogus one. With a bogus address, the Manila VNX backend service will fail to come up, but the deployment should finish. Here is an example from my own test: (overcloud) [stack@rhos-undercloud ~]$ manila service-list +----+------------------+------------------------------+------+---------+-------+----------------------------+ | Id | Binary | Host | Zone | Status | State | Updated_at | +----+------------------+------------------------------+------+---------+-------+----------------------------+ | 1 | manila-scheduler | hostgroup | nova | enabled | up | 2018-05-12T19:42:07.000000 | | 2 | manila-share | hostgroup@tripleo_manila_vnx | nova | enabled | down | None | +----+------------------+------------------------------+------+---------+-------+----------------------------+
Fix proposed for https://review.openstack.org/#/c/568348/ manila::backend::dellemc_vnx::emc_share_backend
Targeting for RH OSP 13 because there's a workaround for OSP 12 as Alan mentioned prior.
Backport to queens -https://review.openstack.org/#/c/570958/
https://review.openstack.org/#/c/570958/
Hi Rajini, Can Dell EMC perform QA on this? Thanks, Sean
If there is an rpm or something, I can ask the VNX storage team to QA it.
Can we get the rpm to Rajini for testing?
We can verify this bug fix,if you can provide tripleo-heat-template and puppet-* rpms.
Dustin, can you help get Rajini the packages she would need to test? Thanks, Sean
Done! Sent them over to Rajini at her Dell email address. Let me know if you need anything else!
Hi Rajini, I tried to deploy Manila VNX today, but failed, the issue is same as before: -------------------------------------------------- Error: Evaluation Error: Error while evaluating a Function Call, Could not find data item manila::backend::dellemc_vnx::share_backend_name in any Hiera data file and no default supplied at /etc/puppet/modules/tripleo/manifests/profile/base/manila/share.pp:215:41 on node osp12ctl0.localdomain + rc=1 + set -e + '[' 1 -ne 2 -a 1 -ne 0 ']' + exit 1 -------------------------------------------------- My test steps as: 1. Upgrade openstack-tripleo-heat-templates-8.0.3-2.el7ost.noarch.rpm on undercloud 2. Upgrade puppet-tripleo-8.3.4-2.el7ost.noarch.rpm on undercloud 3. Deploy overcloud I think Red Hat should also provide new overcloud image which with new puppet-* rpms installed, the puppet-tripleo version on overcloud as: -------------------------------------------------- [root@osp12ctl0 ~]# rpm -qa|grep tripleo puppet-tripleo-8.3.2-7.el7ost.noarch [root@osp12ctl0 ~]# rpm -qa|grep puppet|grep manila puppet-manila-12.4.0-0.20180329035214.6c18418.el7ost.noarch -------------------------------------------------- Thanks Yong
Hi Dustin This didn’t work. In the past we have tested upstream puppet modules using upload-puppet-script. Is there a better way to test these rpms? Using the upload-puppet-modules script we will be able to update the puppet modules when executing the overcloud deployment. # From the undercloud mkdir puppet-modules cd puppet-modules git clone https://git.openstack.org/openstack/puppet-tripleo tripleo # Edit as needed under the tripleo folder cd git clone https://git.openstack.org/openstack/tripleo-common export PATH="$PATH:tripleo-common/scripts" upload-puppet-modules --directory puppet-modules/ Thanks Rajini
(In reply to Rajini Karthik from comment #34) > Hi Dustin > This didn’t work. In the past we have tested upstream puppet modules using > upload-puppet-script. > Is there a better way to test these rpms? > > Using the upload-puppet-modules script we will be able to update the puppet > modules when executing the overcloud deployment. > # From the undercloud > mkdir puppet-modules > cd puppet-modules > git clone https://git.openstack.org/openstack/puppet-tripleo tripleo Instead of cloning the upstream repo, grab a copy of the files from the updated RPM on the undercloud. Try this after you created the puppet-modules directory: % cp -r /usr/share/openstack-puppet/modules/tripleo ~/puppet-modules You should now have a ~/puppet-modules/tripleo directory that contains the fix. Then run the upload-puppet-modules script, as you described.
