yahoo-eng-team team mailing list archive
-
yahoo-eng-team team
-
Mailing list archive
-
Message #05716
[Bug 1234479] Re: ipmi power probing is too fast
Reviewed: https://review.openstack.org/50935
Committed: http://github.com/openstack/nova/commit/c965c2b590a877994604d9163cd52446e0bd4c8a
Submitter: Jenkins
Branch: milestone-proposed
commit c965c2b590a877994604d9163cd52446e0bd4c8a
Author: Joe Gordon <joe.gordon0@xxxxxxxxx>
Date: Thu Oct 3 17:48:13 2013 -0700
Baremetal: Be more patient with IPMI and BMC
Before we called 'power status; power on' in a loop which made the
IPMI/BMCs not behave well. Also the total time we would wait (2.5
seconds wasn't always enough). So make sure power on/off is only called
once and wait up to 10 seconds for the power state change to go into
effect.
This patch has been tested on real baremetal using
https://wiki.openstack.org/wiki/TripleO/TripleOCloud
This bug is also linked to ironic so the change will be made there as
well.
Change-Id: I5a4d7c84ebdf9c1f7d8d0570dbc31764c31f1fc6
Closes-Bug: #1234479
(cherry picked from commit 6e30bbc126a1ad4eb87e735c812832f4fcc0054e)
** Changed in: nova
Status: Fix Committed => Fix Released
--
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1234479
Title:
ipmi power probing is too fast
Status in Ironic (Bare Metal Provisioning):
New
Status in OpenStack Compute (Nova):
Fix Released
Bug description:
baremetal nodes are failing to boot for me:
2013-10-03 01:55:15,392.392 7354 DEBUG nova.openstack.common.processutils [-] Running cmd (subprocess): ipmitool -I lanplus -H 10.10.16.27 -U Administrator -f /tmp/tmpMxP1Xr power on execute /opt/stack/venvs/nova/local/lib/pyt
hon2.7/site-packages/nova/openstack/common/processutils.py:147
2013-10-03 01:55:15,573.573 7354 DEBUG nova.virt.baremetal.ipmi [-] ipmitool stdout: 'Chassis Power Control: Up/On
', stderr: '' _exec_ipmitool /opt/stack/venvs/nova/local/lib/python2.7/site-packages/nova/virt/baremetal/ipmi.py:136
2013-10-03 01:55:15,707.707 7354 DEBUG nova.openstack.common.processutils [-] Running cmd (subprocess): ipmitool -I lanplus -H 10.10.16.27 -U Administrator -f /tmp/tmpRYI8dn power status execute /opt/stack/venvs/nova/local/lib
/python2.7/site-packages/nova/openstack/common/processutils.py:147
2013-10-03 01:55:15,890.890 7354 DEBUG nova.virt.baremetal.ipmi [-] ipmitool stdout: 'Chassis Power is off
', stderr: '' _exec_ipmitool /opt/stack/venvs/nova/local/lib/python2.7/site-packages/nova/virt/baremetal/ipmi.py:136
Note the gap between up and status checking is only 200ms (from IPMI return to status check) but in that interval this hardware doesn't toggle to on.
After the deploy fails, a manual check with
ipmitool -I lanplus -H 10.10.16.40 -U Administrator -P <password> chassis power status
returns
Chassis Power is on
To manage notifications about this bug go to:
https://bugs.launchpad.net/ironic/+bug/1234479/+subscriptions