10:16 a.m.
If I am trying to familiarize myself with the newer containerized
overcloud (Queens & upwards), is there a way to build a TripleO
Quickstart on a single node? As I say, it's mainly to get some
experience in the newer configurations, after which I could probably
grab some more baremetal and do proper setups.
I do find myself wondering, though, with the way Quickstart worked with
images, if it's been doing containers all along...
12:03 p.m.
Пятница, 9 февраля 2018, 19:17 +03:00 от James LaBarre
<jlabarre(a)redhat.com>:
If I am trying to familiarize myself with the newer containerized
overcloud (Queens & upwards), is there a way to build a TripleO
Quickstart on a single node?
Due to I cannot any longer respond from
hotmail.com ( backup address is used )
1. https://www.anstack.com/blog/2018/01/05/tripleo-quickstart-cheatsheet.html
2. this one was done via F27 connection to VIRTHOST
http://lxer.com/module/newswire/view/252059/index.html
I believe it might work in the way suggested by Carlos Camacho (1)
Boris.
As I say, it's mainly to get some
experience in the newer configurations, after which I could probably
grab some more baremetal and do proper setups.
I do find myself wondering, though, with the way Quickstart worked with
images, if it's been doing containers all along...
_______________________________________________
users mailing list
users(a)lists.rdoproject.org
http://lists.rdoproject.org/mailman/listinfo/users
To unsubscribe: users-unsubscribe(a)lists.rdoproject.org
С уважением,
Boris Derzhavets
dba477(a)list.ru
4:39 a.m.
Mentioned bellow my deployment (4 Node HA) via F27 WKS, which worked fine twice on 02/06
and 02/09 now fails with message in log :
2018-02-11 10:19:50 | 2018-02-11 10:19:27Z [overcloud]: CREATE_FAILED Resource CREATE
failed: Error:
resources.AllNodesDeploySteps.resources.ControllerDeployment_Step1.resources[0]:
Deployment to server failed: deploy_status_code: Deployment exited with non-zero status
code: 2
2018-02-11 10:19:50 |
2018-02-11 10:19:50 | Stack overcloud CREATE_FAILED
2018-02-11 10:19:50 |
2018-02-11 10:19:50 | overcloud.AllNodesDeploySteps.ControllerDeployment_Step1.0:
2018-02-11 10:19:50 | resource_type: OS::Heat::StructuredDeployment
2018-02-11 10:19:50 | physical_resource_id: a9db44c8-39f7-4bc6-80b8-534b2c402b48
2018-02-11 10:19:50 | status: CREATE_FAILED
2018-02-11 10:19:50 | status_reason: |
2018-02-11 10:19:50 | Error: resources[0]: Deployment to server failed:
deploy_status_code : Deployment exited with non-zero status code: 2
2018-02-11 10:19:50 | deploy_stdout: |
2018-02-11 10:19:50 | ...
2018-02-11 10:19:50 | "2018-02-11 10:19:11,526 INFO: 22569 -- Removing
container: docker-puppet-heat_api_cfn",
2018-02-11 10:19:50 | "2018-02-11 10:19:11,558 INFO: 22569 -- Finished
processing puppet configs for heat_api_cfn",
2018-02-11 10:19:50 | "2018-02-11 10:19:11,559 ERROR: 22567 -- ERROR
configuring swift_ringbuilder"
2018-02-11 10:19:50 | ]
2018-02-11 10:19:50 | }
2018-02-11 10:19:50 | to retry, use: --limit
@/var/lib/heat-config/heat-config-ansible/a25a263b-579b-4954-bf09-25bda1c4f1c3_playbook.retry
2018-02-11 10:19:50 |
2018-02-11 10:19:50 | PLAY RECAP
*********************************************************************
2018-02-11 10:19:50 | localhost : ok=6 changed=2 unreachable=0
failed=1
2018-02-11 10:19:50 |
2018-02-11 10:19:50 | (truncated, view all with --long)
2018-02-11 10:19:50 | deploy_stderr: |
2018-02-11 10:19:50 |
2018-02-11 10:19:51 | + status_code=1
2018-02-11 10:19:51 | + openstack stack list
2018-02-11 10:19:51 | + grep -q overcloud
2018-02-11 10:20:20 | + openstack stack list
2018-02-11 10:20:20 | + grep -Eq '(CREATE|UPDATE)_COMPLETE'
Thanks
Boris
________________________________
From: users <users-bounces(a)lists.rdoproject.org> on behalf of Boris
<dba477(a)list.ru>
Sent: Friday, February 9, 2018 9:03 PM
To: James LaBarre
Cc: users(a)lists.rdoproject.org
Subject: Re: [rdo-users] Quickstart and containers
Пятница, 9 февраля 2018, 19:17 +03:00 от James LaBarre <jlabarre(a)redhat.com>:
If I am trying to familiarize myself with the newer containerized
overcloud (Queens & upwards), is there a way to build a TripleO
Quickstart on a single node?
