CPU load alert on Enterprise Manager

Hi,
We have Oracle Database R1 (11.1.0.7.0). We see alerts on Enterprise Manager; "CPU Load is 2164260864, crossed warning (10) or critical (20) threshold." frequently.
Our CPU utilization looks normal; CPU in I/O Wait (%) = 0 but Run Queue Length (5-minute average)= 21,474,836.47 is critical. Currently we do not have any performance problems and we do not take a heavy load.
How can I check if this is the real load on our CPU and how can I learn what really causes it?
Thanks,
Hatice

>How can I check if this is the real load on our CPU and how can I learn what really causes it?
might solution/answer be Operating System dependent?
How do I ask a question on the forums?
https://forums.oracle.com/message/9362002#9362002

Similar Messages

Wish to clear tablespace alert at Enterprise Manager 10g

Is that a way to clear tablespace full alert at Enterprise Manager 10g?
I have added the required space to the database but the alert is still there.
Below is the screenshot:
http://i.imgur.com/glve4.png
There is new and valid tablespace alert coming, and I am sure the space adviser is kept running. But I have no idea why the alert has not been cleared.
Note: I do not wish to change or disable the metric. If another tablespace nearly full, I want to be alerted.
Edited by: csmth96 on Jan 18, 2012 3:49 PM

This is the followup on the alert.
The alert is generated based on metrics behind "dba_tablespace_usage_metrics". Without fixing the metric collection process, removal of an alert only caused the alert to be shown some time later. The removal of alert looks working initially but it does not fixed anything.
My database is created at 10.2.0.2 but upgraded to 10.2.0.4 recently. Because the alert is created before the upgrade, it is reasonable to believe that the metric collection problem has started before the upgrade (rather than the upgrade has caused the problem). Having reviewed metalink DocID 736909.1, it is reasonabMyle to believe the view "dba_tablespace_usage_metrics" is the problem source. It summed up the max size of all files in tablespace as the tablespace size. However, before 10.2.0.4, extending a datafile does not increase the max file size (view "v$filespace_usage", the limit of autoextend). If the actual usage is large than the autoextend limit, the usage is more than 100%.
My action is to increase the autoextend limit (max file size) of datafile to the same as allocated file size. No autoextend is actually allowed. The autoextend limit is reset so that "dba_tablespace_usage_metrics" is corrected and it is no longer possible to have more than 100% tablespace usage. The alert is gone several minutes later.
I doubt whether it make sense to compare the actual usage vs autoextend limit. For any concerned DBA, patch 6759910 may be applied at a downtime window.
I conclude the solution provided by "N Gasparotto" is truly correct.
Edited by: csmth96 on Jan 26, 2012 3:26 PM

Alert from Enterprise Manager

Hello, I have an oracle rac with 3 nodes and I have oracle database control enterprise manager (No Grid Control).
Yesterday one node was reboot and I don't know why. I want set a rules from enterprise manager or something similar for notificate to me when a reboot from one of my host suceed.
Any one can help me?

roberto1979es wrote:
Ok,
Can I program one alert for warning when a node is reboot it?When a node is reboot it. !http://www.wackyb.co.nz/skype/smileys_v1.5.0.60/skype-wackyb-0120-doh.gif!
If i understood correctly what you are trying to ask , you are interested in email notification when your node suddenly reboots/restarts.
And if it is so, this is will not be possible in the way like it is for tablespace, memory and similiar alerts.
Although you could consider monitoring alert.log content, because information about database up-down statuses can be found there.
For example , you could do that either by using external tables or by using shell/batch scripts;depending on the os on what your database is, and then after information that indicates database unexpected reboot is found, forward that information further (e-mail notification) .

Alert on enterprise manager

how to get an alert in the enterprise manager when a specific user login??

For this you have to enable auditing(audit by session/by access). see following links
http://www.oracle-base.com/articles/8i/Auditing.php
Auditing in Oracle 10g

False alerts from EnterPrise Manager

Hi All,
Please help in addressing this issue:
I have configured monitoring and alerts for OVD in Enterprise manager and I have been getting false alerts on OVD being down.
Below is my notification alert configuration:
General:
Rule Name: OVD Down
Public: No
Availability:
Up: No
Down: Yes
Corrective Actions on target down: No
Agent Unreachalbe: Yes
Agent Unreachable Resolved: No
Metric Error Detected: Yes
Metric Error Resolved: No
Blackout Started: Yes
Blackout Ended: No
No Metrics I have added, No Policies, No job types and no methods.
My first thought that I have "Metric Error Detected" but I haven't added any metrics. So, I change it No (without restarting EM) but same false results.
Not sure if I have to restart the EM after changing notificaiton rule.
Please help.
Regards,
Sunny

Uninstall LELA and your false alerts will go away.

