Tablespace Threshold Metrics (exceptions)
(I entered a SR but no answer as of yet)
I created to overriding thresholds for Tablespace Full, one for TEMP, the other for UNDOTBS. Problem Those tablespaces still trigger alerts based on the thresholds set for the "default/other" tablespace metrics.
Any one get this working in 10gR2? I believe it worked in 10gR1 Grid.
I dont see it tracing problemTbsp.pl or fullTbsp.pl scripts. I cant figure out which perl script is being run such that I can debug the problem.
Help?
started and stopped many times, upload metrics, clearstate etc.
I believe there is something fundamentally wrong. Has anyone else tried this functionality? Is it broke elsewhere?
Similar Messages
-
Hi all,
my env..
Oracle 8i
solaris
I am trying to check tablespace threshold chek ..below query ...this query is not working in oracle 8i ...but it is working fine in oracle 10g machines ...
what is the reason I was unable to findout...
Please let me know if any one knows
select
(select INSTANCE_NAME||'_'||HOST_NAME from v$instance ) "db_name_host_name",
tablespace_name tablespace,
sum(file_cnt) "# of Files",
(sum(tp)-sum((fs + (tp-tc))))/1048576 "Used (M)",
sum(tc)/1048576 "Allocated (M)",
sum(tp)/1048576 "Max Extend (M)",
round(max(100 - (100*((fs + (tp-tc))/tp)))) " % of Max Used"
from ( select b.tablespace_name, nvl(a.free_space, 0) fs,
b.total_poss_space tp,
b.total_curr_space tc,
b.file_cnt,
b.tablespace_name ts_name
from (
select sum(bytes) free_space,
tablespace_name
from dba_free_space
group by tablespace_name
) a,
select sum(decode(autoextensible, 'YES',
maxbytes,
bytes)) total_poss_space,
sum(bytes) total_curr_space,
count(1) file_cnt,
tablespace_name
from dba_data_files
group by tablespace_name
) b
where a.tablespace_name (+) = b.tablespace_name
group by tablespace_name
having max(100 - (100*((fs + (tp-tc))/tp))) >=80
Error at line 2
ORA-00936: missing expression
Script Terminated on line 1.
Edited by: user603328 on Dec 10, 2010 7:26 AMThanks ur reply...
I am getting below issue....
select
(select INSTANCE_NAME||'_'||HOST_NAME from v$instance ) "db_name_host_name",
tablespace_name tablespace,
sum(file_cnt) "# of Files",
(sum(tp)-sum((fs + (tp-tc))))/1048576 "Used (M)",
sum(tc)/1048576 "Allocated (M)",
sum(tp)/1048576 "Max Extend (M)",
round(max(100 - (100*((fs + (tp-tc))/tp)))) " % of Max Used"
from ( select b.tablespace_name, nvl(a.free_space, 0) fs,
b.total_poss_space tp,
b.total_curr_space tc,
b.file_cnt,
b.tablespace_name ts_name
from (
select sum(bytes) free_space,
tablespace_name
from dba_free_space
group by tablespace_name
) a,
select sum(decode(autoextensible, 'YES',
maxbytes,
bytes)) total_poss_space,
sum(bytes) total_curr_space,
count(1) file_cnt,
tablespace_name
from dba_data_files
group by tablespace_name
) b
where a.tablespace_name (+) = b.tablespace_name
group by tablespace_name
having max(100 - (100*((fs + (tp-tc))/tp))) >=80
Error at line 2
ORA-00936: missing expression
Script Terminated on line 1. -
Script for database send mail for tablespace threshold 90%
Dear All,
Scripts required for send mail from database for following events
1.Tablespace Threshold
2. Db Shutdown & Startup
Database Version : 10.2.0.1.0
OS :- Linux
Thanks & Regards,
Praful KawajiYou have Database Control.
Database Control can already do this.
Why do you want to remain in the Stone Age and continue to use scripts?
Why do you want to reinvent the wheel, when the functionality is already there.
Agreed, you would have to read documentation and/or help text, and you rather don't want to do that.
Which is the exact cause why you want to reinvent the wheel, and pay Larry for functionality you don't use.
Sybrand Bakker
Senior Oracle DBA -
Hi,
I executed the below procedure to implement the tablespace threshold for the tablespace 'TESTING'.
My doubt is, once the threshold reaches where will I get the message like the tablespace 'TESTING' has only 51200 KB left.