Due to I cannot any longer respond from hotmail.com ( backup address is used )
1. https://www.anstack.com/blog/2018/01/05/tripleo-quickstart-cheatsheet.html
2. this one was done via F27 connection to VIRTHOST
http://lxer.com/module/newswire/view/252059/index.html
I believe it might work in the way suggested by Carlos Camacho (1)
Boris.
As I say, it's mainly to get some
experience in the newer configurations, after which I could probably
grab some more baremetal and do proper setups.
I do find myself wondering, though, with the way Quickstart worked with
images, if it's been doing containers all along...
_______________________________________________
users mailing list
users@lists.rdoproject.org<mailto:users@lists.rdoproject.org>
http://lists.rdoproject.org/mailman/listinfo/users
To unsubscribe:
users-unsubscribe@lists.rdoproject.org<mailto:users-unsubscribe@lists.rdoproject.org>
С уважением,
Boris Derzhavets
dba477(a)list.ru
2:07 a.m.
New subject: Quickstart and containers (2)
Attempting via F27 connection to VIRTHOST (32 GB) with 770 MB is Swap
4 Node HA case. Sitill fails with control_memory=7700
ep4.0]: SIGNAL_IN_PROGRESS Signal: deployment f7493897-9e88-464c-bb78-71c2c5fa164f
succeeded
2018-02-12 07:47:10 | 2018-02-12 07:46:45Z [overcloud.AllNodesDeploySteps]: CREATE_FAILED
Resource CREATE failed: Error: resources.ControllerDeployment_Step4.resources[2]:
Deployment to server failed: deploy_status_code: Deployment exited with non-zero status
code: 2
2018-02-12 07:47:10 | 2018-02-12 07:46:46Z
[overcloud.AllNodesDeploySteps.ControllerDeployment_Step4.0]: CREATE_COMPLETE state
changed
2018-02-12 07:47:10 | 2018-02-12 07:46:46Z [overcloud.AllNodesDeploySteps]: CREATE_FAILED
Error: resources.AllNodesDeploySteps.resources.ControllerDeployment_Step4.resources[2]:
Deployment to server failed: deploy_status_code: Deployment exited with non-zero status
code: 2
2018-02-12 07:47:10 | 2018-02-12 07:46:47Z [overcloud]: CREATE_FAILED Resource CREATE
failed: Error:
resources.AllNodesDeploySteps.resources.ControllerDeployment_Step4.resources[2]:
Deployment to server failed: deploy_status_code: Deployment exited with non-zero status
code: 2
2018-02-12 07:47:10 | 2018-02-12 07:46:51Z
[overcloud.AllNodesDeploySteps.ControllerDeployment_Step4.1]: SIGNAL_IN_PROGRESS Signal:
deployment 51c7e7c9-ae7b-41e8-ba62-6767c9730986 failed (2)
2018-02-12 07:47:10 |
2018-02-12 07:47:10 | Stack overcloud CREATE_FAILED
2018-02-12 07:47:10 |
2018-02-12 07:47:10 | overcloud.AllNodesDeploySteps.ControllerDeployment_Step4.1:
2018-02-12 07:47:10 | resource_type: OS::Heat::StructuredDeployment
2018-02-12 07:47:10 | physical_resource_id: 51c7e7c9-ae7b-41e8-ba62-6767c9730986
2018-02-12 07:47:10 | status: CREATE_FAILED
2018-02-12 07:47:10 | status_reason: |
2018-02-12 07:47:10 | Error: resources[1]: Deployment to server failed:
deploy_status_code : Deployment exited with non-zero status code: 2
2018-02-12 07:47:10 | deploy_stdout: |
2018-02-12 07:47:10 | ...
