Sun Microsystems, Inc.  Oracle System Handbook - ISO 7.0 May 2018 Internal/Partner Edition
   Home | Current Systems | Former STK Products | EOL Systems | Components | General Info | Search | Feedback

Asset ID: 1-72-2374670.1
Update Date:2018-03-21
Keywords:

Solution Type  Problem Resolution Sure

Solution  2374670.1 :   VM Takes Long Time to Boot And Grub Menu Delayed  


Related Items
  • Oracle VM
  •  
  • Oracle Database Appliance
  •  
Related Categories
  • PLA-Support>Infrastructure>Operating Systems and Virtualization>Virtualization>Oracle VM
  •  


This document provides the solution to abnormal boot time required by VMs running on ODA and OVS 3.4.x

In this Document
Symptoms
Changes
Cause
Solution
References


Created from <SR 3-16826533061>

Applies to:

Oracle VM - Version 3.4.1 to 3.4.4 [Release OVM34]
Oracle Database Appliance - Version All Versions to All Versions [Release All Releases]
Linux x86-64

Symptoms

VM takes long time to boot. 
Around 30 mins are taken to load the grub menu and once it is loaded, the VM finishes booting in 2 mins.

The qemu-dm log is started once the VM is started and logs several lines in the beginning, and then stops logging for around 30 mins, and then resumes logging.

...........
...........
...........
initialise() failed
cirrus vga map change while on lfb mode
mapping vram to f0000000 - f0400000
platform_fixed_ioport: changed ro/rw state of ROM memory area. now is rw state.
platform_fixed_ioport: changed ro/rw state of ROM memory area. now is ro state.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> the logs are stopped
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> the logs continue after 30 min 
Unknown PV product 3 loaded in guest
PV driver build 1
region type 1 at [c100,c200).
region type 0 at [f3001000,f3001100).
squash iomem [f3001000, f3001100).
region type 1 at [c200,c300).
region type 0 at [f3001100,f3001200).
squash iomem [f3001100, f3001200).
ACPI:debug: write addr=0xb044, val=0x0.
ACPI:debug: write addr=0xb045, val=0x89.
...........
...........
...........

 

Also, only one VCPU is used by the VM for about 30 min after the VM is started, while four VCPU's are assigned to VM. After 30 mins the VM gets 4 VCPU's.

During this time the VM console is not accessible either via "oakcli console" or "xm console" or "xm create -c"

Changes

 No changes

Cause

The problem was that grub2 was not caching the contents of the extent block:
for every block of a file it would read the extent map and then read the next block in the file.

In a typical initramfs that amounts to about 30,000 reads of the same block.

BUG 18812855 - Slow grub read of initramfs image; bare-metal and pvhvm boot time much slower

Solution

Upgrade the grub package to version grub-0.97-94.0.5.el6_7.1.x86_64 or higher, preferably latest and then reinstall grub on the /dev/xxxx (the boot disk)

1.Upgrade existing package using below command   

 

# rpm -Uvh grub-xxxxxxxxx.x86_64.rpm

2.And then reinstall grub on boot disk

 

# grub-install /dev/xxxx

3.Now reboot the VM.

 

References

<BUG:18812855> - SLOW GRUB READ OF INITRAMFS IMAGE; BARE-METAL AND PVHVM BOOT TIME MUCH SLOWER
<NOTE:1639348.1> - ODAVP: How to boot Guest DomU's into the Rescue System for disaster recovery
<NOTE:1581363.1> - "grub-install" Command Fails with Error "/deb/xvda does not have any corresponding BIOS driver."

Attachments
This solution has no attachment
  Copyright © 2018 Oracle, Inc.  All rights reserved.
 Feedback