Database:10.2.0.4
OS: solaris 10
BEGIN
DBMS_SERVER_ALERT.SET_THRESHOLD(
metrics_id => DBMS_SERVER_ALERT.TABLESPACE_BYT_FREE,
warning_operator => DBMS_SERVER_ALERT.OPERATOR_LE,
warning_value => '51200',
critical_operator => DBMS_SERVER_ALERT.OPERATOR_LE,
critical_value => '40960',
observation_period => 1,
consecutive_occurrences => 1,
instance_name => NULL,
object_type => DBMS_SERVER_ALERT.OBJECT_TYPE_TABLESPACE,
object_name => 'TESTING');
DBMS_SERVER_ALERT.SET_THRESHOLD(
metrics_id => DBMS_SERVER_ALERT.TABLESPACE_PCT_FULL,
warning_operator => DBMS_SERVER_ALERT.OPERATOR_GT,
warning_value => '0',
critical_operator => DBMS_SERVER_ALERT.OPERATOR_GT,
critical_value => '0',
observation_period => 1,
consecutive_occurrences => 1,
instance_name => NULL,
object_type => DBMS_SERVER_ALERT.OBJECT_TYPE_TABLESPACE,
object_name => 'TESTING');
END;
/Just curious, why are you setting the second threshold (TABLESPACE_PCT_FULL) limit = 0 %?
Anyway, the view dba_outstanding_alerts does not reflect the status immediately. It may take a few minutes, even if the parameter observation_period = 1 (minute).
-- Here we have set the database to trigger an alert if the any tablespace usage exceeds 85%
SQL> select metrics_name, warning_operator, warning_value
2 from dba_thresholds
3 /
METRICS_NAME WARNING_OPERATOR WARNING_VALUE
Tablespace Space Usage GE 85
SQL> CREATE TABLESPACE "TS_TEST" LOGGING DATAFILE
2 'C:\ORADB\TRAIN10G\DATA\TS_TEST01.DBF' SIZE 400K AUTOEXTEND ON NEXT 100K MAXSIZE 800K
3 EXTENT MANAGEMENT LOCAL SEGMENT SPACE MANAGEMENT AUTO ;
Tablespace created
SQL> CREATE TABLE testtab (testcol CHAR(100)) tablespace TS_TEST;
Table created.
-- Here we wrote a script to insert 1000 rows at a time which will be run repeatedly until we see the threshold exceeded
DECLARE
N INTEGER;
BEGIN
FOR N IN 1..1000 LOOP
INSERT INTO testtab values('TESTTESTTEST');
END LOOP;
COMMIT;
END;
... repeated a few times, checking the freespace each round with the query below until the threshold is exceeded
SQL> SELECT a.tablespace_name
2 ,trunc(decode(a.autoextensible,'YES',a.maxsize-a.bytes+NVL(b.free,0),'NO',NVL(b.free,0))/1024) free_kb
3 ,trunc(a.bytes/1024) size_kb
4 ,trunc(a.maxsize/1024) maxsize_kb
5 ,a.autoextensible ae
6 ,trunc(decode(a.autoextensible,'YES',(a.maxsize-a.bytes+NVL(b.free,0))/a.maxsize*100,'NO',NVL(b.free,0)/a.maxsize*100)) free_pct
7 FROM ( select tablespace_name
8 ,autoextensible,sum(bytes) bytes
9 ,sum (decode(autoextensible,'YES',maxbytes,bytes)) maxsize
10 from dba_data_files
11 group by tablespace_name,autoextensible) a,
12 (select tablespace_name ,sum(bytes) free
13 from dba_free_space
14 group by tablespace_name) b
15 where a.tablespace_name=b.tablespace_name(+)
16 order by 2 asc
17 /
TABLESPACE_N FREE_KB SIZE_KB MAXSIZE_KB AE FREE_PCT
TS_TEST 16 784 800 YES 2
... SQL below was repeatedly run, and after 4 minutes we got the alert in dba_outstanding_alerts....
SQL> select object_name, object_type, reason
2 from dba_outstanding_alerts
3 /
OBJECT_NAME OBJECT_TYPE REASON
TS_TEST TABLESPACE Tablespace [TS_TEST] is [88 percent] fullAlso, you can expand this method to include a stored procedure that is scheduled (DBMS_SCHEDULER) to check dba_outstanding_alerts every so often and compose an email message to be sent to the DBA using UTL_SMTP. -
Hi,
We have metrics like Tablespace Used (%) for which we have set thresholds, but they are not being collected (as confirmed by the All Metrics screen) - why?
Agent Version (OMS) -- 10.1.0.3
Agent Version of target -- 10.1.0.3
No error messages are issued
regards
DilipGo thru the monitoring configuration link at the bottom again. Also try setting the PERL trace in emd.properties to debug and look at the perl.trc file generated for clues.
