It's all about the answers!

Ask a question
QuestionsTagsUsersBadges
Follow this question
Question details

×42,787

question asked: Feb 10 '11, 11:23 a.m.
question was seen: 6,339 times
last updated: Feb 10 '11, 11:23 a.m.

Related questions

FAQ - How this Forum works

Implement RTC High Availability

0
Guowei Jim Hu (1.0k●8●103●53) | asked Feb 10 '11, 11:23 a.m.
Spent five weeks and raised more than 10 PMRs against WAS, RTC and WASIHS and finally get the RTC HA implemented and working on a set of test servers including one IBM HTTP server, two RTC/WAS servers and two DB2 servers follwoing the instructions from http://jazz.net/library/article/207 and http://jazz.net/library/article/464.

I have to say that the documentation can be improved a lot so the next crab-eater won't have to spend the much time I had to. One may argue that most of the setup are not RTC, but WebSphere, IBM HTTP and Web Plugin and DB2. But the goal is to get RTC HA failover work in this complex infrastructure so it would be appreciated by many a lot if the documentation can be added with more details about configuration with the non-RTC components instead of just direct us to a link which just handles general cases, not RTC specific ones like HA failover.

Here are a few missing key points from the documentation I identified to share with the next explorer who dare to try it:

1. After you follow the instruction and install the web plugin on the web server using your primary RTC server's host name, you'll get a generated script and need to copy it over to your primary RTC server and run it there.
When you start to configure your standby RTC server, you have to wonder what to do: to install the web plugin again under the standby server's host name to gennerate the script to use or just use the one already created before? And the solution is to install the web plugin just once and use the same script for both RTC servers;

2. After you run the script above on your RTC server, you can manage the web server from the WAS admin console and you are instructed to regenerate the web plugin configuration file and propagate it bake to the web server. But the issue is you have two RTC servers, evertime you propagate the file from one of them, what you have on the web server from the other propagation is overwriten and you can't get a configuration working for both server.
Teh solution is to forget about the propagation and jusy go back to the web server and manually modify the plugin configuration file to handle both RTC servers;

3. There are instructions asking you to modify the web.xml file on your RTC primary server to replace all "CONFIDENTIAL" to "NONE" but you have to remember to do the same on your standby server otherwise you can get the snoop test work but no luck on RTC connection via the web server;

4. If you want to enable SSL on your web server, then you'll never get it work until you manually install the root singer certificate from your default trust node on your RTC server into the web plugin's keystore on your web server.


There are many more but these are the few major blockers which got me stuck for hours and days.

4 answers

Most liked answers ↓|Newest answers|Oldest answers

0
permanent link
Geoffrey Clemm (30.1k●3●30●35) | answered Sep 13 '11, 1:40 a.m.
FORUM ADMINISTRATOR / FORUM MODERATOR / JAZZ DEVELOPER
It's there ... (I checked just now).
What error message did you get when you tried to access it?

Cheers,
Geoff

On 9/12/2011 2:38 PM, schneidg wrote:
I cannot find article

http://jazz.net/library/article/464.

Has it moved?
0
permanent link
Guido Schneider (3.4k●14●86●115) | answered Sep 12 '11, 2:30 p.m.
I cannot find article

http://jazz.net/library/article/464.

Has it moved?
0
permanent link
Guowei Jim Hu (1.0k●8●103●53) | answered Mar 04 '11, 2:52 p.m.
A friendly warning to those who want to follow the existing RTC HA implementation instructions!!!

We've confirmd with the author of the documentation that the assumption is you start with fresh new RTC servers without data in repository and you need to set the Plublic URL Root property to point to the host name of the front-end HTTP web server.

It is too late for us now as we've been using RTC for years before implementing this solution and we are working with RTC support to find a fix to allow us to keep the current implementation as it is already on production servers.

If anyone on the same boat with us, please share your experience.
0
permanent link
Geoffrey Clemm (30.1k●3●30●35) | answered Feb 10 '11, 9:08 p.m.
FORUM ADMINISTRATOR / FORUM MODERATOR / JAZZ DEVELOPER
Please file a work item against either "Documentation - Website" if you
want those articles to be updated or "Documentation - User Assistance"
if you'd like to see this info added to the on-line help.

The more details you can add to the work item, the better!

Cheers,
Geoff

On 2/10/2011 11:23 AM, ghu wrote:
Spent five weeks and raised more than 10 PMRs against WAS, RTC and
WASIHS and finally get the RTC HA implemented and working on a set of
test servers including one IBM HTTP server, two RTC/WAS servers and
two DB2 servers follwoing the instructions from
http://jazz.net/library/article/207 and
http://jazz.net/library/article/464.

I have to say that the documentation can be improved a lot so the
next crab-eater won't have to spend the much time I had to. One may
argue that most of the setup are not RTC, but WebSphere, IBM HTTP and
Web Plugin and DB2. But the goal is to get RTC HA failover work in
this complex infrastructure so it would be appreciated by many a lot
if the documentation can be added with more details about
configuration with the non-RTC components instead of just direct us
to a link which just handles general cases, not RTC specific ones
like HA failover.

Here are a few missing key points from the documentation I identified
to share with the next explorer who dare to try it:

1. After you follow the instruction and install the web plugin on the
web server using your primary RTC server's host name, you'll get a
generated script and need to copy it over to your primary RTC server
and run it there.
When you start to configure your standby RTC server, you have to
wonder what to do: to install the web plugin again under the standby
server's host name to gennerate the script to use or just use the one
already created before? And the solution is to install the web plugin
just once and use the same script for both RTC servers;

2. After you run the script above on your RTC server, you can manage
the web server from the WAS admin console and you are instructed to
regenerate the web plugin configuration file and propagate it bake to
the web server. But the issue is you have two RTC servers, evertime
you propagate the file from one of them, what you have on the web
server from the other propagation is overwriten and you can't get a
configuration working for both server.
Teh solution is to forget about the propagation and jusy go back to
the web server and manually modify the plugin configuration file to
handle both RTC servers;

3. There are instructions asking you to modify the web.xml file on
your RTC primary server to replace all "CONFIDENTIAL" to
"NONE" but you have to remember to do the same on your
standby server otherwise you can get the snoop test work but no luck
on RTC connection via the web server;

4. If you want to enable SSL on your web server, then you'll never get
it work until you manually install the root singer certificate from
your default trust node on your RTC server into the web plugin's
keystore on your web server.


There are many more but these are the few major blockers which got me
stuck for hours and days.

Your answer

Register or to post your answer.


Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.