2018-02-12 07:47:10 | "a2b168fb9a3d: Pull complete",
2018-02-12 07:47:10 | "5e1193f920e0: Pull complete",
2018-02-12 07:47:10 | "Digest:
sha256:36ca9874c8359103a574ee7311688cdbefa61bec6a8fa5b728add93dc614d7c8"
2018-02-12 07:47:10 | ]
2018-02-12 07:47:10 | }
2018-02-12 07:47:10 | to retry, use: --limit
@/var/lib/heat-config/heat-config-ansible/fa4dfb68-f59a-420d-88f9-1d12528284a3_playbook.retry
2018-02-12 07:47:10 |
2018-02-12 07:47:10 | PLAY RECAP
*********************************************************************
2018-02-12 07:47:11 | Heat Stack create failed.
2018-02-12 07:47:11 | Heat Stack create failed.
2018-02-12 07:47:11 | localhost : ok=6 changed=2 unreachable=0
failed=1
2018-02-12 07:47:11 |
2018-02-12 07:47:11 | (truncated, view all with --long)
2018-02-12 07:47:11 | deploy_stderr: |
2018-02-12 07:47:11 |
2018-02-12 07:47:11 | overcloud.AllNodesDeploySteps.ControllerDeployment_Step4.2:
2018-02-12 07:47:11 | resource_type: OS::Heat::StructuredDeployment
2018-02-12 07:47:11 | physical_resource_id: 16874306-3619-46ab-be0d-c0b91ee790fe
2018-02-12 07:47:11 | status: CREATE_FAILED
2018-02-12 07:47:11 | status_reason: |
2018-02-12 07:47:11 | Error: resources[2]: Deployment to server failed:
deploy_status_code : Deployment exited with non-zero status code: 2
2018-02-12 07:47:11 | deploy_stdout: |
2018-02-12 07:47:11 | ...
2018-02-12 07:47:11 | "a2b168fb9a3d: Pull complete",
2018-02-12 07:47:11 | "5e1193f920e0: Pull complete",
2018-02-12 07:47:11 | "Digest:
sha256:36ca9874c8359103a574ee7311688cdbefa61bec6a8fa5b728add93dc614d7c8"
2018-02-12 07:47:11 | ]
2018-02-12 07:47:11 | }
2018-02-12 07:47:11 | to retry, use: --limit
@/var/lib/heat-config/heat-config-ansible/7be0326a-ff52-4745-918c-f1279eb546f9_playbook.retry
2018-02-12 07:47:11 |
2018-02-12 07:47:11 | PLAY RECAP
*********************************************************************
2018-02-12 07:47:11 | localhost : ok=6 changed=2 unreachable=0
failed=1
2018-02-12 07:47:11 |
2018-02-12 07:47:11 | (truncated, view all with --long)
2018-02-12 07:47:11 | deploy_stderr: |
2018-02-12 07:47:11 |
2018-02-12 07:47:11 | + status_code=1
2018-02-12 07:47:11 | + openstack stack list
2018-02-12 07:47:11 | + grep -q overcloud
2018-02-12 07:47:21 | + grep -Eq '(CREATE|UPDATE)_COMPLETE'
2018-02-12 07:47:21 | + openstack stack list
2018-02-12 07:47:24 | + openstack stack failures list overcloud --long
2018-02-12 07:47:30 | ++ grep FAILED
2018-02-12 07:47:30 | ++ grep 'StructuredDeployment '
2018-02-12 07:47:30 | ++ cut -d '|' -f3
2018-02-12 07:47:30 | ++ openstack stack resource list --nested-depth 5 overcloud
2018-02-12 07:47:58 | + for failed in '$(openstack stack resource list
--nested-depth 5 overcloud | grep FAILED |
2018-02-12 07:47:58 | grep '\''StructuredDeployment '\'' |
cut -d '\''|'\'' -f3)'
2018-02-12 07:47:58 | + echo 'heat deployment-show output for deployment:
51c7e7c9-ae7b-41e8-ba62-6767c9730986'
2018-02-12 07:47:58 | + echo
'######################################################'
2018-02-12 07:47:58 | + heat deployment-show 51c7e7c9-ae7b-41e8-ba62-6767c9730986
