All,
I'm having an intermittent problem with Informix V. 10.xx that arose
about a week ago.
I know I know... I need to upgrade. 8^) In fact a new IBM server
has been quoted.

From time to time it randomly takes about 3-7 seconds to establish a
connection the the DB engine. Once connected the selects all run fast
without latency...as usual.

This connection latency occurs with ODBC connections as well as telnet
dbaccess connections, even from the console.

There are no listenerr thread errors, and user usage has not
increased.
The system has been running for about six years w/o performance
issues.
No recent changes have been made to the system.

The system is a IBM 550 running AIX V5.

Any thoughts?

Thanks in advance for you time.
Steve

Are the dbaccess connections using shared memory or network? What are the
NETTYPE settings in your ONCONFIG file?

Art

Art S. Kagel
Advanced DataTools (www.advancedatatools.com)
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 Wed, Aug 17, 2011 at 12:39 PM, Ripshot <se.bailey (AT) verizon (DOT) net> wrote:

On Aug 17, 12:09*pm, Art Kagel <art.ka... (AT) gmail (DOT) com> wrote:

Thanks for the response.

The dbaccess connections are shared memory.
I've added snippets of the onconfig and sqlhosts files below.

INFORMIXSERVER=ace_ibmserver

ONCONFIG file entries:
DBSERVERNAME aceibm_shm
NETTYPE ipcshm,2,125,CPU # Configure poll thread(s) for
nettype
NETTYPE soctcp,10,40,NET # Configure poll thread(s) for
nettype

SQLHOSTS file entries:
aceibm_shm onipcshm aceibm aceibm_shm

Thanks,
Steve

I assume that there is an onsoctcp entry in the sqlhosts file as well or
delays in the ODBC connections would be the least of your problems ;-)

I don't see anything obvious. I don't know that I would use 10 TCP poll
threads for only 400 connections, I might rather use two or three with
NETTYPE soctcp,3,150 but that's not likely to cause slow response times.

Art
Art S. Kagel
Advanced DataTools (www.advancedatatools.com)
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 Wed, Aug 17, 2011 at 3:08 PM, Ripshot <se.bailey (AT) verizon (DOT) net> wrote:

This is a shot in the dark, but almost every issues at connect time come
from reverse DNS issues.
Are you using /etc/hosts or a DNS to reverse DNS? If using /etc/hosts, are
all the possible client entries there?
If using DNS, what is the timeout?. During that periods, do a "netstat -a"
works smoothly or does it also "hang"?

Regards.

On Wed, Aug 17, 2011 at 5:39 PM, Ripshot <se.bailey (AT) verizon (DOT) net> wrote:



--
Fernando Nunes
Portugal

http://informix-technology.blogspot.com
My email works... but I don't check it frequently...