CentOS LVM Slow Imaging
Using 500gb hard disk
CentOS is LVM partition
I facing slowness when imaging CentOS
is there any reason behind causing the slowness?
what are the configuration/setting i can made to make it faster?
I do tested at the same machine with different OS
Image at Server
CentOS = 11GB
Debian = 7.4 GB
Image Deployment Duration
CentOS = 50min
Debian = 11min
Please do let me know if you need any other info
what do u mean by RAW image type?
This is a mode for partclone when it can’t determine the format of the disk, this is a default mode when no other disk format filters exist, such as with LVM. In linux terms partclone uses dd (disk dump) format to image the disk. This is the slowest and largest format since it copies all disk blocks. Partclone might change to this format if it detects and encrypted drive where it can’t read the partition information because its encrypted.
Hi @wayne-workman can you share with me on what do u mean by RAW image type?
I manage to capture the centOS with LVM just the problem is very slow.
Look carefully at the blue partclone screen. See if it says RAW when capturing the LVM partition. This is why its slow its using block level cloning instead of compressed data cloning. As Wayne said FOS Linux doesn’t support lvm volumes. If you recreate your centos image using standard partitions cloning it should be at comparable speeds to the debian install which uses standard partitions by default.
To clarify, FOG server is using ubuntu. Im trying to capture & deploy different OS.
The experiment carry out with the exact same machine (same cup/disk/ram/network). Image definition is the same, compression ratio, image type etc. FOG server locate at same place.
Things different only the OS type. For CentOS after capture the image on client showing 4xx GB while debian are showing 2X GB. I check the partition type with df command things that are different is CentOS using LVM.
I do try to change to SSD it become faster for centOS but at the same time debian also getting faster.
Hi @Wayne-Workman I manage to capture the centOS with LVM just the problem is very slow.
If were to made some calculation based on the size at server by doubling the debian image size 14.8gb suppose it should take 22min however centOS with just 11GB it takes 50min about 4x difference with debian.
Wayne Workman last edited by
@wt_101 Fog doesn’t support LVM, keep that in mind. If you have more complicated of a setup than a single disk - single partition - single LVM volume, this probably won’t work unless you use RAW image type.
@wt_101 What are your varaiables here. Is the debian and centos installed on the same exact server just different operating systems?
Is the image definition the same (same compression settings)?
Is the source computer the same exact computer each time?
Is the source computer and FOG server plugged into same network jack each time?
During FOG Imaging the source/target computer does all of the heavy lifting of compressing and decompressing the image, the fog server really just moves the data from the network adapter to the local storage drive and manages the overall imaging process. Nothing on the fog server is real intensive CPU usage. So just off the top I would say something has changed on the source computer or your disk subsystem has an issue on the FOG server.
I wrote this post many years ago now: https://forums.fogproject.org/topic/10459/can-you-make-fog-imaging-go-fast Why I bring it up it mentions some performance testing commands. It would be interesting to see if the dd command for created local files on the FOG server have a change between centos and debian. There is also iperf commands to test the network throughput too.