Sporadic external connectivity failures in staging jobs

Bug #1470545 reported by Roman Podoliaka
12
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Fuel for OpenStack
Invalid
High
Fuel DevOps

Bug Description

http://jenkins-product.srt.mirantis.net:8080/view/All/job/5.1.2.test_staging_mirror/202 failed with

AssertionError: Failed tests, fails: 1 should fail: 0 failed tests name: [{u'Check network connectivity from instance via floating IP (failure)': u'Time limit exceeded while waiting for public connectivity checking from VM to finish. Please refer to OpenStack logs for more details.'}]

Looks like we hit a temporarily connectivity issue:

http://paste.openstack.org/show/332206/

New instances work just fine.

Changed in fuel:
assignee: nobody → Fuel Python Team (fuel-python)
milestone: none → 7.0
Changed in fuel:
status: New → Confirmed
Revision history for this message
Vladimir Sharshov (vsharshov) wrote :

Looks like have same nature as https://bugs.launchpad.net/fuel/+bug/1471812

Changed in fuel:
assignee: Fuel Python Team (fuel-python) → Fuel DevOps (fuel-devops)
importance: Undecided → High
Revision history for this message
Vladimir Sharshov (vsharshov) wrote :

Guys, i mark it as High because false alarm for CI is distracting. Feel free to set another importance priority.

Revision history for this message
Igor Shishkin (teran) wrote :

Roman, please ensure you have proper network settings on Fuel Master node and your Controllers compute.
For that purpose please attach output of the following command:

On Fuel Master node:
ip a
ping 8.8.8.8
arp -an

On other nodes:
ip a
ping $MASTERNODE_IP
arp -an

Marking as incomplete and assigning to author since there is no enough details.

Changed in fuel:
assignee: Fuel DevOps (fuel-devops) → Roman Podoliaka (rpodolyaka)
status: Confirmed → Incomplete
tags: added: devops
Revision history for this message
Roman Podoliaka (rpodolyaka) wrote :

Igor, I reported this bug when our team was on the staging mirrors duty, so I don't have any environments for you.

My point here is that it's not a problem with settings on the master node, but rather *irregular* issues with external connectivity, which happen time to time on our Jenkins slaves.

The only reason I filed this bug is that we have this policy about attaching a bug entry to each Jenkins job failure. But as we see, there is interest for this as these connectivity issues continue to happen.

Revision history for this message
Roman Podoliaka (rpodolyaka) wrote :

I suggest we just mark the failed jobs in a special way, rather than continue to file new and new LP issues for the very same problem with external connectivity.

And on fuel-devops team side it's probably worth investigation why those connectivity issues happen rather often on the Jenkins slaves.

Changed in fuel:
status: Incomplete → Confirmed
assignee: Roman Podoliaka (rpodolyaka) → Fuel DevOps (fuel-devops)
Revision history for this message
Igor Shishkin (teran) wrote :

moving to 8.0 since there is no real progress started and it doesn't affect 7.0 release

Changed in fuel:
milestone: 7.0 → 8.0
Dmitry Pyzhov (dpyzhov)
tags: added: area-devops
Revision history for this message
Igor Shishkin (teran) wrote :

Do we met this issue last time?

Igor Shishkin (teran)
Changed in fuel:
status: Confirmed → Incomplete
Revision history for this message
Igor Shishkin (teran) wrote :

Marking invalid according to policy.

Changed in fuel:
status: Incomplete → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.