Best Practice to send AQ message to over 30 databases

Hello, Hoping someone can help. I'm running 11gR2 AQ and wondering what the best practice is to send a message to over 30 databases every few seconds. Due to setup of customer sites this is our only option. However, All databases are in the same data center. I'm thining about doing the multi-consumer queue and have each of the 30 databases subscribe. I believe I can setup an individual propagation queue queue_to_queue for each one but wondering if this is the best thing to do. Also, I was reading about some type of call back (Which I saw on Tom Kyte's website)....but wondering if this would work for 30 db's.
Anyone have experience with this configuration it would be great to hear your ideas. Thanks

Can't you just enqueue to a multi consumer queue and have each subscriber subscribe to that queue?
Also, is there TTL with AQ's such that a subscriber only has a limited window with which it can expect to get the message from the multi consumer queue?

Similar Messages

How to send error message to forms from Database Trigger

Hi, Please help me to send error message to forms from Database Trigger?
RgDs,
Madesh.R.M

You are correct, the On-Error trigger is a Forms trigger. However, if your Form is going to display the error generated by the database stored procedure or trigger - you might not see the database error in your Form unless you check the DBMS_ERROR_CODE in the On-Error trigger and manually display the Error Code and associated Text. I've see this happen with a co-worker. The Form she was working on was based on a table with an Before-Insert trigger. Because she was not explicitely handling the error from the Before-Insert trigger in the Forms On-Error trigger, her Form appeared to halt for no reason at all. Once she added code to the On-Error trigger in the Form to handle the DBMS_ERROR_CODE, she discovered the trigger was producing an error and was able to show the error to the user in the On-Error trigger.
I understand the desire to keep as much as possbile in the database, but with that comes some extra coding in your Forms to handle this. This extra coding could easily be placed in a Forms Library, attached to a Form and called in the On-Error trigger. Your code could look like this:
DECLARE
   /*This example assumes you have an Alert defined
      in your Form called: 'ERROR' */
   al_id    ALERT;
   al_text VARCHAR2(200); /* Max text of a Forms Alert message*/
   al_btn   NUMBER;
BEGIN
IF DBMS_ERROR_CODE != 0 THEN
   /* Error code is ORA-00000 Normal Successful completion
       So only handle non-zero errors */
   al_text := DBMS_ERROR_CODE||':'||DBMS_ERROR_TEXT;
   al_id := Find_Alert('ERROR');
   set_alert_property(al_id, alert_message_text, al_text);
   al_btn := show_alert(al_id);
END IF;
END;Your original question was "How to send error message to forms from Database Trigger?" The answer is you don't because Forms already gets the database error code and database message through the Forms DBMS_ERROR_CODE and DBMS_ERROR_TEXT functions. Look these up in the Forms help and it should clear things up for you.
Craig...
Edited by: CraigB on Jun 14, 2010 4:49 PM
Edited by: CraigB on Jun 14, 2010 4:49 PM
Edited by: CraigB on Jun 14, 2010 4:50 PM
Edited by: CraigB on Jun 14, 2010 4:51 PM
Edited by: CraigB on Jun 14, 2010 4:51 PM

Best practice ? send Object to request or desired pieces of data?

Newbie to this style of programming...
Is the best practice to put the customer object in the session or request object and allow jsp's to declare and instantiate the customer object out of the session/request (then use whatever getters are needed to properly display the desired data)?
Or would it be better to send the customer ID, Name, address etc as a string to the session/request object and just have the JSP declare the strings & instantiate from the session/request object(thus keeping more complicated java code out of the jsp)?
Thanks for the help in advance!

Doesn't this result in more code? If we send the object, we need code to declare and instantiate the object, then use the getters to get the data to display.
If I just send the necessary data, I just need to declare a string = request.getParameter... or just display the request.getParameter.
I actually like the concept of sending the object, it seems cleaner and less likely to result in servlet changes down the road, but i want to make sure there is not some other reason NOT to do this.

