Bug 1257783

Summary: Connection reset by peer during Ironic provisioning
Product: Red Hat OpenStack Reporter: Gaëtan Trellu <gtrellu>
Component: rhosp-directorAssignee: Lucas Alvares Gomes <lmartins>
Status: CLOSED CANTFIX QA Contact: Shai Revivo <srevivo>
Severity: medium Docs Contact:
Priority: medium    
Version: 5.0 (RHEL 7)CC: chlong, gchenuet, gtrellu, lmartins, mburns, nauvray, racedoro, rhel-osp-director-maint, vmindru
Target Milestone: ---   
Target Release: 10.0 (Newton)   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2016-08-22 16:31:27 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
Connection reset by peer. none

Description Gaëtan Trellu 2015-08-28 03:26:50 UTC 
Description of problem:

During an overcloud deployment, sometime some Ironic nodes stay in "wait call-back" state.

After a quick jump on the IPMI console, we can see the "nc" process running but without any activity.

A little bit upper in the console, we can see a fail during the curl command:
  - Recv failure: Connection reset by peer

We just have to power cycle the node and the it pass in "Deploying -> Wait Call-back - > Deploying -> Active" state (because the curl has been performed).

Version-Release number of selected component (if applicable):

rhos-release-0.66-1.noarch
python-rdomanager-oscplugin-0.0.8-44.el7ost.noarch
puddle images 2015-08-07.1

How reproducible:

Try to deploy an overcloud Heat stack.

Actual results:

Ironic node stay in "wait call-back" state.

Expected results:

Ironic node in "active" state.

Additional info:

Have a look on the screenshot
Comment 3 Gaëtan Trellu 2015-08-28 03:29:05 UTC 
Created attachment 1067902 [details]
Connection reset by peer.
Comment 4 chris alfonso 2015-08-28 16:10:27 UTC 
Lucas, would you mind validating there isn't a config issue to look into on this?
Comment 7 Mike Burns 2016-04-07 20:47:27 UTC 
This bug did not make the OSP 8.0 release.  It is being deferred to OSP 10.
Comment 9 Lucas Alvares Gomes 2016-05-19 10:23:22 UTC 
Hi,

Just to confirm, was this bug found in osp 7 right? Cause using "nc" was an old behavior with the previous deploy ramdisk which has been replaced for osp8+.

Can you re validate it please?
Comment 10 Gaëtan Trellu 2016-06-15 15:03:03 UTC 
Hi,

No more issues in OSP-d 8.x

Gaetan
Comment 12 Lucas Alvares Gomes 2016-08-22 16:31:27 UTC 
(In reply to Gaëtan Trellu from comment #10)
> Hi,
> 
> No more issues in OSP-d 8.x
> 
> Gaetan

Thanks!

Closing this bug because we already got rid of the old bash ramdisk osp 8+