Orabbix:"Another way to monitor Oracle reloaded"

[tof233]

Hello,
I have a problem with Orabbix 1.1 . It's working until I got this error messages :

2011-02-02 10:09:17,567 [pool-1-thread-44] INFO  Orabbix - Done with dbJob on database Cody3 QueryList elapsed time 5 ms
 2011-02-02 10:10:17,590 [pool-1-thread-45] INFO  Orabbix - Done with dbJob on database Cody3 QueryList elapsed time 4 ms
 2011-02-02 10:12:17,761 [main] ERROR Orabbix - Error diuring alive testing for dbname ->Cody3org.apache.commons.dbcp.SQLNestedException: Could not retrieve connection info from pool
 2011-02-02 10:13:17,936 [main] ERROR Orabbix - Error diuring alive testing for dbname ->Cody3org.apache.commons.dbcp.SQLNestedException: Could not retrieve connection info from pool

I tried to change Cody3.MaxWait to 200 but still have the same problem.
I noticed it's always after pool-1-thread-45.

Have you an idea on how I could fix it?

[dalle]

please be patience i'm just asking: database is open and the listener is running?
because on my environment don't happen, I have a database that is closed during the night for 3 hour to do a cold backup but when the rman script open the database everything return normal.
if the problem persist put log on debug and send me the complete log on andrea.dallevacche(AT)gmail.com
let me know

Hello,
I have a problem with Orabbix 1.1 . It's working until I got this error messages :

2011-02-02 10:09:17,567 [pool-1-thread-44] INFO  Orabbix - Done with dbJob on database Cody3 QueryList elapsed time 5 ms
 2011-02-02 10:10:17,590 [pool-1-thread-45] INFO  Orabbix - Done with dbJob on database Cody3 QueryList elapsed time 4 ms
 2011-02-02 10:12:17,761 [main] ERROR Orabbix - Error diuring alive testing for dbname ->Cody3org.apache.commons.dbcp.SQLNestedException: Could not retrieve connection info from pool
 2011-02-02 10:13:17,936 [main] ERROR Orabbix - Error diuring alive testing for dbname ->Cody3org.apache.commons.dbcp.SQLNestedException: Could not retrieve connection info from pool

I tried to change Cody3.MaxWait to 200 but still have the same problem.
I noticed it's always after pool-1-thread-45.

Have you an idea on how I could fix it?

[tof233]

Thank you andrea,
There is no discontinuity on the database service.
It really seems to be something on the client side.
I sent you my log file.

[tof233]

Thanks to Andrea, I fixed the problem by removing "locks" and "audit" from QueryList
So now, I have in query.props:

QueryList=archive,dbblockgets,dbconsistentgets,dbhitratio,dbphysicalread,dbversion,hitratio_body,hitratio_sqlarea,hitratio_table_proc,lio_current_read,maxprocs,maxsession,miss_latch,pga_aggregate_target, pga,phio_datafile_reads,phio_datafile_writes,phio_redo_writes,pinhitratio_body,pinhitratio_sqlarea,pinhitratio_table-proc,pinhitratio_trigger,pool_dict_cache,pool_free_mem,pool_lib_cache,pool_misc,pool_sql_area,procnum,session_active,session_inactive,session,session_system,sga_buffer_cache,sga_fixed,sga_java_pool,sga_large_pool,sga_log_buffer,sga_shared_pool,tbl_space,userconn,waits_controfileio,waits_directpath_read,waits_file_io,waits_latch,waits_logwrite,waits_multiblock_read,waits_singleblock_read,hitratio_trigger,lio_block_changes,lio_consistent_read,waits_other,waits_sqlnet,users_locked,uptime,dbfilesize,dbsize

Andrea: I hope, I can quote your explanation...

I have seen from the log file that the connections arrive to 10. This is the limit configured on connection pool.
So there are no more available connections and connection pool returns the error.
I've also seen that the execution time of the queries is very high, so probably there is a query that
slows down the system.
Queries may be slow down the system can be locks and the audit.
I think you should delete from the list QueryList =
the key "audit" and the key "locks"
I hope that will solve your problem.

Thank you

[dalle]

Thanks to Andrea, I fixed the problem by removing "locks" and "audit" from QueryList
So now, I have in query.props:

QueryList=archive,dbblockgets,dbconsistentgets,dbhitratio,dbphysicalread,dbversion,hitratio_body,hitratio_sqlarea,hitratio_table_proc,lio_current_read,maxprocs,maxsession,miss_latch,pga_aggregate_target, pga,phio_datafile_reads,phio_datafile_writes,phio_redo_writes,pinhitratio_body,pinhitratio_sqlarea,pinhitratio_table-proc,pinhitratio_trigger,pool_dict_cache,pool_free_mem,pool_lib_cache,pool_misc,pool_sql_area,procnum,session_active,session_inactive,session,session_system,sga_buffer_cache,sga_fixed,sga_java_pool,sga_large_pool,sga_log_buffer,sga_shared_pool,tbl_space,userconn,waits_controfileio,waits_directpath_read,waits_file_io,waits_latch,waits_logwrite,waits_multiblock_read,waits_singleblock_read,hitratio_trigger,lio_block_changes,lio_consistent_read,waits_other,waits_sqlnet,users_locked,uptime,dbfilesize,dbsize

Andrea: I hope, I can quote your explanation...