Best Practices for Using Service Controller for Entity Framework Database

I'm running into an issue in my first time creating a Web Service with a .NET backend with Azure. I designed a database in Entity Framework and had it create the models, but I couldn't create a controller for the table unless I made the model inherit from
EntityData. Here's the catch, the Database Model has int Id, but EntityData has string Id, so, of course, I'm getting errors. What is best practice for what I'm trying to do?
Michael DiLeo

hi Michael,
Thanks for you posting!
Sorry for I am not totally understanding your issue. Maybe two points need your confirm:
1. I confuse with the "Service controller"? IS your meaning MVC controller? Or ServiceController(http://www.codeproject.com/Articles/31688/Using-the-ServiceController-in-C-to-stop-and-start
2.whether The type of ID in the model is match to the database ? In other words, Is the type of IDin .edmx matched to the database?
By the way, it seems that this issue is more related to EF. You could post this issue on EF discussion for better support.
Thanks & Regards,
Will
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

Best practice to work with Sybase 12.5.3 database version

Hi all,
Is there any document or information about best practices to access a Sybase 12.3 version from universe on BOXIR2?
Thanks.

Hi Marlon,
Have a look to Product Guide of BOXIR2 for [Data access|http://help.sap.com/businessobject/product_guides/boexir2/en/xir2_data_access_guide_en.pdf] see if it helps.:
Regards,
Shweta

Best Practice to migrate existing Message/Central Instance to new hardware?

I've done some searching for this, but it all seems to be fairly old (I found one decent thread from 2008, but it just seemed outdated). I would like to get some updated info & suggestions on a process to follow for this task.
Here is some info about the current config:
I have an existing SAP ERP 6.0 EHP5 system. SID "XYZ"
Everything is currently running on Windows x64.
ABAP only config.
It is currently setup as a distributed system.
DB server is SQL, which is running on its own hardware as a dedicated DB server. I don't need to touch this.
Uses logon groups to distribute the load between 3 App servers.
One of the App Servers is setup as the Message/CI server. This is what I am wanting to move to VM hardware.
The other two App Servers are setup just as Dialog Instances. VM already. I don't need to touch these.
What I am wanting to do:
I am wanting to migrate the existing Message/CI server to VM hardware to get the current Message/CI system off of very old hardware.
I need to keep the system with the same name & IP.
I do not need/want to mess with the DB server at all.
Need to minimize downtime as much as possible.
I do NOT want to go to an HA environment. That is a different project for the future.
Suggestions? I would like to hear what some of you have done.
Minimize downtime? I have a short maintenance window of a few hours coming up in the next few months, and would like to do this changeover during that downtime.
How do I handle setting this up on the new hardware using the same name & IP, especially with the existing system still running.
What I think I need to do: (Some of these may be out of order, which is why I'm asking for suggestions)
Setup new Windows VM server, named differently than current Message/CI server (Done already).
Run SAPINST to run Global Host Preparation on the new Windows VM server.
Run SAPINST to setup ASCS Instance on the new Windows VM server.
Skip the Database Instance installation, since I already have one running.
Run SAPINST to setup Central Instance on the new Windows VM server.
Copy all of the appropriate profile parameters that were on the old server.
Start SAP and make sure everything is running and working correctly.
Copy all job, spool files, etc. from the old server.
What needs to be done in between those steps for the server rename & IP changes?
Which of these steps can I do in advance of my maintenance downtime window, without it affecting the currently running system?
I have a test environment that I am going to test this with. I'm just trying to get a jump start on my instructions, so that I can hopefully do this quickly and easily. I'm very comfortable doing the installs. Just needing some guidance on how to handle this case.
I'm open to suggestions, and any links you can send my way that show some more recent success at doing this. Everything I keep finding talks about doing the migration for the DB. I am not migrating my DB. It is staying where it is. I'm just wanting to migrate my Message/Central Instance to a VM, without affecting the users, and hopefully them never noticing the changes.
Thanks!

