Fuel for OpenStack

  1. Bug #1536924
  2. Comment #21

Comment 21 for bug 1536924

Revision history for this message
Alexander Gordeev (a-gordeev) wrote : Re: Deploy cluster in HA mode with Neutron failed to execute hook 'clear_nodes_info'

Yup, indeed.

Mcollective module from cloud-init automatically stars the service, therefore there's a redundant mcollective' restart in runcmd which must be removed.

Jan 28 08:47:28 node-2 [CLOUDINIT] util.py[DEBUG]: Renaming /etc/mcollective/server.cfg to /etc/mcollective/server.cfg.old
Jan 28 08:47:28 node-2 [CLOUDINIT] util.py[DEBUG]: Writing to /etc/mcollective/server.cfg - wb: [420] 812 bytes
Jan 28 08:47:28 node-2 [CLOUDINIT] util.py[DEBUG]: Running command ['service', 'mcollective', 'start'] with allowed return codes [0] (shell=False, capture=False)
Jan 28 08:47:28 node-2 [CLOUDINIT] helpers.py[DEBUG]: Running config-disable-ec2-metadata using lock (<cloudinit.helpers.DummyLock object at 0x7fa0ffba86d0>)
Jan 28 08:47:28 node-2 [CLOUDINIT] util.py[DEBUG]: Running command ['route', 'add', '-host', '169.254.169.254', 'reject'] with allowed return codes [0] (shell=False, capture=False)
Jan 28 08:47:28 node-2 [CLOUDINIT] util.py[DEBUG]: Writing to /var/lib/cloud/instances/nocloud/sem/config_runcmd - wb: [420] 20 bytes
Jan 28 08:47:28 node-2 [CLOUDINIT] helpers.py[DEBUG]: Running config-runcmd using lock (<FileLock using file '/var/lib/cloud/instances/nocloud/sem/config_runcmd'>)
Jan 28 08:47:28 node-2 [CLOUDINIT] util.py[DEBUG]: Shellified 6 commands.

The issue is floating just because the amount of time between restarts is less than 1 sec. And it seems that the redundant restart often occurs without any side effects.