What is the query you're using to get the tablespace data? Some data is collected more frequently then others, so it depends on the metric.
Thanks for your Respond. I dont make the query by myself. I try to use the build in report elements, i have a german GUI so i try to translate the element caption in german but i dont know how they are exactly named in the english GUI:
- Space-Use by Tablespaces
- Space-Use by Datafiles
Would you recomend to make the query by myself? do you have one i could use for my purposes?The following metric i need to know:
- Tablespace Size
- Used Space in Tablespace (MB and %)
- Autoextend YES / NO
- Extend Size (How much will the Tablespace Grow in each Extend)
- Datafile Size
- Datafile Max Size
I hope i could give some usefull Information.
Not 100% sure, but I think the problem is because the report uses the configuration collection, which is not collected/uploaded as frequently as a regular metric. The datafiles are only collected in the configuration, so you could get the tablespace info much more frequently, but not all the datafiles. Collecting all datafiles every 2-4 hrs would likely kill the target db's as well as increase size of EM repository quickly!
Do you really need the datafile info instantly? If you had a summary report that came out every morning for database space used, tablespace stats, and then maybe a weekly datafile report, would that suffice? Unless your database is growing rapidly, I would guess that there' wont' be a huge change in the datafile info day to day.
The key is, can your database still extend.
Start by looking at the data that's available in these views:
select * from mgmt$db_datafiles;
select * from mgmt$db_tablespaces;
Hello Courtney Llamas,
Thank you again you take your time to help me out. So, i see that it doesent make sense to run all 2-3 houres a new data update, but the frequency is not the problem, its the timing.
i need the reports every morning, when i start working, first thing i do is to make me an overview over all databases. especially the tablespaces, (and there are hundreds of tablespaces to check) because it's a common problem we run in from times to times that they are getting full and cant grow anymore. so we get alarmed by someone in the middle of the night because production stands still in worst cases (its a 24 h shift factory). We have metric alert configured, but you know murphys law you just cant be 100% sure that the alerts are working in every case. at the moment we have already some reports showing the tablespace status, but when they are getting full, i need to check by myself if they are still able to grow, this takes me about 1-2 houres every morning. so a report that includes datafile status would make my life much easier .
so but back to the problem, as i said, the frequency of the dataupdate is not my problem, as i see in the views, all informations are updated within 24 h. but at the wrong time, so my question at this point is, can i configure the time when this data is pulled in the configuration collection?
Unfortunately, there really is no good way to "schedule" a collection time for this as it is collected with a dozen other data points and will likely cause problems. What you could do, is create a custom metric extension, with the datafile query you need (keep it simple don't go for more than what you really need to prevent performance issues), deploy the metric to all target databases, then disable the collection from the gui (metric and collection settings), and cron the collection to run at a specific period using emctl control agent runCollection <target>:<target type> <metric>. Then you can build a report off the data that is collected in mgmt$metric_current.
Okey, i will give that a try thank you!
so i did it like this! works perfectly fine! thank you a lot!