Tag Archives: disk

Thanks for the support in 2016

 

Most people will say that 2016 was a terrible year and can’t wait for 2017. I agree that 2016 was not perfect for many people but it was a great year for linux-toys. I made this blog with the goal to influence people and drive that creative spark that we all have inside. In this blog post I will go over the website statistics and discuss a few of the blog entries that I thought were most influential for the year.

Continue reading Thanks for the support in 2016

My Journey to Improve Disk Performance on the Raspberry Pi

 

I switched to Gluster FS a while ago to provide easier container mobility across my Raspberry Pi Docker Cluster. Gluster worked great and was easy to get up and running but I had very poor performance. The average write speed was about 1 MB/s which is unacceptable for a filesystem that will undergo a lot of writes. I decided that it was time to take action and started looking at kernel parameters that could be changed.

From researching performance over the years, I often came across the following settings and thought that this is the best place to start. Since the Raspberry Pi is not really made for performance, I did not think it was worth the time investigating additional settings.

Continue reading My Journey to Improve Disk Performance on the Raspberry Pi

Running a Gluster Cluster on the Raspberry Pi with Docker

I was always fascinated with distributed filesystems and wanted to learn more about Gluster since it is becoming more popular in larger open-source projects. Since I have a few Raspberry Pi’s, I thought that now is the best time to learn. This blog post will explain how to run Gluster on a two-node Raspberry Pi cluster from a Docker container.

Architecture

  1. Two Raspberry Pi’s (rpi-1 and rpi-2)
  2. Running a Gluster image from a local Docker registry
  3. Hostnames are resolvable in /etc/hosts on both Pi’s
  4. Docker 1.12.x installed

Continue reading Running a Gluster Cluster on the Raspberry Pi with Docker

Using LVM cache on Linux

The Challenge

My home server uses a RAID 1 configuration. I was very disappointed in the performance and wanted to find a way to make it faster. After browsing the Internet one day, I came across news headlines that said CentOS 7  supports LVM cache. I found an old USB thumb drive and decided to take the cache challenge and see how it performs.

The Journey

Here is a simple DD test prior to enabling cache:

dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 6.27698 s, 167 MB/s

dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 5.04032 s, 208 MB/s

dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 3.41007 s, 307 MB/s

dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 2.94413 s, 356 MB/s

Average write speed: 256.5 MB/s

Time to enable caching and try to make the system perform better:

vgextend vg /dev/sdc

lvcreate -L 1G -n cache_metadata /dev/sdc

lvcreate -L 8G -n cache_vol /dev/sdc

lvconvert –type cache-pool –poolmetadata vg/cache_metadata vg/cache_vol

lvconvert –type cache –cachepool vg/cache_vol vg/original_volume_name

 

The write results with caching enabled:

# dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 3.73197 s, 281 MB/s

# dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 1.70449 s, 615 MB/s

# dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 3.91247 s, 268 MB/s

]# dd if=/dev/zero of=/tmp/1G bs=1M count=1000

1048576000 bytes (1.0 GB) copied, 2.18025 s, 481 MB/s

Average write speed: 411.25 MB/s

Conclusion:

When I originally built this machine from used parts on Amazon, I decided to reuse two old Western Digital Green drives which offer low performance and power usage.  I had no idea that they would perform poorly in RAID 1.  I was surprised and glad that a cheap USB flash drive helped me get a significant increase in write performance by an average of 155 MB/s. I find it fascinating how the Linux ecosystem helps people recycle old junk and put it to good use. Hooray.

 

Improving IO performance on Linux

I purchased a new server and have been struggling with IO performance on a RAID 1 setup. I first tried RAID 5 but it was horrible. Through all my struggles, I found a few kernel and filesystem tunables that helped me out.

The virtual machines improved also. They were barely usable.