Thank you

Thank you for your feedback and your patience!

[dalle]

Orabbix 1.2.0-RC1 has been released


On this release:
Modified as suggested from Sean, Session’s graphs now “Active Session” is at the bottom of the
stack so you can see how close you are to reaching the trigger.
Improved scalability and multithreading now Orabbix can scale very easily
Now controls executed on databases are independent of each other
I removed almost completely sequential procedure on software so bottlenecks are higly reduced

Added query parameters:

• <QueryName>.Trim=[true|false] if true the resultset is trimmed (default is true)
• <QueryName>.AddSpaces=[true|false] if true the add a space between columns of resultset(default is true)
• <QueryName>.ExcludeColumnsList=[1,2,3,..n]= exclude from result set the 1st, the 2nd etc.. columns from resultset

Now Orabbix use ojdbc6 for Oracle 11.2
NOTE: This is a Candidare Release
This release is available for download here
and on sourceforge as usual

[Zete]

Is there a way for orabbix to connect to cluster itself not just to an instance on specified host?

I have an oracle rac with two hosts. I don't want to monitor inst1 and inst2 seperately. What I need to monitor is if the database is available at all.

[dalle]

Is there a way for orabbix to connect to cluster itself not just to an instance on specified host?

I have an oracle rac with two hosts. I don't want to monitor inst1 and inst2 seperately. What I need to monitor is if the database is available at all.

Yes is possible.
e.g.
you have two host RAC1 and RAC2 and one instance RACINST
you can write the connection string as follow:
RACINST.Url=jdbcracle:thin:@(DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(P ROTOCOL=TCP)(HOST=RAC1.EXAMPLE.COM)(PORT=1521))(AD DRESS=(PROTOCOL=TCP)(HOST=RAC2.EXAMPLE.COM)(PORT=1 521)))(CONNECT_DATA=(SERVICE_NAME=RACINST)))

Let me know

[Zete]

Works as a charm! Thank you!

[dalle]

Works as a charm! Thank you!

well !!!

[fritz]

Hi Team,

Thanks for your great job.

I'm a zabbix user for 1 year now and i'm very satisfied by this tool.
I'm trying to setup the orabbix pluggin (1.1.0) to monitor Oracle DB 11g R2 running on Windows 2008 R2 systems.
The parameters seems good, the connections is ok, i can check it in the logs.

But i cannot get any values on the zabbix server. Do you have an idea ? i cannot find anything in the logs.

The zabbix server is running on a Linux Suse, actually i download the zabbix appliances (1.8.4) to test the new features.

Any ideas ?

Thanks a lot.

[dalle]

You have followed "installation instruction" i mean you have created one host inside zabbix called exactly as your database inserted on config.props
DatabaseList
after you should import my template and associate to you db
if you don't see any data and is all right (i mean you don't have error on orabbix.log) there is something of wrong on cofiguration.
let me know

[fritz]

Here are the results i polled from orabbix.log :

 2011-02-28 14:55:05,529 [main] INFO  Orabbix - Starting Orabbix Version 1.1.0
 2011-02-28 14:55:05,540 [main] INFO  Orabbix - Orabbix started with pid:23167
 2011-02-28 14:55:05,540 [main] INFO  Orabbix - PidFile -> ./logs/orabbix.pid
 2011-02-28 14:55:05,977 [main] INFO  Orabbix - Connected as ZABBIX
 2011-02-28 14:55:05,980 [main] INFO  Orabbix - --------- on Database -> SITG01
 2011-02-28 14:55:06,364 [pool-1-thread-1] INFO  Orabbix - Done with dbJob on database SITG01 QueryList elapsed time 340 ms
 2011-02-28 14:55:06,603 [pool-1-thread-2] INFO  Orabbix - Done with dbJob on database SITG01 QueryList elapsed time 288 ms

I applied the Oracle template on my host. Everything is green in the items section.

[dalle]

Here are the results i polled from orabbix.log :

 2011-02-28 14:55:05,529 [main] INFO  Orabbix - Starting Orabbix Version 1.1.0
 2011-02-28 14:55:05,540 [main] INFO  Orabbix - Orabbix started with pid:23167
 2011-02-28 14:55:05,540 [main] INFO  Orabbix - PidFile -> ./logs/orabbix.pid
 2011-02-28 14:55:05,977 [main] INFO  Orabbix - Connected as ZABBIX
 2011-02-28 14:55:05,980 [main] INFO  Orabbix - --------- on Database -> SITG01
 2011-02-28 14:55:06,364 [pool-1-thread-1] INFO  Orabbix - Done with dbJob on database SITG01 QueryList elapsed time 340 ms
 2011-02-28 14:55:06,603 [pool-1-thread-2] INFO  Orabbix - Done with dbJob on database SITG01 QueryList elapsed time 288 ms

I applied the Oracle template on my host. Everything is green in the items section.

Orabbix is working fine! you should see inside latest data of in zabbix interface undee "SITG01" host all the data retrieved.

[fritz]

Problem solved. I named my host with a wrong name. Sorry about that.
Thanks for your help.