Event monitoring / notification - AQ, Alerts, or what ?

I need to know whether Oracle Advanced Queuing, or some other Oracle database facility is able to support the event notification requirements for our event driven user interface.
I am working on moving an application to Oracle (Oracle database is my main expertise but I haven't ever used AQ). There are varying numbers of GUI clients logged in at a time, plus some background services. The application design depends on updating entities in the database and having an event notification sent to all interested GUI clients. I need to support this in the oracle environment.
Key issues are:
- Database triggers to initiate notification of updates
- Different event topics for each notifiable table. Many rows in each table.
- Every event notification must arrive. DBMS_ALERT only delivers the most recent alert.
- Clients come and go fairly often, and the client set is not fixed. Clients are only interested in events that occur while the client is connected.
- We don't want to be polling a table - we want to either wait for events or be notified of them.
Is there a way of doing this using Oracle Advanced Queuing multiconsumer queues?
Is there a better way of doing this with some other tool that comes with the 10G database?
The descriptions of AQ I have read are oriented to having a limited, fixed set of message consumers, whereas our environment has a varying set of subscribers, so I am concerned about:
- How heavyweight is the subscribe / unsubscribe step ? I have not seen references to it being done on the fly like this.
- Is there a way of making the subscription be dropped when the database connection is dropped, or are they always persistent ?
DBMS_ALERT won't do because if many alerts are raised while a listener is busy, only the latest alert is delivered.
We are running the 10GR2 database but not the application server.
Thanks for your advice,
David Penington

David,
I have the same requirement as you.
Basically what I have put in place is:
- Package with sets of wrapper method to ease queue related process (enqueue, dequeue, listen, ...)
- Related object type, queue table and different queues based on different events (update, insert, delete, ...)
- Triggers in certain tables that are related to events described above (will only enqueue messages)
- Application that subscribe to different queues and then calling a blocking listen in a separate thread (the listen call will be unblocked with the queue name as the out parameter)
- I activated the Queue Monitor to clean up the consumed and expired messages
- Other information: Oracle version 9i; non-persistent queue was not yet tested - OCI + raw :(
So anyway, everything is ready and when I do an update on a given record I will have all the applications notified. I launch several application instances and everything is working as I expected.
But then I have another problem: it generates too many redo logs (which in turn become archive logs). I observed this when doing a performance test; it was something like updating 20 records for 10000 times with delay between interval set to 0,5 second. That was when I noticed the excessive generation - which is finally normal because everything is done with transaction integrity.
Until now I haven't find any solution yet... I was thinking about NO LOGGING in the tablespace specially created for the queue table, but it was not at all the solution - well I tried my luck and it didn't work :)
Is there anyone out there that has the same requirement as me and David and has the solution for this? Maybe another way of implementing the mechanism...
Looking forward to have some feedbacks.
Regards,
Kiky Shannon

Similar Messages

How do I see my all-day events in Notification Center?

I cannot see my all-day calendar events in Notification Center. How do I change this so that they show without me having to go to the Calendar app to see them?

What IOS are you using?
With IOS7:
Settings > Notification Center > Calendar Dy View: ON
then scroll down to INCLUDE section
      Tap Calendar
          ALERT STYLE: Alert
          Calendar Alert > choose a sound here
          Show in Notification Center: ON
          Include > 5 items
          Show in Lock Screen: ON

Event monitor cannot find PowerShell generated events

I have a basic event monitor that only looks for an event ID and source. The event I am monitoring is generated via Powershell from a remote system rather than being locally created. I'm not sure if that matters, but it sure seems like it.
If I change the event expression in the monitor to look for other events it will generate alerts, it just cant see my PS events. I've tried changing the ID number, the source, using different params, etc... I'm convinced that this is due to the event creation
process and PS must be creating the event in a different manner than SCOM is looking for. Maybe its a bug where SCOM just ignores events created by proxy.
I found this: http://www.systemcentercentral.com/monitoring-events-written-from-a-remote-server-with-scom/
But that is for Rules. I still tried adding <AllowProxying>false</AllowProxying> to the XML of the MP...SCOM didn't like
it.
Has anyone ran into this before?
We are using SCOM 2012R2 fully updated.
- Slow is smooth and smooth is fast.