2018-02-12 07:47:59 | WARNING (shell) "heat deployment-show" is deprecated,
please use "openstack software deployment show" instead
2018-02-12 07:48:00 | + echo
'######################################################'
2018-02-12 07:48:00 | + echo 'puppet standard error for deployment:
51c7e7c9-ae7b-41e8-ba62-6767c9730986'
2018-02-12 07:48:00 | + echo
'######################################################'
2018-02-12 07:48:00 | + heat deployment-show 51c7e7c9-ae7b-41e8-ba62-6767c9730986
2018-02-12 07:48:00 | + jq -r .output_values.deploy_stderr
2018-02-12 07:48:00 | + sed -r 's:\x1B\[[0-9;]*[mK]::g'
2018-02-12 07:48:01 | WARNING (shell) "heat deployment-show" is deprecated,
please use "openstack software deployment show" instead
2018-02-12 07:48:01 | + echo
'######################################################'
2018-02-12 07:48:01 | + for failed in '$(openstack stack resource list
--nested-depth 5 overcloud | grep FAILED |
2018-02-12 07:48:01 | grep '\''StructuredDeployment '\'' |
cut -d '\''|'\'' -f3)'
2018-02-12 07:48:01 | + echo 'heat deployment-show output for deployment:
16874306-3619-46ab-be0d-c0b91ee790fe'
2018-02-12 07:48:01 | + echo
'######################################################'
2018-02-12 07:48:01 | + heat deployment-show 16874306-3619-46ab-be0d-c0b91ee790fe
2018-02-12 07:48:02 | WARNING (shell) "heat deployment-show" is deprecated,
please use "openstack software deployment show" instead
2018-02-12 07:48:03 | + echo
'######################################################'
2018-02-12 07:48:03 | + echo 'puppet standard error for deployment:
16874306-3619-46ab-be0d-c0b91ee790fe'
2018-02-12 07:48:03 | + echo
'######################################################'
2018-02-12 07:48:03 | + jq -r .output_values.deploy_stderr
2018-02-12 07:48:03 | + sed -r 's:\x1B\[[0-9;]*[mK]::g'
2018-02-12 07:48:03 | + heat deployment-show 16874306-3619-46ab-be0d-c0b91ee790fe
2018-02-12 07:48:03 | WARNING (shell) "heat deployment-show" is deprecated,
please use "openstack software deployment show" instead
2018-02-12 07:48:04 | + echo
'######################################################'
2018-02-12 07:48:04 | + exit 1
________________________________
From: users <users-bounces(a)lists.rdoproject.org> on behalf of Boris
<dba477(a)list.ru>
Sent: Friday, February 9, 2018 9:03 PM
To: James LaBarre
Cc: users(a)lists.rdoproject.org
Subject: Re: [rdo-users] Quickstart and containers
Пятница, 9 февраля 2018, 19:17 +03:00 от James LaBarre <jlabarre(a)redhat.com>:
If I am trying to familiarize myself with the newer containerized
overcloud (Queens & upwards), is there a way to build a TripleO
Quickstart on a single node?
Due to I cannot any longer respond from hotmail.com ( backup address is used )
1. https://www.anstack.com/blog/2018/01/05/tripleo-quickstart-cheatsheet.html
2. this one was done via F27 connection to VIRTHOST
http://lxer.com/module/newswire/view/252059/index.html
I believe it might work in the way suggested by Carlos Camacho (1)
Boris.
As I say, it's mainly to get some
experience in the newer configurations, after which I could probably
grab some more baremetal and do proper setups.