If you're using VMWare, there is a tool provided to virtualize existing systems. Basically it makes a copy of the old system (while it is down) and recreates it as a VM with the same name, IP, etc. I haven't double-checked to see if this process is 'officially' supported by SAP, and I don't know the technical details about how it's done or how long it will take (VMWare is managed by the Network Operations group in my team, so as the SAP sysadmin I'm basically a customer of theirs). However, we have virtualized a few smaller systems that needed to get off failing hardware in a hurry and it worked very well, so I would expect it would for an SAP dialog instance as well. As you're not copying a database, just the app server, it probably would fit within your maintenance window.
If this route works for you, then there is no need to actually install the SAP system on the VM instance... you just copy the old server as-is, then make any adjustments required due to no longer needing old drivers, etc, plus potentially having different memory parameters.

Best practices for handling large messages in JCAPS 5.1.3?

Hi all,
We have ran into problems while processing larges messages in JCAPS 5.1.3. Or, they are not that large really. Only 10-20 MB.
Our setup looks like this:
We retrieve flat file messages with from an FTP server. They are put onto a JMS queue and are then converted to and from different XML formats in several steps using a couple of jcds with JMS queues between them.
It seems that we can handle one message at a time but as soon as we get two of these messages simultaneously the logicalhost freezes and crashes in one of the conversion steps without any error message reported in the logicalhost log. We can't relate the crashes to a specific jcd and it seems that the memory consumption increases A LOT for the logicalhost-process while handling the messages. After restart of the server the message that are in the queues are usually converted ok. Sometimes we have however seen that some message seems to disappear. Scary stuff!
I have heard of two possible solutions to handle large messages in JCAPS so far; Splitting them into smaller chunks or streaming them. These solutions are however not an option in our setup.
We have manipulated the JVM memory settings without any improvements and we have discussed the issue with Sun's support but they have not been able to help us yet.
My questions:
* Any ideas how to handle large messages most efficiently?
* Any ideas why the crashes occur without error messages in the logs or nothing?
* Any ideas why messages sometimes disappear?
* Any other suggestions?
Thanks
/Alex

* Any ideas how to handle large messages most efficiently? --
Strictly If you want to send entire file content in JMS message then i don't have answer for this question.
Generally we use following process
After reading the file from FTP location, we just archive in local directory and send a JMS message to queue
which contains file name and file location. Most of places we never send file content in JMS message.
* Any ideas why the crashes occur without error messages in the logs or nothing?
Whenever JMSIQ manager memory size is more lgocialhosts stop processing. I will not say it is down. They
stop processing or processing might take lot of time
* Any ideas why messages sometimes disappear?
Unless persistent is enabled i believe there are high chances of loosing a message when logicalhosts
goes down. This is not the case always but we have faced similar issue when IQ manager was flooded with lot
of messages.
* Any other suggestions
If file size is more then better to stream the file to local directory from FTP location and send only the file
location in JMS message.
Hope it would help.

Best practice for lots of DP's over slow links

Hi All,
We have an established environment consisting of 1 primary server (with separate SQL server), 10 secondary sites and (potentially) 2000+ sites, each with a DP.
Currently we have all the DP's in place as push DP's with rate limits set between 1% (daytime) and 75% (night time) to throttle the traffic. All but 100 of the sites are on variable links that vary between 2Mb and 16Mb depending on the traffic
needed. The 100 other sites are on very slow links (<1mb).
Network utilisation hardly ever exceeds 50% of the available bandwidth in normal use but can burst to 100% on rare occasions
We are concerned about the load on the primary server when deploying applications to the 2000+ sites, it takes a long time to get the application to all the DP's.
I have been given the task of re-designing the environment to meet the requirements:
SCCM traffic cannot block or affect normal network traffic (financial info etc)
A critical patch to the OS must be deployed as fast as possible to ALL DP's and then to all clients
A large deployment (>1GB) must not block the deployment of smaller but higher priority packages
My reasoning is that if I were to set the 2000+ site DP's to be pull DP's with the 10 secondary sites as content locations (prioritised to the secondary site each DP reports to, then any others) then that would meet points 2 and 3.
Point 1 is the sticky one... Normally I could use rate limits (as now) or pulse mode to limit the bandwidth used between the secondary site and the DP but a pull DP cannot use anything other than BITS.
What would be the settings for setting up BITS?
Where would be the better area? client settings (requires client on the DP) or via GPO?
Any suggestions as to preferred limits to set in BITS?