I tried the deploy on Yong's env, but FAILED again. Here are the steps: 1. Used the latest heat-templates and puppet-tripleo packages. (undercloud) [stack@manager ~]$ rpm -qa | grep heat-template openstack-tripleo-heat-templates-8.0.3-2.el7ost.noarch (undercloud) [stack@manager ~]$ rpm -qa | grep puppet-tripleo puppet-tripleo-8.3.4-2.el7ost.noarch 2. Ran `upload-puppet-modules` provided by Alan. 3. Deployed the overcloud. But failed. Cinder failure: Error: Evaluation Error: Error while evaluating a Resource Statement, Cinder::Backend::Emc_vnx[tripleo_dellemc_vnx]: has no parameter named 'storage_vnx_pool_names' expects a value for parameter 'storage_vnx_pool_name' at /etc/puppet/modules/tripleo/manifests/profile/base/cinder/volume/dellemc_vnx.pp:42 on node osp12ctl0.localdomain Manila failure: Error: Evaluation Error: Error while evaluating a Resource Statement, Manila::Backend::Dellemc_vnx[tripleo_manila_vnx]: has no parameter named 'network_plugin_ipv6_enabled' has no parameter named 'emc_ssl_cert_verify' has no parameter named 'emc_ssl_cert_path' at /etc/puppet/modules/tripleo/manifests/profile/base/manila/share.pp:215 on node osp12ctl0.localdomain Root causes: I found the puppet-cinder and puppet-manila are out-of-date on the undercloud. (undercloud) [stack@manager ~]$ rpm -qa | grep puppet-cinder puppet-cinder-12.4.1-0.20180329071637.4011a82.el7ost.noarch (undercloud) [stack@manager ~]$ rpm -qa | grep puppet-manila puppet-manila-12.4.0-0.20180329035214.6c18418.el7ost.noarch The fixes for above failures were merged: https://review.openstack.org/#/c/562532/ and https://review.openstack.org/#/c/574504/. My questions: 1. Could you share me the related rpm packages of puppet-cinder and puppet-manila containing above fixes? 2. And after getting these two packages, I need to run `upload-puppet-modules` on them together with puppet-tripleo, right? Thanks! -Ryan
I just emailed the latest puppet-cinder and puppet-manila RPMs for OSP-13 (will ship in 13z2) to Rajini. BTW, I thought Dell EMC's DCI has access to the latest OSP-13 puddles. Is this not in place? If it is, then the latest RPMs should already be available. It will certainly be more expedient to get RPMs from DCI than to request them individually.
This has been tested successfully. I installed the latest puppet-cinder and puppet-manila rpms, and deployed Cinder and Manila SUCCESSFULLY. The version of rpms: openstack-tripleo-heat-templates-8.0.3-2.el7ost.noarch.rpm puppet-tripleo-8.3.4-2.el7ost.noarch.rpm puppet-cinder-12.4.1-0.20180628102250.641e036.el7ost.noarch.rpm puppet-manila-12.4.0-2.el7ost.noarch.rpm (overcloud) [stack@manager ~]$ openstack volume service list +------------------+-------------------------------+------+---------+-------+----------------------------+ | Binary | Host | Zone | Status | State | Updated At | +------------------+-------------------------------+------+---------+-------+----------------------------+ | cinder-scheduler | osp12ctl0 | nova | enabled | up | 2018-08-07T07:55:02.000000 | | cinder-volume | hostgroup@tripleo_dellemc_vnx | nova | enabled | up | 2018-08-07T07:55:02.000000 | +------------------+-------------------------------+------+---------+-------+----------------------------+ (overcloud) [stack@manager ~]$ manila service-list +----+------------------+------------------------------+------+---------+-------+----------------------------+ | Id | Binary | Host | Zone | Status | State | Updated_at | +----+------------------+------------------------------+------+---------+-------+----------------------------+ | 1 | manila-scheduler | hostgroup | nova | enabled | up | 2018-08-07T07:55:26.000000 | | 2 | manila-share | hostgroup@tripleo_manila_vnx | nova | enabled | up | 2018-08-07T07:55:25.000000 | +----+------------------+------------------------------+------+---------+-------+----------------------------+ Thanks! -Ryan
Setting Verified by Partner.
This bug is marked for inclusion in the errata but does not currently contain draft documentation text. To ensure the timely release of this advisory please provide draft documentation text for this bug as soon as possible. If you do not think this bug requires errata documentation, set the requires_doc_text flag to "-". To add draft documentation text: * Select the documentation type from the "Doc Type" drop down field. * A template will be provided in the "Doc Text" field based on the "Doc Type" value selected. Enter draft text in the "Doc Text" field.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2018:2574