“Byobu is a Japanese term for decorative, multi-panel screens that serve as folding room dividers. As an open source project, Byobu is an elegant enhancement of the otherwise functional, plain, practical GNU Screen.”

https://wiki.ubuntu.com/MeetingLogs/openweekKarmic/Byobu

SSH into your server and install the two packages:

sudo apt-get install byobu landscape-common update-motd

Next, run byobu-config:

Change byobu settings with byobu-config

You can change Byobu’s colors if you’d like. I like the “dark” theme.

The important setting to change is the last one: “Byobu currently does not launch at login“. Select that and press return. Now you can go back to the menu and exit byobu-config (tab to the exit choice and hit return).

Now log out and log back in again. You’ll now access Byobu automatically.

Byobu as your session manager

The new MOTD (that’s “Message of the Day”, which is the text that is displayed when you login) has system information, thanks to landscape-common. This will let you know if your almost out of memory or disk space, or if you’re under heavy load.

Now let’s look at the information Byobu provides. The bottom row of Byobu contains, from left to right, your Distribution logo and name, uptime, load, cpu frequencies, ram amount and usage, and the date.

The upper row are your windows. Press F2 to create a new window. Along the bottom of the screen you’ll now see two entries in the top row of the footer. Press F3 to move to the previous window and F4 to move to the next window.

Press F8, then give the window a name, like “echo”. I like to name each tab with the task I’m doing, like “console”, “mysql”, and “top”.

You can press F7 to enter copy/scrollback mode. Use the arrow and page-up/down keys to move around your session. Press the spacebar to start copying. Now move your cursor somewhere else. Then press space again to copy the selection. Byobu doesn’t have a paste hotkey, so you have to use the screen hotkey, which is CTRL-a ] (thats CTRL-a, then let go of control and press ]).

You can press F12 to quickly lock your terminal, to grab a quick cup of coffee (or read this blog!).

Let’s explore how to maintain a session. Issue the following command:

echo hi

That will print out “hi” on the screen. Now let’s exit this session. Press the Byobu hotkey F6 to “Detach” from Byobu.

You’ll notice you’re still logged in, this is because Byobu is an application that runs on top of your session. So now you have to exit again to get back to your local machine.

Now SSH back into your machine. You’ll notice that command is still on the screen! “Detaching” does not end your session, it just detaches you from it. That means you can log in, run a long running task, and then Detach and come back in the morning when your task is done.

For extended information on Byobu, check out its Ubuntu Man Page.

• Replacing your current ruby installation with a patched version
• Installing a gem to patch the code and then updating your application to load the patch.

The latter approach only postpones the inevitable, so I opted for the former. If you’re using the debian packages, it’s a simple call to aptitude update. However, if you’ve installed from source, this presents a few challenges.

The upgrade is actually quite straightforward. According to the REE Documentation, you simply need to run the new installer over top of your previous installation. This includes specifying the exact same directory name that you used to install the previous version.

As you may know, REE installs by default into /opt/ruby-enterprise-1.8.6-<year_month_day_release>. However, after completing this, you’re left with a directory name that doesn’t reflect what’s actually installed in there. If you’ve a little neurotic like I am, you’re going to want to rename that directory. However, be careful. Your Passenger configuration is likely pointing to the directory that you want to rename, so you’ll have to remember to update your apache/nginx configuration to reflect this.

When we install REE we create a symlink at /opt/ruby which points to the version of ruby that we want passenger to run. This allows us to quickly swap in/out different versions of ruby for testing/upgrading/etc.

Considering we have dozens of client applications running on distinct servers each with their own versions of REE installed, I decided to write a script to minimize the manual effort required to upgrade several production servers.