-
Hi
I want to create a threshold on the tablespace when it is 80 % Used.
I want a mail to be sent on my email address and also when user inserts data in that TBS he should receive the Alert.
Please AdviseUse Oracle Enterprise manager database control, go to preferences and assign a email server. Then on the home page-->manage metrics link-->Edit Threshod button-->set your metric warning/critical thresholds here and wait for notifications in email box when the threshod is reached.
Sami. -
How to change "tablespace allocation" Metric chng from 7 days to 1 day?
Hi
I got to change metric collection time from 7 days (i.e once in every week) to every day. I have not find any tab to modify, please let me know how can I modify.
Here are the details.
Metric Name: Tablespace Allocation None
Sub tabs : Tablespace Allocated Space (MB) Not Set
Tablespace Used Space (MB) Not Set
Thanks,
RKAh, you can't actually change it, at least not through the front end. From the Oracle® Enterprise Manager Oracle Database and Database-Related Metric Reference Manual
10g Release 2 (10.2)
Part Number B25986-01:
This metric calculates the space allocated for each tablespace. It is not intended to generate alerts. Rather it should be used in conjunction with the Allocated Space Used (MB) metric to produce an historical view of the amount of space being used and unused by each tablespace.
For all target versions, the collection frequency for this metric is every 7 hours.
The metric is only designed for historical information, looks a typo on the 7 hours, mine is also set to 7 days. -
Tablespace threshold checking without grid control
Hi all,
DB oracle 10g
solaris,
It is new env ....we are not having grid configuration ....
Please any one provide to me a script for monitoring threshold limits like >90% critcal alert ....
client wants to these alerts are configure to mail when ever tablespace reaches thershold....
please help me out if any one having any shell ,sql scripts....
thanks,
kkuser603328 wrote:
Hi all,
DB oracle 10g
solaris,
It is new env ....we are not having grid configuration ....How about dbcontrol? It has all the same functionality as Grid, only for a single database instance. It can monitor and email as specified. It doesn't cost any extra. If your client has Oracle 10g, then he has, and has paid for, dbcontrol. If they don't know that, you owe it to them to show them.
Please any one provide to me a script for monitoring threshold limits like >90% critcal alert ....
client wants to these alerts are configure to mail when ever tablespace reaches thershold....
please help me out if any one having any shell ,sql scripts....
thanks,
kk -
Tablespace threshold alerts?
Maybe 2 silly questions...
Say if I have a tablespace 80% full and then I set the warning threshold down to 75%, will the alert be generated immediately or delayed (how much?)
If I don't get warning even after 1 day ( with restart database ), what was wrong?Hi,
I have checked the dba_threshold view the threshold has been set correctly as I saw in EM. (warning: 75, critical: 85, while my tablespace is now 77% full )
and there are no rows for dba_outstanding_alerts.
I remember there was an exam question like this as well but I don't know the correct answer. -
Track ip route metric threshold
Hi guys
Is there someone who can explain the follow output:
c3750#sh track
Track 1
IP route 8.8.4.4 255.255.255.255 metric threshold
Metric threshold is Up (EIGRP/28416/11)
147 changes, last change 00:00:04
Metric threshold down 210 up 190
First-hop interface is Vlan1
Track 2
IP route 8.8.8.8 255.255.255.255 metric threshold
Metric threshold is Up (OSPF/20/20)
40 changes, last change 01:13:36
Metric threshold down 255 up 254
First-hop interface is Vlan1
I've not found documentation about line :
(EIGRP/28416/11) and (OSPF/20/20)
I've not understood which is the metric that track match with his configuration:
Thanks many much in advance
Daniele
Some configuration detail:
track 1 ip route 8.8.4.4 255.255.255.255 metric threshold
threshold metric up 190 down 210
track 2 ip route 8.8.8.8 255.255.255.255 metric threshold
c3750#sh ip route
O E2 8.8.8.8 [110/20] via 192.168.0.100, 00:00:04, Vlan1
D EX 8.8.4.4 [170/28416] via 192.168.0.1, 00:00:31, Vlan1As far as tracking goes, it will be "Up" as long as there is a route to 7.1.1.7/32 in the routing table of the routers configured with track 1. It must be a /32 route due to the "255.255.255.255" in the track 1 statement. With that being said, track 1 will be "Down" since e1/0 on r7 is most likely not configured with a /32 mask. I say most likely because it is conneced to R2 and ethernet interfaces cannot be configured with a /32 mask. Attempting to would result in "Bad mask /32 for address 7.1.1.7". To bring track 1 up, the mask needs to match the route for 7.1.1.7 exactly as seen in the routing tables of R3 and R4.