Filesystem Tweaks: /etc/fstab

Change your mount options to:

defaults,noatime,data=writeback,barrier=0,nobh
Linux Kernel tunables:
sysctl -w vm.swappiness=0
sysctl -w vm.dirty_background_ratio=5
sysctl -w vm.dirty_ratio=5
sysctl -w vm.vfs_cache_pressure=200
The dirty tunables controls the portion of memory that the kernel will store pages in. I decreased the value to 5 so that I will have more memory free.  The cache_pressure tunable tells the kernel how to quickly free up the cache.

Convert root partition to LVM and mirror the root disk

1. Create a tar backup of your filesystem.

# tar czpf /root/redhat.tar –exclude=/var/tmp/portage/* –exclude=/root/* –exclude=/usr/portage/* –exclude=*.deb –exclude=/tmp/* –exclude=*.rpm –exclude=/sys/* –exclude=/proc/* –exclude=/dev/* –exclude=/mnt/* –exclude=/media/* –exclude=/home/* –exclude=/var/lib/libvirt/images/* –exclude=/oracle/* –exclude=redhat.tar

2. Use fdisk to create /boot and 1 LVM partition on the new disk.

/dev/sda1 * 1 100 803218+ 83 Linux
/dev/sda2 101121601 975956782+ 8e Linux LVM

3. Set /dev/sda1 to be bootable.

# parted /dev/sda set 1 boot on

4. Create the new LVM partition.

# pvcreate /dev/sda2
# vgcreate vg /dev/sda2
# lvcreate -L 200G /dev/vg -n root
# mkfs /dev/vg/root
# mkfs /dev/sda1
# mount /dev/vg/root /mnt
# mount /dev/sda1 /mnt/boot

5. Extract the tar file to /mnt

# tar xpf /root/redhat.tar -C /mnt/

6. Modify the following files:

/mnt/boot/grub/menu.list

Modify the kernel line to support LVM by adding the following LVM details:

rd_LVM_VG=vg rd_LVM_LV=root

Also ensure that initrd and kernel does not have /boot/ in the location.

Example:

kernel /vmlinuz-2.6.32-279.2.1.el6.x86_64 ro root=/dev/mapper/vg-root LANG=en_US.UTF-8 SYSFONT=latarcyrheb-sun16 KEYBOARDTYPE=pc KEYTABLE=us crashkernel=128M rd_LVM_VG=vg rd_LVM_LV=root rhgb quiet

initrd /initramfs-2.6.32-279.2.1.el6.x86_64.img

/mnt/etc/fstab:

Change the /boot and / entries to LVM:

/dev/sda1 /boot ext4 defaults 0 0
/dev/mapper/vg-root / ext4 defaults 1 1

7. Mount and configure the new environment:

# mount /dev/vg/root /mnt
# mount /dev/sda1 /mnt/boot
# mount -o bind /sys /mnt/sys
# mount -o bind /dev /mnt/dev
# mount -o bind /proc /mnt/proc
# grep -v rootfs /proc/mounts > /mnt/etc/mtab

Modify /mnt/etc/mtab and add:

/dev/sda1 /boot ext4 rw 0 0

# chroot /mnt

8. Install GRUB and reconfigure the ram disk image:

# grub-install –recheck /dev/sda
# dracut -force

9. Unmount and reboot:

Type exit to exit the chroot environment
# cd /
# umount /mnt/*
# umount /mnt
# reboot

Set your system to boot from the disk known as /dev/sda

10. Initialize and format your original boot disk.

Just like we did for /dev/sda. (1 bootable partition for /boot and 1 Linux LVM partition.

Device Boot Start End Blocks Id System
/dev/sdc1 * 1 100 803218+ 83 Linux
/dev/sdc2 101 121601 975956782+ 8e Linux LVM

11. Add /dev/sdc to the volume group.
# vgextend /dev/vg /dev/sdc2

12. Format the boot partition on the drive and set it bootable:
# mkfs /dev/sdc1
# parted /dev/sdc set 1 boot on

13. Mirror the boot disk:

# lvconvert -m1 /dev/vg/root