If your DPs belong to secondary sites, then there is no load on the primary site server to distribute content. Content flows from the site server that the DP belongs to which in this case are secondary site servers.
You will never prevent complete network usage using built-in software facilities. If those are your requirements you should discuss QoS with the network team and/or check out one of the two alternate content providers (and eliminate all or most of your DPs):
Adaptiva OneSite and 1E Nomad.
Jason | http://blog.configmgrftw.com | @jasonsandys

Best practice to send report to disabled users

Few users are not on CRM yet. They are setup as deactivated users and user lookup on the Account is a look-up to their user record. They would like a to get an automated weekly email/report
showing the activity (appointments, email, etc.) scheduled for the upcoming week for each of their accounts (where they are the user on the Account)
Here the option we thought of. Can you think of others that are relatively simple to develop?
Give them access to CRM and develop a Dashboard
Dynamic worksheet they can save to their desktop
Scheduled report?   can I setup a report schedule to be sent to the user?

Hi,
    The simplest option is option 1. If they need to access the account in CRM, they would need access to the system. It would be easier and it will not cause any other major customisations etc.
    As the users are disabled users, other 2 options are not valid options.
Hope this helps.
Minal Dahiya
blog : http://minaldahiya.blogspot.com.au/
If this post answers your question, please click "Mark As Answer" on the post and "Vote as Helpful"

Best practice for sending client ai files

If a client requests ai files...and they don't have Illustrator - but they want the orig files for printers and vendors, should I save the file using legacy or just save as a normal cs4 ai file?
Thx.
Chemol

in this case it's just a simple 2-colour logo (same file you helped me with earlier) - no effects, nothing to be changed. Nothing mentioned in the contract, I just want them to have the original file so they can provide it to print providers if need be in the future for whatever projects they use it for.
Just a simple situation, wanting to provide most versatile format for them in the long run. I have CS4, and have been told by one past client (for a signage project) that the printer was unable to open the file - I think they had an older version of AI....
Just curious - any insight would be helpful.
Thx.

Best Practices for sending meeting link in advance?

Greetings,
So our client is asking for the meeting login link well in advance of the meeting (to distribute to his staff), but of course we're still testing and setting pods up in the meeting room.
Can we provide the link/invite, but prohibit their ability to login in advance and putter around, seeing things that aren't ready for prime time?
Oh, and please forgive asking another question that should probably be a second post, but has anyone come up with a default graphic/pdf/ppt slide to be the first thing people see as they enter? Seems like it would be useful to provide some generic, graphical instructions to ease the uninitiated into the Connect world, pointing out what they may need to set up to particpate fully...
thanks,
Steve

Sure. When you create the meeting there is a choice you have to make in terms of meeting security. Just choose the option that everyone must login.
You then make sure that you do not start the meeting. Or if you start the meeting so you can load assets/pods, then choose the option on the meeting access menu to place participants on hold. That way the room is on hold until you are ready to start.
You can download any JPG image as your room background.
There is not much the attendee needs to setup with Adobe Connect.
Add a note to the notes pod to advise them to check their computer speakers. One way I do this is to upload an MP3 file with songs in it and have that playing. It's a nice way to join a meeting and it tells the host that people are prepared for sound/video.

Best Practice for closing a blocking connection to the database.