With the current topology any issue causing eigrp on R2 to not advertise 7.1.1.7/32 will take track 1 down on both R3 and R4. The HSRP priority on both routers will decrement 10 which means that R3 will never take priority. No failover will occur as a permanent state in a converged network.
Another point to make in this scenario is that tracking 7.1.1.7/32 could cause instability in the network. If connectivity between R2 and R7 fails, HSRP will failover. This topology change causes updates and and traffic disruptions in an unrelated part of the network. Traffic destined to R2 or possibly another router off R2 is now affected even though only connectivity to R7 is of concern. Also, R2's hold timer for R7 could cause long delays for eigrp updates to make it down to R3 and R4.
Optimally, R3 and R4 should track the status of an object specifically related to their own connectivity to R2 such as interface line-protocol or IP. This will give the network stability, increase routing accuracy, cause faster failover, and better avaibility, etc, etc, etc.
Check out Configuring Enhanced Object Tracking @ http://www.cisco.com/c/en/us/td/docs/ios-xml/ios/ipapp/configuration/15-mt/iap-15-mt-book/iap-eot.html
Ian -
Hello - I'm new to OEM 12c and I'm still getting my head around the monitoring aspects of it. I just added two temporary tablespaces to a database. When I look at the tablespaces via OEM12c, I notice that the Avaliable Space Used (%) Alerts column has a check mark for all the tablespaces except the two I just created. The original TEMP tablespace does have check in the column. Will these two new tablespaces be monitored for used space?
Thanks!
ShawnI'm assuming you're on DB 11.2 or higher... Temp and Undo are typically excluded from monitoring due to their cyclical nature. Thresholds must explicitly be set from 11.2 and higher. Please see this note for more details. Another explanation might be timing... if you just added them, and the tablespace collection hasn't run, you won't see that check because the metric hasn't validated them yet. If you go back after the collection runs, you should see (if they were a normal tablespace and the temp/undo monitoring wasn't in play)...
How to - Exclude UNDO and TEMP Tablespaces From The Tablespace Used (%) Metric (Doc ID 816920.1) -
Hai,
i am registering new product using with adsplice in 11.5.10.2 environment.
in newproducts.txt
product=ego
base_product_top=*APPL_TOP*
oracle_schema=ego
sizing_factor=100
main_tspace=?????????/
index_tspace=????????
temp_tspace=?????
default_tspace=??????
what wew ill mention for main_tspce????????? simlirly to index and default also.
Regards
dbaIt is recommended not to have autoextend on for any tablespace as it will be difficult to manage datafiles that grow very large. Example if your datafile are very big in size then while doing an image copy of the files if you get any copy error in between then you have to start the copy from beginning. But if the files were small and you were doing a copy and after few files you get the copy error then you can continue from the last failed file onwards. Another example is if you don't have chunk of small disks to store your image copy backupsets then small files will fit on those disks. Recommended size of the file is maximum 8 to 10GB.
Also set Tablespace%userd metrics warning and critical threshold values so that you get an alert once a tablespace reaches that threshold. -
How to capture tablespace warning and critical messages
Hi,
Where in the database dictionary, the tablespace threshold percent (warning and critical) defined by the user is stored.
Requirement is to set threshold percent (warning and critical) in OEM and in a function (for User-Defined Metrics) or procedure (for scheduler) check for Tablespace PCT_FULL against the set threshold percent and display message accordingly.
Thanks & Regards,
Sanju.Sounds like this is default behaviour of the metrics for tablespace full?
Here is the basic query for OEM metrics:
select
target_type,
target_name,
metric_column Metric,
decode(warning_threshold,' ',null, decode(warning_operator,'Equal To','=','Not Equal','!=','Greater Than','>','Less Than','<',warning_operator) || ' ' || warning_threshold) Warning,
decode(critical_threshold,' ', null, decode(critical_operator,'Equal To','Not Equal','!=','=','Greater Than','>','Less Than','<',critical_operator) || ' ' || critical_threshold) Critical
from mgmt$metric_collection
--where target_type = 'oracle_emd'
--and metric_column = 'HostInfovszGrowth'
--and warning_threshold != '20'
--where target_type = 'host'
where-- metric_column like 'tnsPing'
-- (warning_threshold != '80' and critical_threshold != '95')
--and
target_type = 'oracle_database'
--and target_type = 'host'
--and
--target_type = 'oracle_listener'
--target_type = 'oracle_emd'
--and metric_column like 'generic%'
--and warning_threshold like '%ORA%'
and target_name = 'XXXSOMEHOSTNAMExxxx'
and metric_column = 'pctUsed'
order by 1,3,2 -
Cell definition / Cell exception
Hi,
Can you please explain me What are the situations we are using Cell definition and Cell exception in BEx Query??
