RE: IBM HTTP Server -- MIDRANGE-L

Are you running multiple IP addresses on one interface card?  Perhaps that
is causing the problem.  I like to define one IP address to one interface
and have at least 2 interface cards, then use Virtual IPs.  Set routes in
your primary router to each VIP address using both physical addresses.
Split the primary/secondary routes.

i.e..
Lin01 = 10.10.1.1 Physical
Lin02 = 10.10.1.2 Physical
VIP01 = 10.1.1.1
VIP02 = 10.1.1.2

Route for: VIP01 = 10.10.1.1 primary
                 = 10.10.1.2 secondary

           VIP02 = 10.10.1.2 primary
                 = 10.10.1.1 secondary

This helps to balance the load across your interfaces.  Each interface is on
a separate IOP in a separate BUSS.  Each interface is plugged into a
separate switch/hub.

Of course if you have only one physical interface, try using Virtual IPs for
your servers instead of multiple IPs on one interface.  Then setup your
routes.

There is more information on TCP/IP performance in the Redbooks.
AS400 Routing and Load Balancing Techniques might help.  I cannot find the
one I am thinking about at this time.



Christopher K. Bipes    mailto:ChrisB@Cross-Check.com
Sr. Programmer/Analyst  mailto:Chris_Bipes@Yahoo.com
CrossCheck, Inc.        http://www.cross-check.com
6119 State Farm Drive   Phone: 707 586-0551 x 1102
Rohnert Park CA  94928  Fax: 707 586-1884

If consistency is the hobgoblin of little minds, only geniuses work here.
Karen Herbelin - Readers Digest 3/2000

-----Original Message-----
From: MEovino@ESTES-EXPRESS.COM [mailto:MEovino@ESTES-EXPRESS.COM]
Sent: Tuesday, November 21, 2000 9:36 AM
To: MIDRANGE-L@midrange.com
Subject: IBM HTTP Server



It had been performing pretty nicely for me up until now.  The two specific
changes we've made recently are:

1.  Moving the server from port 8080 to port 80
2.  Turning BindSpecific on so it would bind to one IP address (on port
80), and binding our Domino HTTP server to another IP address (also on port
80)

It still works, but it's pretty slow at times.  If I look at the active
jobs for QHTTPSVR, they look like this (set mail reader to fixed width
fonts for this to make sense):

Opt  Subsystem/Job  User        Type  CPU %  Function        Status
     QHTTPSVR       QSYS        SBS      .0                   DEQW
       DEFAULT      QTMHHTTP    BCH      .2  PGM-QZHBHTTP     CNDW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW
       DEFAULT      QTMHHTTP    BCI      .0                   TIMW

That first job under QHTTPSVR is ALWAYS in CNDW.  I don't remember if this
used to be the case before the changes.  CPU utilization on the box is
greatly DECREASED after the changes (generally around 20%), so I don't
think it's that.

Anyone have any ideas?
+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---