Why are system boots so slow with large non-thin LVM snapshots?
Issue
- A simple cold boot of the two hypervisors we're using (production and test) takes up to 50 minutes. The system spends almost all of that time 'hung' at the 'Setting up Logical Volume Management' stage of the boot.
- This system really did take 47 minutes to boot in total, 'hanging' for most of that time in the LVM setup stage. I'm curious as to why it takes so long. Over the past couple of years I've noticed that boots take a bit longer when non-thin LVM snapshots exist and have significant allocated storage ... but nothing as bad as waiting 47 minutes for a simple cold boot!
- Can you please advise on whether such extreme delays are typical for a system using LVM like this? And why that is the case? And whether there is anything we can do to speed up a boot?
Environment
- RHEL6.5GA kernel version 2.6.32-431.el6
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