I guess I should have been more specific. I am using PowerShell to create an event, it isnt an actual PowerShell event.
write-eventlog -computername Server -logname Application -source Orchestrator -eventID 1 -message "This is an example of the events created with PS that SCOM cannot detect."
Maybe I should explain what I'm doing. I have an Orchestrator runbook that does stuff with things. If one of the runbook activities fails, it will send an event to the machine it was trying to run the activity on. SCOM is monitoring for that PowerShell generated
event stating that the runbook failed.
- Slow is smooth and smooth is fast.
I think you haven't native source "Orchestrator" in PoSh log.
Vladimir Zelenov | http://systemcenter4all.wordpress.com
I run a New-EventLog command first to create the Orchestrator source. Otherwise the Write-EventLog command would fail if the source wasn't there. I can see the event there in the logs.
- Slow is smooth and smooth is fast.

Monitoring template and alert noifications, how ?

I have set up database monitoring template with some metric tresholds for warning/alerting. Also, I applied this to the databases. Now, my question is how will i be notified (say via e-mail) once an alert has generated ? I see notification methods and notification rules, where it has its own metric settings. Do i have to apply notification rules to these databases for notifications. If so, what is the use of monitoring template ? Can i set up notifications via e-mail within the monitoring template ?
thanks,

In EM12c, go to the 'Enterprise' menu, then 'Job', then 'Library'. There should be an out-of-the-box job called "DISABLE TABLESPACE USED (%) ALERTS FOR UNDO AND TEMP TABLESPACES" that you can run against your database targets, and that job will disable database-generated alerts.
If a previous admin has set up EM12c-generated alerts for undo and temp, then yes, a monitoring template will take care of disabling them. Either remove those alerts from the existing monitoring template and apply them to your targets, or create a new monitoring template based on one database's current settings, remove those warning/critical thresholds for your undo and temp tablespaces, and apply it to your targets.

Double Calendar events and notifications

Hello everyone,
I am getting double Calendar events and notifications, that is, instead of getting and seeing one notification/calendar event, I get two of the same one.
Yesterday I was setting up iCloud so it may have something to do with that but I don't know what.
Also, in my Calendar Settings, I have the same calendars listed ( Work, Home, Personal, etc) under "From my Mac" and the another under iCloud.
This might have something to do with it too ...
But I still don't know what I need to do in order to stop getting double notifications and alerts.
Any ideas, suggestions??
thanks!!

okey-dokey, and now imagine I have 250 events in calendar
sitting one hour somewhere and changing some of them in such a manner - will make anyone mad... unless he has non-nokia device... mua-ha-ha
Of course it's related to the bug you described (I've encountered it too).
But regarding my initial problem as in header of topic - I'm pretty sure it does exist.
I deleted ALL entries in my Nokia calendar and have been ultimately suprized to hear reminder popping up. I even searched through calendar and alarms to see if anything left by chance. No - both were empty. Presume this bug is somehow related to the second one you described.Message Edited by sidream on 16-Oct-200704:14 PM

Is there a way to exclude specific calendars from Notifications and alerts?

I have 4 calendars: 1 Exchange and 1 Yahoo that includes 2 shared calendars (wife and MIL). I activate the Yahoo calendars only when I need them and do not need any yahoo reminders at all. When I swipe down the Notifications page, it includes all four calendars including the shared Yahoo calendars. What's worse, I get alerts from all four calendars.
I only want Exchange alerts and notifications. Is there a way to exclude the Yahoo calendars from Notifications and alerts?
Right now, the only way I can eliminate them is by deactivating the Yahoo account in Settings/Mail, Contacts, Calendars.

Is there a way to only choose specific contacts from an existing group and add to a new group?
I mean there has be a faster way than typing everybody by hand..

IPS event monitor and reports not working

Dear after upgrading my IPS from E3 to E4  the event monitor and reports not working, can you please advice my to solve this issues