Apply CPU /PSU patches through Enterprise Manager Grid Control.

How to apply PSU patches from 10.2.0.5 GRID Grid Control to my test/development/qa environment.
Are there any step by step documents availible. Did any one applied these patches using Oracle Enterprise Manager Grid control.

Please read
Oracle® Enterprise Manager Administrator's Guide for Software and Server Provisioning and Patching
10g Release 5 (10.2.0.5.0)
3 Infrastructure Requirements
http://download.oracle.com/docs/cd/B16240_01/doc/em.102/e14500/infrastructure_setup.htm#BABEIFBC
For the correct setup to do you patching.
Regards
Rob
http://oemgc.wordpress.com

Alerts on Enterprise Manager 10.2.0.1.0

hy gurus...
Recently, One of my Tablespaces ran out space and I got one alert at the home page of EM. Howewver I've already fixed it and the alert did not desappear.
Is there storage in a fixed Table ? How can I clean these alerts ?
tks
Keen

Ok. You need to delete from the MGMT_SEVERITY table - use the timestamp and a method of referencing the correct record. This table has a trigger called SEVERITY_DELETE will will fire to delete records from MGMT_CURRENT_SEVERITY, MGMT_SEVERITY_ANNOTATION, and MGMT_NOTIFICATION_LOG.
HTH,
Jason

Error loading Management Modules of Enterprise Manager 8.2.3.5

Hi all,
actually we have installed the Wily Enterprise Manager 8.2.3.5 on Linux x64.
Starting the manager we get the following errors in the IntroscopeEnterpriseManager.log
2/23/11 03:41:09 PM CET [INFO] [Manager.ManagementModule] Loading Management Module "MSFT .NET - Duet"
2/23/11 03:41:09 PM CET [INFO] [Manager.ManagementModule] Loading Management Module "SAP J2EE - Duet"
2/23/11 03:41:09 PM CET [INFO] [Manager.ManagementModule] Loading Management Module "SAP J2EE - NetWeaver PI(XI)"
2/23/11 03:41:09 PM CET [INFO] [Manager.ManagementModule] Loading Management Module "SAP Navigation"
2/23/11 03:41:09 PM CET [INFO] [Manager.ManagementModule] Loading Management Module "SAP Host"
2/23/11 03:41:09 PM CET [INFO] [Manager.ManagementModule] Loading Management Module "SAP Business Objects"
2/23/11 03:41:10 PM CET [ERROR] [Manager]
2/23/11 03:41:10 PM CET [ERROR] [Manager]
2/23/11 03:41:10 PM CET [ERROR] [Manager]
2/23/11 03:41:13 PM CET [ERROR] [Manager]
The error indicates a problem during the load of the Management Modules.
Also we observed problems calling dashboards from the Console in the wily Introscope Webview.
For e.g. the dashboard J2EE CPU and memory Detail
Calling this dashbord ends in the following error:
An error occurred fetching this dashboard from the Enterprise Manager: null
What could be the problem for the errors during the Management Modul load ?
Thanks in advance for your help.
Kind regards
Anton

Hello Anton,
Please download the SAP Management Modules from the marketplace.
Check note 797147 and note 1579474 about Management Modules usage.
Best regards,
Guilherme Balbinot