Background:
Using toplink/eclipselink to manage the the connection to the database.
We support 3 databases, Oracle, Mysql, and MSSQL.
We host the an application locally (SAAS) and we also distribute it to our clients so they can run it on premise.
Problem:
From time to time the application will execute a route that query the DB and will take 10+ minutes. When we find these queries we generally fix them, but it takes time to fix the problem and roll new code to production. We have also found that if an end user executes the query to many times they can potentially bring the service down - maxing out the database connection, or putting a huge load on the DB effectively rendering the system to a crawl.
During this time period the IT team need a way to abort a transaction. The current approach, for our SAAS solution, to scan the database find to the long running query and kill it. The application is smart enough to recover and everything seems to work fine. This doesn't always work for our on premise customers - often they don't have a DBA or adequate resources to monitor the system. Therefore, I'm looking for a way to have the application kill/close any query that is lasting longer then X minutes. We currently track the duration of every thread so I can get the time the thread has been actively running.
Stack Trace
Currently we keep track of how long each thread is running waiting for a response from the database such as the stacktrace below, I would like to grab the connection and physically close it. Since we are using a java socket and it's in a blocking state there is no way for us to interrupt the thread. The only option is to wait for it to receive a response from the db - kill the sql process - or to close the java socket.
java.net.SocketInputStream.socketRead0(Native Method)
java.net.SocketInputStream.read(SocketInputStream.java:129)
com.mysql.jdbc.util.ReadAheadInputStream.fill(ReadAheadInputStream.java:113)
com.mysql.jdbc.util.ReadAheadInputStream.readFromUnderlyingStreamIfNecessary(ReadAheadInputStream.java:160)
com.mysql.jdbc.util.ReadAheadInputStream.read(ReadAheadInputStream.java:188)
com.mysql.jdbc.MysqlIO.readFully(MysqlIO.java:1931)
com.mysql.jdbc.MysqlIO.reuseAndReadPacket(MysqlIO.java:2380)
com.mysql.jdbc.MysqlIO.checkErrorPacket(MysqlIO.java:2909)
com.mysql.jdbc.MysqlIO.sendCommand(MysqlIO.java:1600)
com.mysql.jdbc.MysqlIO.sqlQueryDirect(MysqlIO.java:1695)
com.mysql.jdbc.Connection.execSQL(Connection.java:2998)
com.mysql.jdbc.Connection.execSQL(Connection.java:2927)
com.mysql.jdbc.Statement.executeQuery(Statement.java:956)
org.eclipse.persistence.internal.databaseaccess.DatabaseAccessor.executeSelect(DatabaseAccessor.java:854)
org.eclipse.persistence.internal.databaseaccess.DatabaseAccessor.basicExecuteCall(DatabaseAccessor.java:573)
org.eclipse.persistence.internal.databaseaccess.DatabaseAccessor.executeCall(DatabaseAccessor.java:501)
org.eclipse.persistence.sessions.server.ServerSession.executeCall(ServerSession.java:536)
org.eclipse.persistence.internal.queries.DatasourceCallQueryMechanism.executeCall(DatasourceCallQueryMechanism.java:205)
org.eclipse.persistence.internal.queries.DatasourceCallQueryMechanism.executeCall(DatasourceCallQueryMechanism.java:191)
org.eclipse.persistence.internal.queries.DatasourceCallQueryMechanism.executeSelectCall(DatasourceCallQueryMechanism.java:262)
org.eclipse.persistence.internal.queries.DatasourceCallQueryMechanism.executeSelect(DatasourceCallQueryMechanism.java:244)
org.eclipse.persistence.queries.DataReadQuery.executeNonCursor(DataReadQuery.java:188)
org.eclipse.persistence.queries.DataReadQuery.executeDatabaseQuery(DataReadQuery.java:144)
org.eclipse.persistence.queries.DatabaseQuery.execute(DatabaseQuery.java:664)
org.eclipse.persistence.queries.DataReadQuery.execute(DataReadQuery.java:130)
org.eclipse.persistence.internal.sessions.AbstractSession.internalExecuteQuery(AbstractSession.java:2243)
org.eclipse.persistence.internal.sessions.AbstractSession.executeQuery(AbstractSession.java:1181)
org.eclipse.persistence.internal.sessions.AbstractSession.executeQuery(AbstractSession.java:1165)
org.eclipse.persistence.internal.sessions.AbstractSession.executeQuery(AbstractSession.java:1125)
Question
What is the best way to grab the connection and close it - from another thread?