Hi All,
Filter settings below:
The filter works partially as I don't get alerts on the IPS itself.
Firewall LOG:
4          Feb 14 2014          15:33:22                              39715                    514          IPS requested to drop UDP packet from SOURCE_VLAN_NUMBER:/39715 to DESTINATION_VLAN_NUMBER:/514
IPS LOG (when enabled):
evIdsAlert: eventId=1352793300955167909 vendor=Cisco severity=low
originator:
    hostId: SSM02
    appName: sensorApp
    appInstanceId: 1192
time: Feb 14, 2014 15:33:22 UTC offset=0 timeZone=GMT00:00
signature:   description=IP Fragment Too Small id=1206 version=S212 type=anomaly created=20030801
    subsigId: 0
    sigDetails: Too many small IP fragments in datagram
interfaceGroup: vs0
vlan: 0
participants:
    attacker:
      addr: 172.x.x.x locality=OUT
      port: 39715
    target:
      addr: x.x.x.x locality=OUT
      port: 514
      os:   idSource=unknown type=unknown relevance=relevant
alertDetails: InterfaceAttributes: context="single_vf" physical="Unknown" backplane="GigabitEthernet0/1" ;
riskRatingValue: 50 targetValueRating=medium attackRelevanceRating=relevant
threatRatingValue: 50
interface: GigabitEthernet0/1 context=single_vf physical=Unknown backplane=GigabitEthernet0/1
protocol: udp
Our next step is to make a service policy exception on the firewall itself. We are also considering reloading the IPS device or at least the analysis engine.
Thanks for all your help so far. Any more suggestions are most welcome. I'll keep you up to date.
Regards
Mariusz

Missed event monitor with event reset

I have pretty simple request for Scom monitoring, but I don’t know how to do it. My application generates an event in custom log with name Mpis. Event level is Informational, Event ID is 10 and source is Mpis. This event is created every
5 minutes. If application failed, event isn’t created. I have a need to create an alert, if event isn’t created more than one hour and monitor must go into unhealthy state (red). If same event come back, monitor must resolve itself and go back to healthy (green).
I try doing this with missed event monitor with event reset, but they don’t work as I expected, probably because I use the same event for trigger an alert and resolve an alert. Any idea ?

There must be something misconfigured. Missed Event Detection -> Windows Event Reset is the correct one.
Use Event ID 10 and Source Mpis as expression in both Simple Event Expression and Missing Event Expression. in Missing Event Detection, use "Based on fixed simple recurring Schedule" and choose 60 minutes/1 hour.
I have configured this in my environment, and this works perfectly; if no events are logged within an hour, i get an alert. If i then create the event, the alert Closes.
www.coretech.dk - blog.coretech.dk

Monitoring Notification Mailer

Guys,
We have an issue with notification mailer needing to be restarted on a regular basis. Is there a common method of monitoring notification mailer. Can a cron be setup to send an alert when it goes down.
Thanks in advance,

Hello
First off, please check the WF mailer log file on the concurrent tier under $APPLCSF/$APPLLOG. The file is named FNDCP*txt - check the latest ones. You should be able to see errors there pointing out to the issue that is killing the mailer.
Now, to answer the question directly, you can check fnd_svc_components. There you will see the status of the WF mailer and if it appears not running then you can create your alert based on it. A query like this might do:
select component_name, component_status
from fnd_svc_components
where component_type='WF_MAILER';
Regards,
Alejandro

NSM 3.0.4 Event Monitor Service

NSM AD 3.0.4 on W2008
This install has been running fine for months. Now the Event Monitor service will not start. Does anyone have any ideas on what might be the cause? Suggestions on a fix?
Mike

On 2/14/2013 6:10 PM, mfruech wrote:
>
> NSM AD 3.0.4 on W2008
>
> This install has been running fine for months. Now the Event Monitor
> service will not start. Does anyone have any ideas on what might be the
> cause? Suggestions on a fix?
>
> Mike
>
>
Mike,
Does the Event Monitor log file show anything? You should be able to see
where the Event Monitor starts by looking at the log's timestamps
(there's a large blob of XML at that timestamp too), and what follows is
a sequence of checks making sure the service can start up correctly with
all its appropriate dependencies. Anything with a 'Result = ' or 'rc='
followed by a non-zero number is potentially a problem. Let us know what
those are, and we can troubleshoot the Event Monitor from there. Thanks!
- NFMS Support Team

Can't see events in Event Monitoring in IME

Hello,
I have a AIM-SSM installed succesfully, and "show events alert" output does display some alerts as expected.
The problem is, there is no events in Event Monitoring in IME. Have tried to select "Realtime" or Time Range, it doesn't work.
I've checked the configuration, and everything looks fine.
Any suggestions for this?
Thanks and regards,
-Alejin

Better to post your topic in the iPhoto community here.
This forum is more or less for troubleshooting the App Store.
You'll receive more feedback from iPhoto users in that forum.

Notification from Alert

Hello Friends,
The business needs notification in the worklist along with email from an event based alert or periodic alert. Is it possible to do this from an alert.
If not, is there any method to set up an event based or periodic alert in the Work flow. Can any one tell how to use the workflow for making alert.
Many many thanks in advance.

Hi,
We had a similar requirement to send a notification from the alert.
We created the alert and called the conccurent program which will trigger the workflow.
Below are the stepsfor the same:
Create the event/periodic alert and give the action type as "Concurrent Program".and give the concurrent program detials as below
Application :Human Resources
Program Name :XX(Eg:Alert Workflow Initiation CP) (this is a plsql stored procedure type with six parameters for this program )
Arguments:YYY(eg:"N" &person_id "&name" "&job" "&org" "&username")
Below is the sample code to call the workflow from plsql stored procedure
begin
l_item_key := to_char(SYSDATE,'YYMMDDHH24MISS');
wf_engine.CreateProcess
(itemtype => 'IRC_WF',
itemkey => l_item_key,
process => 'PER_APPLICATION_PRC'
l_mail_content := chr(10)||chr(10)||'Dear '||p_name||':'||chr(10)||chr(10)||
'We appreciate your interest in the '||l_job||' position. A member of the Human Resources team will review your information and contact you if your skills and experience is a match for our needs at this time.'||chr(10)||chr(10)||
'Once again, thank you for your interest in employment at XXX.'||chr(10)||chr(10)||
'Sincerely,'||chr(10)||
'XXX';
wf_engine.SetItemAttrText(itemtype => 'IRC_WF',
itemkey => l_item_key,
aname => 'WWPER_MESG_HTML',
avalue => l_mail_content);
wf_engine.StartProcess
(itemtype => 'IRC_WF',
itemkey => l_item_key);
COMMIT;
end;
Get back to me if you need further details
Thanks,
Anuradha
Edited by: user10411683 on Dec 2, 2008 11:12 PM

Customers should receive event triggered notification.

Hi,
I am suppose to do the e-mail design documentation using sap crm on 1. Customer Receives Event Triggered Notification. and 2. Customer receives the marketing information. on these two topics got no idea on these things at all. Could any one please help me in finding the material for these and also please describe the overview of these two things.
Thanks and Regards,
Venkat.
Message was edited by:
venkat ratnam

Hi,
I cant say about the documentation but what I know is we have to perform the Scott configuration in the system,Maintain the distriution list in custom table for email detials of the customer & develop an exit which will send the required information on event of saving the document..Hope this will help..
Thanks

Best Approach for Performance/Event Monitoring

I'm completely new to SAP and would appreciate the input of some more experienced folks.
I have been given two tasks for SAP: performance monitoring of the SAP application/associated modules (e.g. SAP Memory utilization, number of processes, etc.) and alarm event monitoring (anything that would be characterized as a error and logged).
For the performance monitoring I would like a access the metrics once every 5 minutes to keep a running history. Several things I've started to look at are the Alert Manager and perhaps a Perl API interface.
For alarm events, I assume there is some SAP module that can monitor for these and generate SNMP traps or perhaps a custom application that does log mining. As part of this I am trying to determine where error events are written to in SAP (e.g. a log file??).
If anyone has insight or can point me in the right direction, I would be very appreciative. Thanks in advance.
Ed

Dear Ed,
If Wily Introscope is configured for your Portal, which shoold be the case because SAP delivers it as with SAP Solution Manager, all metrics which oyu mention would be available.
If you are running a testing instance which is not integrated with Solution Manager, you could try the Web Administrator
http://<host>:<port>/nwa -> Availability and Performance -> Resource Monitoring
Best Regards,
Sylvia

Insert date on event module notification email

Hello community,
I'd like to add the date of the event booked on the event module notification email, but those tags are not available in the toolbox on that panel and when I use those listed in the documentation, they don't work and simply display as {tag_ }, even though they work in the detailed event layout.
I'm thinking it can be done with a module tag /liquid layout, but the things I've tried so far aren't giving me what I need.
Can you help?
The events are daily (basically admission) so there is no category, but the date should be sent on the confirmation email because it serves as the ticket.
Thank you!

PLEASE! There must be some way to insert the date on the events module notification email.
HELP????

Event monitoring / notification - AQ, Alerts, or what ?

Similar Messages

Maybe you are looking for