Health Check version 12 vs 13

Toad for Oracle
21

Hi John,

I found it. In the output window on the server in the HealthCheck i see those 2 tablespaces (ignore extent on) and when Toad is sending the message to me in Outlook there is no information off those 2 tablespaces. Something is going wrong by sending the output.
I'll investigate it next week, or you have the solution.

Gr. Jan

22

The only thing that I can think of is - look at the health check action that is sending the email. Make sure that it is set to "Ignoring Autoextend". If that's the setting that you want.

With "Ignoring Autoextend"

  • Toad will report tablespaces as almost full, even if the tablespaces have autoextend enabled and tablespace will just grow when it runs out of space.

With "Observing Autoextend"

  • If a tablespace is almost full, but tablespace has autoextend enabled and there is still room for tablespace to grow before max size, then Toad will not include it in the report.
  • If a tablespace is almost full, and tablespace has autoextend enabled and, but max size is reached, then Toad will include it in the report.
  • If a tablespace is almost full, and tablespace has autoextend disabled, then Toad will include it in the report.

Toad 12.1 did not have the Ignoring/Observing option (it behaved as "Ignoring autoextend"), so I think this is the difference that we've been looking for.

23

Hi John,

I found it. First, I check de prod databases and one database has 2 tablespaces who are less than 10%. With ignore option. Now when i starting the check on the left window in Automation Designer

image
image721×314 20.8 KB
and run the check there is now output on those 2 tablespaces from PGISGEO. Also when i open the app in the right windows

image
image774×502 39.3 KB
i've got the same info, now 2 tablespaces info with less then 10%. Also the HTML output file as parameter set to healtcheckPROD.html and looking into it, the same info as i get in the mail. But............ when i open the report output and let it be run, then ....

image

image
image892×485 38.5 KB

and the saved HTML file also containd the wright info

image

So thats strange, isn't it?

So i hope i told you enough about this and give you some "home" work.

Gr. Jan

24

Thanks Jan.

It looks like what is happening is that when the health check runs without the GUI, it is not running the health checks on the differrent databases. It appears to be running them all against the same database.

I'll have this fixed soon.

25

Try it in the beta just released. It should be fixed.

26

Hi John,

Installed the BETA and it seem good now. I'll try it several days and looking if i see any differences between our older version and this beta version. I'll keep you informed.

Gr. Jan

1 Like
27

Hi John,

So far so good. We saw only that sometimes the task isn't run because a password isn't correct. In the execution log you can see the error. But we are missing on which database this is wrong. When running it manuay to see which database it is, it's just running fine. Strange. Our solution is to remove all the database from the check and add them again.

image
image1318×310 29.2 KB

Is it possible to add the info from which database this is wrong, to connect to the database. I think i saw it once a time ago, that the password wasn't correct or the database wasn't running. And all this in the screen by running it manualy.

Gr. Jan

28

It would be nice to identify that in the log. I'll take a look.