Oracle 10g Enterprise Manager console giving Alert log

Dear all
I have One HP server running Oracle 10g on Redhat Enterprise Linux 3.0 and two other systems with Oracle 9i and 8i. I have written sql query in my Oracle10g system which updates the database in 9i.
The sql statement is
CREATE MATERIALIZED VIEW ADASNAP REFRESH FAST START WITH SYSDATE NEXT SYSDATE+1 AS
SELECT * FROM ADA@WCMMISLINK;
CREATE MATERIALIZED VIEW BGTABSNAP REFRESH FAST START WITH SYSDATE NEXT SYSDATE+1 AS
SELECT * FROM BGTAB@WCMMISLIN;
The above query is sheduled to run every day .
and i get the following Alert log in 10g enterprise manager console
Generic Alert log ORA-12012: error on auto execute of job 54
ORA-04052: error occured when looking up remote object WCMM.SYS@WCMMSERVERLINK
ORA-00604: error occured at recursive SQL level 3
ORA-12514: tns no listener
ORA-06512: AT "SYS.DBMS_SNAPSHOT" line 1883
can you guid me why the above error are occuring
the ORA-12514: tns no listener
the above error why it is occuring , then listener is running on oracle 10g server and other 8i and 9i server also and i am able to connect from oracle10g to other oracle 8i and 9i from sql plus and viceversa..
what could be the problem with tns listener
Regards
Niranjan

Hi Ugonic
Sorry both are same, it's spelling mistake it's " WCMMISLINK". Acutally, my database is getting updated but this alert log is getting generated in enterprise manager console of 10g.
The sql statement is
CREATE MATERIALIZED VIEW ADASNAP REFRESH FAST START WITH SYSDATE NEXT SYSDATE+1 AS
SELECT * FROM ADA@WCMMISLINK;
CREATE MATERIALIZED VIEW BGTABSNAP REFRESH FAST START WITH SYSDATE NEXT SYSDATE+1 AS
SELECT * FROM BGTAB@WCMMISLINK;
The above query is sheduled to run every day .
and i get the following Alert log in 10g enterprise manager console
Generic Alert log ORA-12012: error on auto execute of job 54
ORA-04052: error occured when looking up remote object WCMM.SYS@WCMMSERVERLINK
ORA-00604: error occured at recursive SQL level 3
ORA-12514: tns no listener
ORA-06512: AT "SYS.DBMS_SNAPSHOT" line 1883
can you guid me why the above error are occuring
Regards
Niranjan

Enterprise Manager - extremely slow to load initial tree

Group,
Has anyone else encountered any long delays waiting for the CAPS Enterprise Manager to display the initial tree of servers/serivces? We are finding that it takes on average 5 minutes to load the initial treeset in any given environment. We have on average 5 different servers that are to be loaded (all within the same network). Sun has provided a host of patches to try in different combinations but none seem to improve our response times. We do have services that are using eGate SRE IQ Managers (for bridging between old and new services) but I would think this would be supported. Any suggestions would be appreciated (CAPS 5.1.2 on Solaris 9/10)
Edited by: chrisallen21 on Jul 7, 2008 6:54 AM

John,
Can you quote the Incident number you have open? We have opened one as well and would like to reference it. We did find that for domains that are down/unreachable, it takes a default 2 minutes to give up looking for it which accounts for a fair amount of the wait time but not all of it (i.e. it took 10 minutes to populate the treeview and only 4 minutes for used looking for unreachable domains). I believe the parameter to change the default timeout is connectTimeout=120000 in the monitor.properties file in the conf directory. Our current tree has about 5 servers (about 7 domains in total) with approximately 50 services deployed.
Edited by: chrisallen21 on Jul 8, 2008 4:40 AM

Enterprise Manager Security Alert

Everytime time I bring up Enterprise Manager on one of my Servers (Windows 2003), I get a security alert. I thought this was normal until the security people at my office told me my server was at risk due to this alert. I have thought that this is normal until I check the detail of the security alert and found out that it says "This certificate cannot be verified up to a trusted certification authority".
Detail:
Issue to: mybd.everbody.gov
Issue by: mybd.everbody.gov
Valid from 12/2/2009 to 12/1/2019
Any suggestion will be appreciated.

You might want to check
Oracle® Enterprise Manager Advanced Configuration
10g Release 2 (10.2)
B16242-03
http://download.oracle.com/docs/cd/B14099_19/manage.1012/b16242/security2.htm#i1039745
It describes all about security and the way you should handle Security Certificate ALerts in the browser
Regards
Rob

Oracle Enterprise Manager installed but page is not loading

hi, I have installed Oracle Enterprise Manager on a Oracle Linux 5.8 box. version being Oracle.x64.11gR2. The installation was successful and I am able to start the oms services
via <ORACLE_INSTALL_HOME>/middleware/oms/bin ./emctl start oms. However, when I browse the location using browser
i.e. https://localhost:7801/em the page is indefinitely waiting for response. Any particular log files or any specfic pre-requisite I am missing that could be causing this. Kindly provide your inputs.

