Author: paul.deluca

Can I safely ignore any mismatch_cnt errors on my Fusion-io array?

The root user of a system using Linux software RAID-1 or RAID-10 of Fusion-io ioDrives (aka IBM High IOPS SSD Adapters) may occasionally receive a message similar to the following:

/etc/cron.weekly/99-raid-check:

WARNING: mismatch_cnt is not 0 on /dev/md0

For RAID-1 and RAID-10 arrays using Linux software RAID, these messages are normal and considered a known bug in Red Hat Enterprise Linux 5. These messages will be removed in a future release of the operating system.

Reference:

How do I change a UniVerse tuning parameter?

The current settings for the configurable parameters are stored in the uvconfig file, located in the UV account directory (/u2/uv). A command called uvregen uses the uvconfig file to create another file in the UV account directory named .uvconfig, which contains an encrypted version of the current configurable parameter settings. The .uvconfig file is used during the startup of UniVerse.

When you run the uvregen program, it does the following:

  • Verifies that the values in the uvconfig file are reasonable
  • Creates a new .uvconfig file in the UV account directory
  • Resets the master key if you are using encryption

To change configurable parameter values, you can edit the uvconfig file manually. When you change configurable parameter settings, you must save them in the uvconfig file. You must restart UniVerse for the new settings to take affect.

Make sure you understand the effect of any changes you make to the default values. A small change can have a significant impact on your system. It is impossible to document the impact of each of these parameters so that you can predict the effect of a value change. If you are not sure about how to change the value of any parameter, please consult with a UniVerse support specialist.

To change the value of a configurable parameter, complete the following steps:

Stop Eclipse services that rely on UniVerse (SYSTEM.ADMIN, JBoss):

Change the working directory to the UV account directory (make sure the UV account directory is in the current path).

cd /u2/uv

Make sure all users are logged out of UniVerse and stop all UV processes:

bin/uv

Press ESC
At the prompt, type:

MASTER OFF ALL
Q

Shut down UniVerse

bin/uv -admin -stop
Before making changes to a configuration file, always make a backup.

Edit the uvconfig file in the UV account directory using your preferred text editor (for example, vi):

vi uvconfig

When you’ve saved the changes, apply them by executing the UniVerse uvregen program:

bin/uvregen

Start up UniVerse:

bin/uv -admin -start

Start up any dependent services (SYSTEM.ADMIN, JBoss, etc.):

Resources:

How do I shrink a Linux filesystem?

WARNING: storage maintenance, especially reducing filesystems, can potentially result in errors or data loss. Plan and act accordingly. Always create backups and follow best practices.

For this example, we’ll be using the standard /dev/datavg/eclipse LV that’s mounted at /u2/eclipse. We’ll be shrinking the filesystem to 50GB.

Begin by unmounting the file system:

umount /u2/eclipse

Force a filesystem check:

e2fsck -f /dev/datavg/eclipse

Reduce the file system. The number specified here is the final, target size of the filesystem, not the amount by which it is reduced.

resize2fs /dev/datavg/eclipse 50G

The logical volume must also be reduced by the same amount:

lvreduce -L 50G /dev/datavg/eclipse

Mount the filesystem and run df to verify the change was made:

mount /u2/eclipse
df -h /u2/eclipse

Resources:

  • https://access.redhat.com/kb/docs/DOC-5531
  • http://blog.shadypixel.com/how-to-shrink-an-lvm-volume-safely/

How do I configure an IBM dumb terminal?

NOTE: The IBM 3151 or 3153 dumb terminals are end-of-life and therefore not supported by IBM or Eclipse. If you need a terminal for your server, please see Where can I buy an IBM dumb terminal for my IBM pSeries server?

How to configure an IBM 3151 or 3153 dumb terminal (aka console):

  • Press Control-“-“ (the minus key) to get to the setup screen
  • Verify the following settings:
  • Emulation = Wyse 50+
  • Enhanced = on
  • Comm Mode = Full duplex
  • Host/Printer = EIA/None
  • EIA Baud Rate = 19200 (9600 for older servers)
  • AUX Baud Rate = same as EIA
  • Language = US
  • EIA Data Format = 8/1/N
  • Aux Data Format = 8/1/N
  • Sessions = One
  • To save the settings and exit, press Control-“-“ (the minus key)

Note: the keys pressed must be the control in the lower-left part of the keyboard and the minus sign in the upper-right (numeric keypad).  If CONTROL-MINUS does not work, try CONTROL-SETUP or CONTROL-SCROLL-LOCK.  To navigate around, use the arrow key.  The space bar will present a list of options.

How do I view or reset the attention indicator on my IBM AIX server?

The attention indicator (amber LED) is lit on an IBM AIX server to notify the administrator of potential problems. Once you have verified that no problems exist through running diagnostics or analysis of the errpt, you may reset this indicator.

To disable the attention indicator on your IBM AIX server:

  • Log into the server as root via Eterm (telnet), PuTTY (telnet, ssh), the HMC or the local console.
  • Launch diagnostics: 
    diag
  • Press ENTER to continue
  • Select Task Selection
  • Select Identify and Attention Indicators
  • Select Set System Attention Indicator to NORMAL
  • Press ESC-7 or F7
  • Press ESC-0 to exit

For assistance with running diagnostics, or to answer any questions regarding your IBM server, you may contact IBM directly.

References: