I'm setting up a new IDS 11.7 instance on a new server. What is the proper way to create informix rawdevices in Red Hat Enterprise Linux 5.5 (32-bit X86)?

method #1: ln -s /dev/cciss/cxdxpx /etc/udev/devices/dbsname
add "informix" group and user ownership in /etc/udev/rules.d/50-udev.rules
add "0660" perms in /etc/udev/rules.d/50-udev.rules
start_udev (starts udev and creates rawdevices)
raw -qa (check if raw devices have been created)

method #2: edit /etc/sysconfig/rawdevices and add raw device bindings in there
ln -s /dev/raw/rawX /opt/IBM/informix/devices/dbsname
service rawdevices start
chkconfig rawdevices on

method #3: add device name entries in /etc/udev/rules.d/60-raw.rules
ACTION="add", KERNEL=="<device name>", RUN+="raw /dev/raw/rawx %N"
set ownership and permissions on rawdevices in /etc/udev/rules.d/60-raw.rules
ACTION="add", KERNEL=="raw*", OWNER="informix", GROUP="informix", MODE="0660"
ln -s /dev/raw/rawX /opt/IBM/informix/devices/dbsname
start_udev (starts udev and creates rawdevices)
raw -qa (check if raw devices have been created)


DISCLAIMER

This transmission and the information contained herein and/or in any attachments hereto and/or in any attachments thereto is privileged and/or confidential and is intended ONLY for the use of the individual(s) and/or entity(s) named above. If you are not the intended recipient, you are strictly prohibited from disclosing, printing, copying, using or disseminating this transmission and/or any such attachments and any information contained in this transmission and/or in any such attachments. ANY unauthorized interception of this transmission and/or any such attachments is a violation of federal criminal law.

If you have received this transmission in error, please notify the sender immediately and delete the transmission and all such attachments

Any of those will get the job done. It just depends on what your site
standard is.

Art

Art S. Kagel
Advanced DataTools (www.advancedatatools.com)
IIUG Board of Directors (art (AT) iiug (DOT) org)
Blog: http://informix-myview.blogspot.com/

Disclaimer: Please keep in mind that my own opinions are my own opinions and
do not reflect on my employer, Advanced DataTools, the IIUG, nor any other
organization with which I am associated either explicitly, implicitly, or by
inference. Neither do those opinions reflect those of other individuals
affiliated with any entity with which I am affiliated nor those of the
entities themselves.



On Thu, Nov 18, 2010 at 7:39 PM, June Nebab LKINY <jnebab (AT) idealcut (DOT) com>wrote:

This is unnecessary and will quite possibly be slower. See other
recent thread:

https://groups.google.com/group/comp...a4908ebbadb977

Regards,
Doug Lawry

On Nov 19, 7:21*pm, Doug Lawry <la... (AT) nildram (DOT) co.uk> wrote:
Hi Doug,

That htread has Art Kagel saying that raw devices are still faster,
why did you find it slower?

Hi David.

Sorry - I only just saw your last posting on this. We found bulk
operations such as a level 0 "ontape" either unchanged or slower with
raw compared to block devices on recent Linux versions. This surely is
expected: with IO_DIRECT, block devices should be fine and there is no
need to go via another pseudo device type which is in any case now
deprecated.

Cheers,
Doug

"Doug Lawry" <lawry (AT) nildram (DOT) co.uk> wrote

Do you have any comparative figures for cooked files also, Doug?

cheers

No, Neil, as I don't use cooked files on principle!