I don't think it's because of ageing out because:
1. I am trying to view the plan almost at the same time that the statement is run.
2. I am the only person running anything on the database. Shared pool is about 700M, SGA size is 2GB. PGA is 7GB.
What else would you need to know?

Sending email notifications/alerts via Enterprsie Manager 11g

I have Oracle 11g on Linux...and i have set up SMTP of my Enterprise Manager Database control. I click on Test and an email was sent successfully.
What i dont know is how to add more emails as the receipient of the alerts.
I have setup metrics like number of sessions, cpu usage, buffer cache hits, swapping, tablespace usage, etc...
Upon saving and waiting for transactions to come in, alerts were shown the the EM homepage..but i did not receice any emails... :(
What step did I miss? Any advise....
Edited by: InquisitiveDBA on Feb 24, 2013 7:30 AM

Hi,
I've had loads of issues with this as well. I haven't done it in a while but here are the notes that one of my guys has on it:
Configuration For Administrators
1. Roles: Public
2. System Privileges: View any target
3. Target Privileges: View the database on which the job runs
4. Subscribe to Rules: Subscribe to the rule containing (or that will contain) the job in question
For the Job you want notification sent out on
1. Access: The administrator should be added with View level access
2. E-mail Notification: Only affects e-mails sent to SYSMAN group
Notification Rules
1.Public: Yes
2.Jobs: The required job should be added with the appropriate status
3.E-mail Notification: Does not have to be checked
4.Subscribers: The administrator from above
Notification Schedule
1.Administrator: The administrator from above
2.Schedule: All time slots should be filled in with the appropriate e-mail address(es)
Hopefully that helps. I couldn't explain it in detail without going through the process myself...

