Regarding Block Live Migration on DevStack-Nova Error [closed]

asked 2016-11-15 13:46:08 -0500

ajayraop gravatar image

updated 2016-11-30 08:50:39 -0500

rbowen gravatar image

I'm using Devstack-mitaka branch to set up a in-house multi-node cloud with 1 controller node and 5 compute node. The set-up process was quite easy and controller node can launch vm's across compute nodes and these can be accessed across the subnet. As I'm experimenting on these systems, I haven't set up shared storage server. I'm now trying to achieve Block Live migration across these nodes but the active log states on dashboard shows error for this particular operation. Going through n-cpu.log log (nova compute log), I find this particular exception related to oslo_messaging rpc -

2016-11-15 13:53:00.810 [01;31mERROR oslo_messaging.rpc.dispatcher [ [01;36mreq-e3a2f83c-fbc1-4677-86ea-f9f07416e33a [00;36madmin admin [01;31m] [01;35m [01;31mException during message handling: Remote error: ProcessExecutionError Unexpected error while running command.

with few DEBUG comments. I searched online for solution, but couldn't find any satisfying solution. Is there place I can look for help i.e website or people I can contact regarding this matter.

The system configuration on each system is, i7 processor , 8GB RAM, 1TB harddisk, Ubuntu 14.04 Desktop connected via a small Dlink 100 Mbps router.

n-cpu.log with only log related to block live migration request-Id "req-e3a2f83c-fbc1-4677-86ea-f9f07416e33a" below

2016-11-15 13:51:42.335 DEBUG nova.objects.instance [req-e3a2f83c-fbc1-4677-86ea-f9f07416e33a admin admin] Lazy-loading 'vcpu_model' on Instance uuid 1b79133c-5933-4997-8db4-5aa485336c34 from (pid=21881) obj_load_attr /opt/stack/nova/nova/objects/instance.py:895
2016-11-15 13:51:42.341 INFO nova.virt.libvirt.driver [req-e3a2f83c-fbc1-4677-86ea-f9f07416e33a admin admin] Instance launched has CPU info: {"vendor": "Intel", "model": "Haswell", "arch": "x86_64", "features": ["pge", "avx", "clflush", "sep", "rtm", "vme", "dtes64", "invpcid", "tsc", "fsgsbase", "xsave", "vmx", "erms", "xtpr", "cmov", "hle", "smep", "ssse3", "est", "pat", "monitor", "smx", "pbe", "lm", "msr", "nx", "fxsr", "syscall", "tm", "sse4.1", "pae", "sse4.2", "pclmuldq", "acpi", "fma", "tsc-deadline", "mmx", "osxsave", "cx8", "mce", "de", "tm2", "ht", "pse", "lahf_lm", "abm", "popcnt", "mca", "pdpe1gb", "apic", "sse", "f16c", "ds", "pni", "rdtscp", "avx2", "aes", "sse2", "ss", "ds_cpl", "bmi1", "bmi2", "pcid", "fpu", "cx16", "pse36", "mtrr", "movbe", "pdcm", "rdrand", "x2apic"], "topology": {"cores": 4, "cells": 1, "threads": 2, "sockets": 1}}
2016-11-15 13:51:42.342 DEBUG nova.virt.libvirt.config [req-e3a2f83c-fbc1-4677-86ea-f9f07416e33a admin admin] Generated XML ('<cpu>\n  <arch>x86_64</arch>\n  <model>Haswell</model>\n  <vendor>Intel</vendor>\n  <topology sockets="1" cores="4" threads="2"/>\n  <feature name="abm"/>\n  <feature name="acpi"/>\n  <feature name="aes"/>\n  <feature name="apic"/>\n  <feature name="avx"/>\n  <feature name="avx2"/>\n  <feature name="bmi1"/>\n  <feature name="bmi2"/>\n  <feature name="clflush"/>\n  <feature name="cmov"/>\n  <feature name="cx16"/>\n  <feature name="cx8"/>\n  <feature name="de"/>\n  <feature name="ds"/>\n  <feature name="ds_cpl"/>\n  <feature name="dtes64"/>\n  <feature name="erms"/>\n  <feature name="est"/>\n  <feature name="f16c"/>\n  <feature name="fma"/>\n  <feature name="fpu"/>\n  <feature name="fsgsbase"/>\n  <feature name="fxsr"/>\n ...
(more)
edit retag flag offensive reopen merge delete

Closed for the following reason question is off-topic or not relevant by ajayraop
close date 2016-12-07 12:00:36.966749

Comments

I see "An error occurred while deallocating network". Something wants this migration to roll back. Perhaps there is a clue in n-cpu.log on the source compute node, and also the nova-cond.log from the controller?

Bernd Bausch gravatar imageBernd Bausch ( 2016-11-15 17:42:51 -0500 )edit

By the way, I hope you are not using devstack for anything else but testing, learning and proof of concept.

Bernd Bausch gravatar imageBernd Bausch ( 2016-11-15 17:43:41 -0500 )edit

At the source node --> " 2016-11-15 13:52:43.035 [01;31mERROR nova.compute.manager [ [01;36mreq-e3a2f83c-fbc1-4677-86ea-f9f07416e33a [00;36madmin admin [01;31m] [01;35m[instance: 1b79133c-5933-4997-8db4-5aa485336c34] [01;31mPre live migration failed at pgcne-16 [00m ".

ajayraop gravatar imageajayraop ( 2016-11-17 22:33:42 -0500 )edit

pgcne-16 is the destination node.

ajayraop gravatar imageajayraop ( 2016-11-17 22:33:55 -0500 )edit

No entry in n-cond.log file for this request Id

ajayraop gravatar imageajayraop ( 2016-11-17 22:35:01 -0500 )edit