The reason for changing the namespace is to guarantee that they appear as different versions. The namespace is part of the strong name of the DLL when deployed. Yes they will be deployed to separate servers because a server can only belong
to one farm. But when you are doing a migration it won't be clear that you need to install the 2013 version since the .DLL will report that it exists in the 2010 farm already. It will work without changing the namespace, but it would be best to differentiate
versions at that level to avoid confusion.
Paul Stork SharePoint Server MVP
Principal Architect: Blue Chip Consulting Group
Blog: http://dontpapanic.com/blog
Twitter: Follow @pstork
Please remember to mark your question as "answered" if this solves your problem.

Best practice for storing price of an item in database ?

In the UK we call sales tax, VAT, which is currently 17.5%
I store the ex-VAT price in the database
I store the current VAT rate for the UK as an application variable (VAT rate is set to change here in the UK in January)
Whenever the website display the price of an item (which includes VAT), it takes the ex-VAT price and adds the VAT dynamically.
I have a section in the website called 'Personal Shopper' which will happily search for goods in a fixed priced range eg. one link is under £20, another is £20-£50
This means my search query has to perform the VAT calculation for each item. Is this practice normal, or is it better to have a database column that stores the price including VAT ?

I'm also based in the UK, and this is what we do:
In our Products table, we store a Product Price excluding VAT and a VAT rate ID, which is joined off to a VAT Rates table. Therefore in order to calculate selling price yes, this is done at the SQL level when querying back data. To store the net, vat and gross amounts would be to effectively duplicate data, hence is evil. It also means that come January we only have to update that one row in one table, and the whole site is fixed.
However.
When someone places an order, we store the product id, net amount, vat code id, vat amount and vat percentage. That way there's never any issue with changing VAT codes in your VAT codes table, as that'll only affect live prices being shown on your website. For ever more whenever pulling back old order data you have the net amount, vat amount and vat percentage all hard-coded in your orders line to avoid any confusion.
I've even seen TAS Books get confused after a VAT change where in some places on an order it recalculates from live data and in others displays stored data, and there have been discrepancies.
I've seen many people have issues with tax changes before, and as database space is so cheap I'd always just store it against an order as a point-in-time snapshot.
O.

Best practice from passing messages from servlets

Is there a best practice for passing user messages (typically errors) back to the page from servlets?
e.g. http://localhost:4502/content/geometrixx/en.html?message=Some user error message
Dan

Well I suppose that answer to that question depends somewhat on your requirements, but I would say using a query string as you have indicated would be less than ideal because the page with the message would not be cached. No depending on your requirements and what sort of message you are passing that might be OK - especially if you message is highly personalized.
If however you have a limited number of standard messages to display a more common approach is to have each message have it's own page and then to configure the servlet to redirect to the appropriate page based on the desired message.
If you want the user to end up on the same page they submitted to the servlet from then another common approach would be to have the post to the servlet be AJAX and then display the message client side without having to change the URL.

Best practice for OSB to OSB communication

Cross posting this message:
I am currently in a project where we have two OSB that have to communicate. The OSBs are located in different security zones ("internal" and "secure"). All communication on a network level must be initiated from the secure zone to the internal zone. The message flow should be initated from the internal zone to the secure zone. Ideally we should establish a tcp connection from the secure zone to the internal zone, and then use SOAP over HTTP on this pre-established connection. Is this at all possible?
Our best approach now, is to establish a jms-queue in the internal zone and let both OSBs connect to this. All communication between the zone is then done over JMS. This is an approach that would work, but is it the only solution?
Can the t3/t3s protocol be used to achieve our goal? I.e. To have synchronous commincation over a pre-established connection (that is established the in opposite direction of the communication)?
Is there any other approach that might work?
What is considered best practice for sending messages from a OSB to another OSB in a more secure zone?
Edited by: hilmersen on 11.jun.2009 00:59

Hi,
In my experience in a live project, we have used secured communication (https) between internal service bus and DMZ/external service bus.
We also used two way SSL with customers.
The ports were also secured by firewall in between them.
If you wish more details, please email [email protected]
Ganapathi.V.Subramanian[VG]
Sydney, Australia
Edited by: Ganapathi.V.Subramanian[VG] on Aug 28, 2009 10:50 AM

Best Practice to send AQ message to over 30 databases

Similar Messages

Maybe you are looking for