Oracle Enterprise Manager 10gR4 is Released

Here is the press release. Material is already posted on OTN. Enjoy!
Oracle(R) Enterprise Manager Delivers Unique New Capabilities for Top-Down Application Management
Tuesday November 13, 8:15 am ET
New Capabilities Address Cross-Tier Performance Diagnostics, Integrated Incident and Problem Management and Application Data Sharing
ORACLE OPENWORLD, SAN FRANCISCO, Nov. 13 /PRNewswire-FirstCall/ -- Building upon its "top-down" approach to managing application and Information Technology (IT) infrastructure resources, Oracle today announced key new enhancements to Oracle® Enterprise Manager 10g. The new enhancements deliver a broad range of capabilities, across the application lifecycle, for custom and SOA-based applications and pre-packaged applications and include faster performance diagnostics for production applications; simplified incident and problem management through integration with Oracle's Siebel CRM HelpDesk and PeopleSoft Enterprise HelpDesk applications; expanded capabilities to manage Oracle Business Intelligence; and easy and secure data sharing across the application lifecycle.
Oracle's top-down approach to application management uniquely addresses the need for an end-to-end application perspective -- whether technology-focused or business-focused -- while enabling administrators to virtually eliminate the confusion caused by low-level system alerts that cannot be easily correlated to business exceptions. This has been a common and costly shortfall in traditional systems management software and frequently contributes to prolonged application downtime.
"Helping customers drive better business results is the essence of Oracle's top-down approach to application management," said Richard Sarwal, senior vice president of Systems Management Products, Oracle. "With the addition of Connectors enabling closed-loop incident and problem tracking with Siebel CRM and PeopleSoft Enterprise HelpDesk applications and breakthrough application performance management capabilities using Oracle Enterprise Manager's Application Diagnostics for Java feature, Oracle takes another significant step in fulfilling our top-down application management strategy."
"With this new release, Oracle is expanding both the breadth and depth of the application management capabilities available through Oracle Enterprise Manager, providing features designed for detailed diagnosis of Java application performance issues in production environments, with the ability to trace transactions between Java application servers and database sessions," according to Tim Grieser, Program Vice President, Enterprise System Management Software, IDC
Taking Application Performance Management to the Next Level
The new Oracle Application Diagnostics for Java capability in Oracle Enterprise Manager can monitor and diagnose performance problems in production Java applications with minimal impact on application performance and availability. Unlike most Java performance management tools that utilize bytecode instrumentation, Oracle Enterprise Manager's new feature has a highly innovative diagnostics engine that introduces less than one percent performance overhead on production applications and does not require IT operations to have a deep understanding of the application logic. The new feature also has a hot-pluggable architecture that enables the diagnostics engine to be deployed in heterogeneous middleware environments, without shutting down the application server. Oracle Application Diagnostics for Java is included in the Oracle Application Server Diagnostics Pack and the new Oracle Diagnostics Pack for Non-Oracle Middleware.
Improving Performance Diagnostics Across Application Tiers
Diagnosing end user performance problems has become increasingly difficult because of the increased deployment of multi-tiered web applications and the dynamic nature of web-based applications. With Oracle Application Diagnostics for Java, administrators now have powerful diagnostic capabilities by enabling application administrators to trace in-flight transactions and correlate user request activities in the Java application to corresponding activity in the database. Similarly, database administrators can quickly identify Java applications that may be causing resource contention in the database. This cross-tier diagnostics helps accelerate problem resolution by identification of resource bottlenecks across tiers.
Tighter Integration for Siebel CRM Diagnostics and Oracle Enterprise Manager
Siebel CRM diagnostic capabilities have been integrated with Oracle Enterprise Manager, providing a single integrated tool for problem diagnostics in Siebel CRM and its underlying infrastructure. Siebel CRM administrators can use Oracle Enterprise Manager for end-to-end management of Siebel CRM and to simplify Siebel CRM diagnostics based on user and server performance analysis. In addition, Siebel CRM customers can benefit from broader application support including Siebel CRM 8.0, 7.8 and 7.7.
Enhancing Management Capabilities Across the Application Lifecycle
New Management Connectors for Siebel CRM HelpDesk and Peoplesoft Enterprise HelpDesk applications allow the deployment of end-to-end IT incident and problem management solutions capable of unprecedented levels of automation and closed-loop remediation. The connectors enable Siebel CRM and PeopleSoft Enterprise HelpDesk customers to benefit from an integrated solution that spans the service lifecycle from incident logging and tracking to top-down management of applications including problem management, patching and provisioning across the technology stack. The bi-directional integration between Oracle Enterprise Manager and Siebel CRM and PeopleSoft Enterprise HelpDesk applications helps streamline key support processes by automatically creating service requests for incidents captured in Oracle Enterprise Manager. Support analysts using these help desk products are automatically empowered with respective corrective actions to accelerate problem resolution.
Expanded Capabilities for Business Intelligence Management
Oracle Management Pack for Business Intelligence is a new Oracle Enterprise Manager pack that extends the benefits of top-down application management to Oracle Business Intelligence Enterprise Edition Plus. Using the BI management pack, customers can receive proactive alerts on impending problems, troubleshoot and tune performance problems, manage configurations, proactively enforce configuration policies, and enforce service level agreements on Oracle Business Intelligence Enterprise Edition Plus deployments.
Easy and Secure Data Sharing Across the Application Lifecycle
Also new is the Oracle Data Masking Pack, to protect sensitive production data by using masking rules to replace confidential or sensitive production information with realistic but scrubbed data. This pack provides a scalable, secure solution for application data sharing for a wide variety of needs ranging from application development, testing, quality assurance as well as business processing. See related news release dated today.
Availability and Pricing
The enhancements listed above are generally available today with Oracle Enterprise Manager 10g Release 4.
Oracle Application Server Diagnostics Pack and Oracle Diagnostics Pack for Non-Oracle Middleware are priced at $6,000 each per CPU. Management Connectors are priced at $5,000. The Oracle Management Pack for Business Intelligence is priced at $10,000 per CPU. The Oracle Data Masking Pack is priced at $10,000 per CPU of the server where data is being masked.
For more information about Oracle Enterprise Manager 10g Release 4 and the family of Plug-ins, Connectors and Management Packs and to download free evaluation software go to: http://www.oracle.com/enterprisemanager/. Terms, conditions and restrictions apply.
About Oracle Enterprise Manager 10g
Spanning applications, middleware, and database management, Oracle Enterprise Manager 10g delivers integrated enterprise management for Oracle and non-Oracle systems to more than 21,000 Management Pack customers worldwide. Employing a unique "top-down" approach to managing application and IT infrastructure resources, Oracle Enterprise Manager allows customers to focus on what matters for its business -- greater agility, better service quality and lower operational costs. With a broad set of administration, configuration management, provisioning, end-to-end monitoring, service level management, and security capabilities, Oracle Enterprise Manager 10g helps customers manage service levels, proactively isolate business exceptions before they become emergencies, and remediate issues at any level spanning application and heterogeneous infrastructure -- all within one management solution. Learn more at http://www.oracle.com/enterprisemanager/.
About Oracle
Oracle (Nasdaq: ORCL - News) is the world's largest enterprise software company. For more information about Oracle, visit our Web site at http://www.oracle.com.
Trademarks
Oracle is a registered trademark of Oracle Corporation and/or its affiliates. Other names may be trademarks of their respective owners.
This announcement is provided to you solely for information purposes, is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. Many factors can materially affect Oracle's product development plans and the nature and timing of future product releases. The development, release, and timing of any features or functionality described remains at the sole discretion of Oracle. This information may not be incorporated into any contractual agreement with Oracle or its subsidiaries or affiliates. Oracle specifically disclaims any liability with respect to this information.
Message was edited by:
Chung Wu