I do find myself wondering, though, with the way Quickstart worked with
images, if it's been doing containers all along...
_______________________________________________
users mailing list
users@lists.rdoproject.org<mailto:users@lists.rdoproject.org>
http://lists.rdoproject.org/mailman/listinfo/users
To unsubscribe:
users-unsubscribe@lists.rdoproject.org<mailto:users-unsubscribe@lists.rdoproject.org>
С уважением,
Boris Derzhavets
dba477(a)list.ru
5:01 p.m.
(meant to send this to the general list)
On 02/09/2018 01:03 PM, Boris wrote:
Пятница, 9 февраля 2018, 19:17 +03:00 от James LaBarre
<jlabarre(a)redhat.com>:
If I am trying to familiarize myself with the newer containerized
overcloud (Queens & upwards), is there a way to build a TripleO
Quickstart on a single node?
Due to I cannot any longer respond from hotmail.com ( backup
address is used )
1.
https://www.anstack.com/blog/2018/01/05/tripleo-quickstart-cheatsheet.html
2. this one was done via F27 connection to VIRTHOST
http://lxer.com/module/newswire/view/252059/index.html
I believe it might work in the way suggested by Carlos Camacho (1)
Boris.
I tried the script as suggest by Carlos Camacho that you linked above.
As user "stack" it would fail when it was running the teardown process,
and would kill the user task before it finished.
So as a test I ran it as root, and it went through the
provision/teardown steps, but in the process wiped out everything in the
/home/stack directory (and I think removed the user since I can no
longer log into it.
Using a combination of both instructions, I cloned the git repos locally
for tripleo-quickstart tripleo-quickstart-extras tripleo-environments
tripleo-heat-templates tripleo-ui (needed the heat templates for Carlos'
directions). I'll have to poke through the templates & see if I can
figure out more.
Used the commands:
===================================
export CONFIG=/home/stack/deploy-configHA.yaml
export VIRTHOST=127.0.0.2
export OOOtimestamp=`date +%Y%m%d_%H%M`
bash ./tripleo-quickstart/quickstart.sh --clean --release queens
--teardown all --tags all -e @$CONFIG $VIRTHOST |& tee
~/OOOlog_${OOOtimestamp}.txt
4:54 p.m.
HI James,
Not for an "overcloud" I think. But you could check out some of the
work we did in Pike, Queens, and now Rocky for the new containerized
undercloud. See the upstream docs here:
https://docs.openstack.org/tripleo-docs/latest/install/containers_deploym...
There are two ways to approach it:
One is 'undercloud deploy' which is an experimental interface that
lets you interact with the heat templates to drive a single node
installation much like you would when creating an Overcloud.
The second approach which we are actively building is a command which
will consume the same undercloud.conf that instack-undercloud uses to
drive the same installer framework that 'undercloud deploy' provides.
Quickstart has some playbooks that can drive 'undercloud_deploy'...
but I think we are switching it over to 'undercloud install'
exclusively. So if either one suits your needs you might be able to
get quickstart to do this all for you on a single node as well.
We've got lots of ideas on how to improve the single node use case in
the future BTW. Hoping to prototype some things at the upcoming PTG in
Dublin.
Dan
On Fri, Feb 9, 2018 at 11:16 AM, James LaBarre <jlabarre(a)redhat.com> wrote:
If I am trying to familiarize myself with the newer containerized
overcloud (Queens & upwards), is there a way to build a TripleO
Quickstart on a single node? As I say, it's mainly to get some
experience in the newer configurations, after which I could probably
grab some more baremetal and do proper setups.
I do find myself wondering, though, with the way Quickstart worked with
images, if it's been doing containers all along...
_______________________________________________
users mailing list
users(a)lists.rdoproject.org
http://lists.rdoproject.org/mailman/listinfo/users
To unsubscribe: users-unsubscribe(a)lists.rdoproject.org
3:37 p.m.
On 02/09/2018 05:54 PM, Dan Prince wrote:
HI James,
Not for an "overcloud" I think. But you could check out some of the
work we did in Pike, Queens, and now Rocky for the new containerized
undercloud. See the upstream docs here:
https://docs.openstack.org/tripleo-docs/latest/install/containers_deploym...