#!/bin/sh
# Author: John Trupiano
# Script to upgrade an REE installation on a hot server and maintain sane directory names
 
if [ "$(whoami)" != "root" ]; then
  echo "You need to be root to run this!"
  exit 2
fi
 
RF_RELEASE=58677
REE_VERSION=20090610
REE=ruby-enterprise-1.8.6-$REE_VERSION
URL=http://rubyforge.org/frs/download.php/$RF_RELEASE/$REE.tar.gz
 
# Determine what the most recent version of REE is that is installed
MOST_RECENT_REE_VERSION=`ls /opt | awk -F"-" '$4 > max && $2 == "enterprise" { max=$4; maxline=$0 }; END { print max }'`
MOST_RECENT_REE=ruby-enterprise-1.8.6-$MOST_RECENT_REE_VERSION
 
WORKING_DIR=/root/src
 
echo "Going to update $MOST_RECENT_REE to $REE"
 
echo "Back up previous release"
cp -R /opt/$MOST_RECENT_REE /opt/$MOST_RECENT_REE.bak
 
echo "Download new release"
mkdir -p $WORKING_DIR
cd $WORKING_DIR && wget $URL
 
echo "Untar and install over the previous release for 'upgrade' according to REE manual"
tar xzf $REE.tar.gz
./$REE/installer --auto /opt/$MOST_RECENT_REE
 
echo "Shuffle folder names to remain sane"
mv /opt/$MOST_RECENT_REE /opt/$REE && mv /opt/$MOST_RECENT_REE.bak /opt/$MOST_RECENT_REE
 
echo "Actually symlink in the new version of REE"
rm /opt/ruby && ln -s /opt/$REE /opt/ruby
 
echo "Clean up after ourselves"
rm -f $WORKING_DIR/$REE

Simply upload this script to your server, su – up to root and run it. Then restart apache/nginx so that passenger will pick up the new version of ruby. All of your gems will be maintained, and your most recent version of REE will remain in tact at /opt/ruby-enterprise-1.8.6-<old_version> in case you need it again.

View the script on github

2) Run:
> tune2fs -o journal_data_writeback /dev/sdXY
Where /dev/sdXY is replaced by the partition that you want to boost

4) Edit fstab

> nano -w /mnt/sdXY/etc/fstab

Find the line that references sdXY. It will look something like:

# /dev/sda2
UUID=be2f0ac2-4683-4550-bcd1-704a1a840b3e / ext4 relatime,errors=remount-ro 0 1

The first entry is the UUID (although on your system this could just be /dev/sdXY). The second entry is the path (/ for me). Third is the fstype (ext3/4). Fourth are the options. Fifth is for dump and sixth is pass. See man fstab(5) for more info.

Change the options to:

noatime,data=writeback,barrier=0,nobh,errors=remount-ro

(you can leave all of yours in place, if they weren’t the same as mine.

The main ones are replacing atime/relatime with noatime. This causes the FS to not write read-times to a file when read. Think about it. Writing to the FS for every read of the FS? crazy!

Next is data=writeback. This means that metadata for files can be written lazily after the file is written. This will not cause file system corruption, but it may cause the most recent changes to be lost in the event of a crash (so you may jump back into the past a bit).

Next is barrier, which is slightly more dangerous:

barrier=<0|1(*)> This enables/disables the use of write barriers in
the jbd code. barrier=0 disables, barrier=1 enables.
This also requires an IO stack which can support
barriers, and if jbd gets an error on a barrier
write, it will disable again with a warning.
Write barriers enforce proper on-disk ordering
of journal commits, making volatile disk write caches
safe to use, at some performance penalty. If
your disks are battery-backed in one way or another,
disabling barriers may safely improve performance.

Next is nobh:

bh (*) ext4 associates buffer heads to data pages to
nobh (a) cache disk block mapping information
(b) link pages into transaction to provide
ordering guarantees.
“bh” option forces use of buffer heads.
“nobh” option tries to avoid associating buffer
heads (supported only for “writeback” mode).

You can skip barrier and nobh if you’d like. noatime and data=writeback are the big ones.

6) Reboot to your system.

If you have any trouble booting, just boot a recovery disk and revert the fstab changes.

EDIT: Updated to no longer require recovery disk booting thanks to Nicolas Alpi’s response post.