Me in IT Linux and Open Source consultancy in infrastructural environments

i have investigated some time to look at the performance of Solaris with Oracle on it. Here is some information that might be relevant.

maxcontig:
definition: maximum contiguous i/o size

• check:
  (input) # fstype -v /dev/dsk/c0t0d0s0 | grep maxcontig
  (output) maxcontig 128 rotdelay 0ms rps 167
• default:
  (output) maxcontig 128 rotdelay 0ms rps 167
• change:
  for existing filesystem:
  (input) # tunefs -a 256 /dev/dsk/c0t0d0s0
  for new filesystems:
  (input) # newfs -C 256 /dev/rdsk/c0t0d0s0

maxphys:
definition: largest physical i/o operation allowed

• check:
  (input) # adb -k
  (input) maxphys/D
  (output) maxphys: 131072
• default:
  (output) maxphys: 131072
• change:
  add or change in /etc/systems and reboot
  (input) set maxphys 1048576
  when using a metadevices, also set this one same value as "set maxphys value".
  (input) set md:md_maxphys 1048576

directio:
definition: ufs feature to bypass the buffer-chache

• check:
  (input) # mount | grep directio
• default:
  not enabled.
• change:
  change options in /etc/vfstab and reboot:
  (input) /dev/md/dsk/d0 /dev/md/rdsk/d0 / ufs 1 no logging
  directio
  or manually mount the filesystem:
  (input) mount -o directio /dev/md/dsk/d0 /

tcp_time_wait_interval
definition: the time in milliseconds a tcp connection stays in time_wait state

• check:
  (input) # ndd /dev/tcp tcp_time_wait_interval
  (output) 240000
• default on solaris 8 systems:
  (output) 240000
  default on solaris 9 systems:
  (output) 60000
• change:
  (input) # ndd /dev/tcp -set tcp_time_wait_interval 60000\r\n

here are some interesting settings that i tested. it seemed to work, so here you are.

/etc/system
set nfs:nfs_nra=10
set nfs:nfs3_nra=10
set nfs:nfs_max_threads=24
set nfs:nfs3_max_threads=24
set ncsize=8192
set tcp:tcp_xmit_hiwat 65536
set tcp:tcp_recv_hiwat 65536
set nstrpush=9
set rlim_fd_cur=1024
set rlim_fd_max=1024
set sq_max_size=0x19
ce:ce_taskq_disable=1

/etc/vfstab
netappfiler:/some/volume - /a/mountpoint nfs no yes hard,intr,bg,proto=tcp,vers=3,rsize=32768,wsize=32768,largefiles

/platform/sun4u/kernel/drv/ce.conf
interrupts=1;
accept-jumbo=1;

Here is a way to use ssh-agent, ssh-add without having to type you password every session. Just once (when the machine has been rebooted) you will have to enter you passphrase, from then on you are authenticated. Here is the script you will have to add to your .bashrc or .profile in you homedirectory:

Have fun, I have been using this great hint for quite some time now, it seems to work perfectly!

here is a very simple cool trick. Imagine you would have a file (test.txt) with a few million lines in it looking something like this:

Imagine you would like to know how many lines there are with "123" in the second column. try something like this:

This will open the file test.txt ("cat test.txt") and give the results to awk ("|awk"). awk will look for 123 in column 2. ("match ($2,/123/)". these results are passed to wc, a counter, that can count lines with the -l option. ("| wc -l")

Think of another example; you'd like to find all pids where the connected tty is ttys000:

Here is a very simple script illustrating the use of expect:

Have you ever wondered how to get variables from a command line in a decent way?

Imagine you wrote a script that must check certain files how easy would it be to just enter the file on the command line interface as an option like this:

Here is some code to do just that. You will need to change the options according to your wishes.

Here is a cool script, used to backup this site and a few others hosted here. The script should be self explanatory.

Some modifications have been made for publishing, not to reveal to much information...

Here is a script to check if your Linux box is ready for the US DST 2007 Energy Act. Save it, chmod it, run it.

#!/bin/bash

status=0

glpackagecheck() {
 rpm=$(rpm -q tzdata)
 case $rpm in
  tzdata-2005m*)
   :
  ;;
  tzdata-2006*)
   :
  ;;
  *)
   echo "RPM tzdata-2005m-1 is not installed. You installed: $rpm"
   status=$(($status+1))
  ;;
 esac
}

glfilecheck() {
 if [ ! -d /usr/share/zoneinfo/America/ ] ; then
  echo "The /usr/share/zoneinfo/America directory does not exist."
  status=$(($status+4))
 fi
}

glfunctioncheck1() {
 storedtz=$(echo $TZ)
 export TZ="America/Adak"
 date | grep 'HA.T' > /dev/null
 if [ $? -gt 0 ] ; then
  echo "Unable to switch timezone."
  status=$(($status+8))
 fi
 export TZ="$storedtz"
}

glfunctioncheck2() {
 zdump -v CST6CDT | grep 2007 | grep "Mar 11" > /dev/null
 if [ $? -gt 0 ] ; then
  echo "Wrong roll over date configured for CST6CDT."
  status=$(($status+16))
 fi
}

glpackagecheck
glfilecheck
glfunctioncheck1
glfunctioncheck2

if [ "$status" -gt 0 ] ; then
 echo "Not all tests ended with success, the status is: $status."
else
 echo "All test successful, you are prepared for the US DST change."
fi

Here is a script to rotate any logfile. Optionally a filesize can be given. It contains some extra features that are not (yet) in the logrotate facility.

I am not sure it works, but hey, you can tweak it! ;-)