`syslog_aggregator/0' is not running after update

[Upendra Sharma]

Hi,
I am trying to deploy CF on OpenStack (I have Grizzly version deployed over here)
I have closely followed these instructions (http://docs.cloudfoundry.com/docs/running/deploying-cf/openstack/)
In the final step (Step 7: Install Cloud Foundry using BOSH ) I downloaded the cf/138 release following these instructions ( http://docs.cloudfoundry.com/docs/running/deploying-cf/common/cf-release.html) -- except that I ran ./update after doing

git clone -b release-candidate git://github.com/cloudfoundry/cf-release.git.

I uploaded the release and after create a proper manifest files (edited version of the demo.yml). I was not sure about the domain name so I left it as it is. The top of my demo.yml looks like this
<%
director_uuid = "7867a0f8-d48d-4037-9378-6fcb52eb396e"
protocol = "http"
ip_address = "9.47.173.169"
common_password = "c1oudc0wc1oudc0w"
root_domain = "#{ip_address}.xip.io"
deployment_name = "cf-demo"
%>

Then I ran bosh deploy. The command ran for a while and eventually gave task error;
bosh tasks recent
| 12 | error | 2013-09-08 18:48:37 UTC | admin | create deployment         | `syslog_aggregator/0' is not running after update |

The debug log of create-deployment task is long but the place where it shows the error are shown below:

bosh task 12 --debug

D, [2013-09-08T18:48:37.160169 #19619] [task:12] DEBUG -- : Deleted lock: lock:deployment:cf-demo
E, [2013-09-08T18:48:37.173615 #19619] [task:12] ERROR -- : `syslog_aggregator/0' is not running after update
/var/vcap/packages/director/gem_home/gems/director-1.5.0.pre.975/lib/director/instance_updater.rb:86:in `update'
/var/vcap/packages/director/gem_home/gems/director-1.5.0.pre.975/lib/director/job_updater.rb:63:in `block (5 levels) in update'
/var/vcap/packages/director/gem_home/gems/bosh_common-1.5.0.pre.975/lib/common/thread_formatter.rb:46:in `with_thread_name'
/var/vcap/packages/director/gem_home/gems/director-1.5.0.pre.975/lib/director/job_updater.rb:60:in `block (4 levels) in update'
/var/vcap/packages/director/gem_home/gems/director-1.5.0.pre.975/lib/director/event_log.rb:58:in `track'
/var/vcap/packages/director/gem_home/gems/director-1.5.0.pre.975/lib/director/job_updater.rb:59:in `block (3 levels) in update'
/var/vcap/packages/director/gem_home/gems/bosh_common-1.5.0.pre.975/lib/common/thread_pool.rb:83:in `call'
/var/vcap/packages/director/gem_home/gems/bosh_common-1.5.0.pre.975/lib/common/thread_pool.rb:83:in `block (2 levels) in create_thread'
/var/vcap/packages/director/gem_home/gems/bosh_common-1.5.0.pre.975/lib/common/thread_pool.rb:67:in `loop'
/var/vcap/packages/director/gem_home/gems/bosh_common-1.5.0.pre.975/lib/common/thread_pool.rb:67:in `block in create_thread'


Please let me know what I am doing wrong over here.

Thanks in advance,
-upendra

[Ferran Rodenas]

Hi Upendra,

I ran a similar problem today using cf-140. Try changing the job order, so nats is deployed before syslog_aggregator:

 ...

jobs:

  - name: nats

    template:

      - nats

    instances: 1

    resource_pool: common

    networks:

      - name: default

        default: [dns, gateway]

  - name: syslog_aggregator

    template:

      - syslog_aggregator

    instances: 1

    resource_pool: common

    persistent_disk: 65536

    networks:

      - name: default

        default: [dns, gateway]

  ...

- Ferdy

2013/9/9 Upendra Sharma <upendra...@gmail.com>

To unsubscribe from this group and stop receiving emails from it, send an email to bosh-users+...@cloudfoundry.org.

[Upendra Sharma]

Some more additional info
- bosh gem versions
   gem list bosh

*** LOCAL GEMS ***

bosh_aws_cpi (1.5.0.pre.975)
bosh_cli (1.5.0.pre.975)
bosh_cli_plugin_micro (1.5.0.pre.975)
bosh_common (1.5.0.pre.975)
bosh_cpi (1.5.0.pre.975)
bosh_openstack_cpi (1.5.0.pre.975)
bosh_registry (1.5.0.pre.975)
bosh_vcloud_cpi (0.4.9)
bosh_vsphere_cpi (1.5.0.pre.975)

- bosh director info
bosh status
Config
             /root/.bosh_config

Director
  Name       microbosh-openstack
  URL        https://9.47.173.162:25555
  Version    1.5.0.pre.975 (release:fbb73b1d bosh:fbb73b1d)
  User       admin
  UUID       7867a0f8-d48d-4037-9378-6fcb52eb396e
  CPI        openstack
  dns        enabled (domain_name: microbosh)
  compiled_package_cache disabled
  snapshots  disabled

Deployment
  Manifest   /root/bosh-workspace/deployments/cf/demo.yml
 
- stemcell version(s) you are using
bosh stemcells

+---------------+---------+--------------------------------------+
| Name          | Version | CID                                  |
+---------------+---------+--------------------------------------+
| bosh-stemcell | 975     | 087ad071-8471-451a-8e4d-d418c5ae02ee |
+---------------+---------+--------------------------------------+

- the release you are trying to deploy
 bosh releases

+------+----------+-------------+
| Name | Versions | Commit Hash |
+------+----------+-------------+
| cf   | 138*     | adca9c45+   |
+------+----------+-------------+

-- deployment manifest is demo.yml (http://docs.cloudfoundry.com/docs/running/deploying-cf/openstack/install_cf_openstack.html)

[Upendra Sharma]

Thanks Fredy, reordering nats and syslog_aggregator did correct the problem.

--
-u

[Ferran Rodenas]

Cool, thanks. I'll modify the instructions with the updated deployment manifest.

- Ferdy

[Dr Nic Williams]

Completely anally retentively: the docs do state they are for v138; and that's for exactly this reason. Its hard to write documentation that is future proof :)

thx @ferdy for fixing that already in cf-docs xoxo https://github.com/cloudfoundry/cf-docs/commit/b16301ec86385dce1b15aeababb2784a3097e8b3

Nic

Dr Nic Williams

Stark & Wayne LLC - consultancy for Cloud Foundry users

http://drnicwilliams.com

http://starkandwayne.com

cell +1 (415) 860-2185

twitter @drnic