[NETVIRT-801] An instance can't get an IP when Restarting the compute node Created: 23/Jul/17  Updated: 03/May/18  Resolved: 06/Apr/18

Status: Resolved
Project: netvirt
Component/s: General
Affects Version/s: Carbon
Fix Version/s: None

Type: Bug Priority: Medium
Reporter: Itzik Brown Assignee: Unassigned
Resolution: Cannot Reproduce Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified
Environment:

Operating System: Linux
Platform: All

Attachments: Text File karaf.log     File logs.tar.gz    
External issue ID: 8877

 Description   

Deploying OpenStack Pike with Carbon SR1.
After restarting a compute node and launching an instance the Instance doesn't get an IP from DHCP.
After restarting Open vSwitch and running the DHCP client again the Instance gets an IP.

It seems that the following flow is missing
table=0, n_packets=0, n_bytes=0, priority=4,in_port=2,vlan_tci=0x0000/0x1fff actions=write_metadata:0x180000000000/0xffffff0000000001,goto_table:17

Version
=======
OpenStack Pike
OpenDaylight Carbon SR-1
Open vSwitch 2.7

Steps to reproduce
==================
1. Restart the compute node
2. Launch an instance on the above compute node
3. Verify the instance is not getting an IP
4. Restart the openvswitch on the compute node
5. Restart the DHCP client and verify the instance gets an IP.

 Comments   
Comment by Itzik Brown [ 23/Jul/17 ]

Attachment logs.tar.gz has been added with description: Out out logs,flows etc.

Comment by Itzik Brown [ 23/Jul/17 ]

Attachment karaf.log has been added with description: Karaf log after compute restart

Comment by Jamo Luhrsen [ 24/Jul/17 ]

unsure if it's related, but I don't recall seeing this ERROR before:

2017-07-23 04:36:18,830 | ERROR | assiveConnServ-7 | StalePassiveConnectionService | 285 - org.opendaylight.ovsdb.library - 1.4.1.Carbon-redhat-1 | Error in checking stale connections)
java.util.concurrent.CancellationException: Task was cancelled.

It's coming a few times in the karaf.log

Comment by Hari Prasidh [ 20/Sep/17 ]

I have tested the same scenario with ocata setup and observed instance getting IP even after restart the compute node.

Generated at Wed Feb 07 20:22:30 UTC 2024 using Jira 8.20.10#820010-sha1:ace47f9899e9ee25d7157d59aa17ab06aee30d3d.