Regs
KuttanHi,
in exception reporting you select and highlight objects that are in some way different or critical. Results that fall outside a set of predetermined threshold values (exceptions) are highlighted in color or designated with symbols. This enables you to identify immediately any results that deviate from the expected results.
Exception reporting allows you to determine the objects that are critical for a query, both online, and in background processing.
http://help.sap.com/saphelp_nw04s/helpdata/en/1a/615f64816311d38b170000e8284689/frameset.htm
You can define threshold values (exceptions) for a query. Data that varies from these thresholds is highlighted in color or marked with icons. You can use these exceptions to identify deviations from expected results at a glance.
http://help.sap.com/saphelp_nw04s/helpdata/en/9d/76563cc368b60fe10000000a114084/frameset.htm
Regards
Andreas -
Second Undo Tablespace in Oracle10g v.2...
Hi ,
Having connected as sys i created another undo tablespace ... except for the first one created by default....
SQL> CREATE UNDO TABLESPACE INFO_UNDO_TBL DATAFILE 'C:\oracle\product\10.2.0\oradata\INFO\INFO_UNDO_DFL.ORA' SIZE 10M;
Tablespace createdI have two questions...
1) When i issue the command:
SQL> SHOW PARAMETER UNDO_TABLESPACE;
NAME TYPE VALUE
undo_tablespace string UNDOTBS1It does not refer the new undo tablespace..... I suppose that i have to write:
sql> alter system set undo_tablespace=UNDOTBS1,INFO_UNDO_TBL scope =BOTH... or something like this....Is that correct....??????
2)The Oracle Doc (Concepts) points out.....
"In automatic undo
management mode, each Oracle instance is assigned one (and only one) undo
tablespace."
That means i cannot issue the command "alter system set undo_tablespace=UNDOTBS1,INFO_UNDO_TBL" (if it is correct) since the default undo tablespace runs in AUTO mode.....????
SQL> SHOW PARAMETER UNDO_MANAGEMENT;
NAME TYPE VALUE
undo_management string AUTOMany thanks....
SimHi ,
I dont think you can have two undo tablespaces assigned for a database.This is what you get in the alert log files even if you are deceived by the statement "System Altered"
SQL> alter system set undo_tablespace='UNDOTBS1,UNDOTBS2' scope=spfile;
System altered.
SQL> sho parameter undo
NAME TYPE VALUE
undo_management string AUTO
undo_retention integer 900
undo_tablespace string UNDOTBS1,UNDOTBS2
Alert log extract..............
Errors in file j:\oracle\product\10.2.0\admin\orcl10g\udump\orcl10g_ora_1724.trc:
ORA-30012: undo tablespace 'UNDOTBS1,UNDOTBS2' does not exist or of wrong type
Errors in file j:\oracle\product\10.2.0\admin\orcl10g\bdump\orcl10g_reco_1808.trc:
ORA-30012: Message 30012 not found; No message file for product=RDBMS, facility=ORA
Tue Oct 02 15:37:41 2007
Errors in file j:\oracle\product\10.2.0\admin\orcl10g\bdump\orcl10g_smon_1760.trc:
ORA-30012: Message 30012 not found; No message file for product=RDBMS, facility=ORA
Instance terminated by USER, pid = 1724
ORA-1092 signalled during: alter database open...
Regards,
-Praveen
http://myracle.wordpress.com
Maybe you are looking for
-
Change BSEG-ZUONR during the Post Goods issue.
Hi We wanted to Update the BSEG-ZUONR with sales Order no( traverse back to the Original SO during reversal) ( Basically Assignment field (ZUONR) in the Accounting side) after performing the Post Goods issue USING TRANSACTION VL02N. Is there any us
-
Can you select part of a document for a word count and can you determine word count from an entire pages document?
-
TS3212 Error message on trying to set up itunes ??
"itunes setup.exe is not a valid WIN32 application" Can anybody please help me - I am on Windows 7
-
I downloaded the updated Firefox 4 on my macbook pro. I am currently using Firefox 3.5 with no problem. With the new download the firefox will not connect to the internet. Help please?
-
Hi all Can anybody tryed BC to XI or XI to BC senario. Tahnkx