There are two ways to approach it:
One is 'undercloud deploy' which is an experimental interface that
lets you interact with the heat templates to drive a single node
installation much like you would when creating an Overcloud.
The second approach which we are actively building is a command which
will consume the same undercloud.conf that instack-undercloud uses to
drive the same installer framework that 'undercloud deploy' provides.
Quickstart has some playbooks that can drive 'undercloud_deploy'...
but I think we are switching it over to 'undercloud install'
exclusively. So if either one suits your needs you might be able to
get quickstart to do this all for you on a single node as well.
We've got lots of ideas on how to improve the single node use case in
the future BTW. Hoping to prototype some things at the upcoming PTG in
Dublin.
I haven't been able to do any level of quickstart build now. I actually
went all the way back to the cheatsheet I had made for myself when I was
working with Pike on RHEL 7.4, a sheet which gave me the steps to
successfully build a quickstart, single-system install. Using those
exact steps, the setup kicks me off the system about a minute into the
setup. This is with RHEL 7.4, 7.5 against Pike as well as Queens.
I am at a loss now, seeing as I can't even duplicate a setup that worked
for me in December.
8:44 a.m.
On Tue, Feb 20, 2018 at 4:37 PM, James LaBarre <jlabarre(a)redhat.com> wrote:
On 02/09/2018 05:54 PM, Dan Prince wrote:
> HI James,
>
> Not for an "overcloud" I think. But you could check out some of the
> work we did in Pike, Queens, and now Rocky for the new containerized
> undercloud. See the upstream docs here:
>
> https://docs.openstack.org/tripleo-docs/latest/install/
containers_deployment/undercloud.html
>
> There are two ways to approach it:
>
> One is 'undercloud deploy' which is an experimental interface that
> lets you interact with the heat templates to drive a single node
> installation much like you would when creating an Overcloud.
>
> The second approach which we are actively building is a command which
> will consume the same undercloud.conf that instack-undercloud uses to
> drive the same installer framework that 'undercloud deploy' provides.
>
> Quickstart has some playbooks that can drive 'undercloud_deploy'...
> but I think we are switching it over to 'undercloud install'
> exclusively. So if either one suits your needs you might be able to
> get quickstart to do this all for you on a single node as well.
>
> We've got lots of ideas on how to improve the single node use case in
> the future BTW. Hoping to prototype some things at the upcoming PTG in
> Dublin.
I haven't been able to do any level of quickstart build now. I actually
went all the way back to the cheatsheet I had made for myself when I was
working with Pike on RHEL 7.4, a sheet which gave me the steps to
successfully build a quickstart, single-system install. Using those
exact steps, the setup kicks me off the system about a minute into the
setup. This is with RHEL 7.4, 7.5 against Pike as well as Queens.
I am at a loss now, seeing as I can't even duplicate a setup that worked
for me in December.
Please write a launchpad bug https://bugs.launchpad.net/tripleo
Include detailed errors, logs, the full command you are using to invoke
quickstart
We will address your issue from launchpad.
Thanks
_______________________________________________
users mailing list
users(a)lists.rdoproject.org
http://lists.rdoproject.org/mailman/listinfo/users
To unsubscribe: users-unsubscribe(a)lists.rdoproject.org
4:47 p.m.
On 02/22/2018 09:44 AM, Wesley Hayutin wrote:
Please write a launchpad bug https://bugs.launchpad.net/tripleo
Include detailed errors, logs, the full command you are using to
invoke quickstart
We will address your issue from launchpad.
Thanks
Entered at https://bugs.launchpad.net/tripleo/+bug/1751929
Let me know of any further information you need, or anything else you
want me to try out. Since the most basic run worked I'd like to see if
it does the same with a RHEL 7.5 system with Queens rather than Pike.
2263
days inactive
2280
days old
8 comments
5 participants
participants (5)
-
Boris -
Boris Derzhavets -
Dan Prince -
James LaBarre -
Wesley Hayutin