This is expected behavious. It is because Enterprise Manager cannot run with the new server name. You can either go through loads to change the config files or simply drop and recreate the console.

Enterprise Manager Newbie Q's

Greetings All,
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
Windows 2008 R2Last night one of the production users received an ORA-01652: unable to extend temp segment by 128 in tablespace TEMP error.
Hence, my manager has requested that I
“Please configure our database servers to notify us via email when critical thresholds are exceeded.
This is a standard config option within enterprise manager”.I have very little experience/exposure with EM. In fact I’ve found several terms that I think all refer to the same thing (correct me if I’m wrong please), for example : "OEM Database Control", "OEM Grid Control", and something called "The Business Service Management (BSM)" capabilities of Oracle Enterprise Manager.
Are these different products/tools, or different implementations of the OEM?
We are not a RAC shop; Is the term GRID is also used to indicate a RAC environment?
I am looking for some good "How To Use/Configure OEM" sources of information. Any suggestions would be appreciated.
I’m looking a purchasing the following books, any recommendations on these or others?
Oracle Enterprise Manager 10g Grid Control Handbook (Oracle Press) - Werner De Gruyter
Oracle Enterprise Manager Grid Control 11g R1: Business Service Management - Ashwin Kumar Karkala, Govinda Raj Sambamurthy
It seems that there is a data repository for the OEM. Is this generally required? Can I set things up such that I can get e-mail notifications when tablespace(s) exceed a certain usage threshold without the repository?
I know I’ve rambled about some, thanks for your patients.
Thanks for your time,
Bob

I can answer a couple of your questions...
"OEM Database Control" is essentially "Oracle Enterprise Manager for a single database". "OEM Grid Control" (or, the current release, "OEM Cloud Control" is essentially "Oracle Enterprise Manager for several databases", plus the ability to monitor server hosts, web applications, and quite a few other non-database items. They are similar, but different products. Database Control generally gets installed along with the database software itself, while Grid/Cloud Control is a completely separate installation of a standalone product. I have no idea what the BSM bit refers to.
Database Control uses the one monitored database as its data repository (in the SYSMAN schema). Grid Control also requires a repository, but you are allowed to install a separate database, without additional licensing fees, to serve as the repository for Grid Control so long as you don't use it for anything else (other than as an RMAN recovery catalog). If you don't want to install another database to serve Grid/Cloud Control, you can use an existing database, but it's generally not recommended.
Neither Database Control nor Grid/Cloud Control will function without a repository. Either of them will allow you to set up alert notifications when tablespace(s) exceed your configured thresholds, or when various other events occur (DB shuts down, filesystem getting full, CPU load too high, too many sessions, whatever you like).
The term "Grid" is used in "Oracle Grid Infrastructure", which is part of the clustering package that supports RAC. "Grid" is just a marketing term there, RAC and Grid Control are orthogonal concepts, neither requires the other but they can also be used together.
I can't comment on the books.

CPU load alert on Enterprise Manager

Similar Messages

Maybe you are looking for