I have partially fixed the problem, i managed to connect to database oracle 11g and some other 12.
The datadase to which i could not connect is a db where the upgrade from oracle 11 to 12 has had some problem.
The strange thing is that with another pc with the same version of toad but with os windows 7 the connection is successful
Recently downloaded the trial version of TfO 12.10.0.30. Up to then I was using an older version (9.6 Commercial), which works rather good, though I hate the debugger for not showing variables in outer scopes or being able to evaluate expressions or accessing PL/SQL tables and other collections like "dbms_sql.varchar2_table" and the objects stored therein. Why can't it evaluate something like "data_map_table(5).EXISTS('key')" ? The debugger in 12.10.0.30 may be better than that in 9.6, that's for sure, but with a memory footprint of 2.5 GB (see below) ? Anyway, the company had me evaluate the newest TfO version, which is 12.10.0.30.
So, here are the problems I had running TOAD 12.10.0.30 on Windows 7 Professional on x64, with Oracle 11.2 client against 11.2 server; saw no difference when running it on 11.2 client against 12c server)
The second and third issues are significant and prevent my company from buying a new TOAD version. We could take the giant memory footprint, but this "hanging" think made us loose work on some occasions, not much, since you can access the safety copies of the editors in the User Files folder to reconstruct the changes which were lost by killing the process, but it's ANNOYING. You must save edits each time before you hit F9 or F5, but sometimes you are too fast hitting F9 and then it's too late to save.
Has anyone encountered similar problems or is there something which can I could do to get a more stable (and not so hungry) version ? Of course, I do not think, a new version would take only 380 KB memory like Toad 9.6, but 2.500 KB ist way too much,
rgds, Werner
Phyllis' suggestion is the way to go, but this appears to be a bug in 12.1. The latest version of Toad (12.10) does not have this problem.
Michael
Thanks for reponding!
As far as the edit/test-cycle is concerned - I also have only one connection open, to a schema consisting of about 2500 objects (tables, views, packages, procedures... the whole bunch), so the schema IS rather big (about 1.5 GB);
I used only one editor with three tabs, package header, body and one for the test case. I use Code Insight frequently with cached results (will now try and switch it off). I use an instance of Toad 9.6 running in parallel on my 2nd monitor to check results.
I'll report the results.
[quote user="Michael Staszewski"]
> Minor glitch: Open editor window and enter "CREATE OR REPLACE\ NPROCEDURE ...
In Options on the Execute/Compile page disable the "Treat blank line as statement terminator " option. This option has a couple of minor bugs including this one.
> But this is not good: Working on a >300 KB package, i.e. loading the package source and enter the edit, compile, test-cycle...
What does your edit/test-cycle consist of? Also, are you connecting to hugeschema like APPS on Oracle E-Business Suite and/or using Code Insight? If so try disabling the "Cache Code Insight results" option on the Editor|Code Assist page in Options. Do things improve? I created a 300KB procedure that has nothing more than a ton of select statements within. Editing, compiling, and executing it hung around 290 MB for me. I just have a single connection open with a single Editor and one tab for the procedure.
Under what context are you comparing the two versions? Are you invoking Code Insight with SELECT or other DML as standalone statements or are they located within PL/SQL? If located within PL/SQL please test it in standalone statements outside of PL/SQL and let me know if the two are comparable. There is much more internal work done behind the scenes in 12.10, but that work *should* be minimal with only those 4 included items you have set. Other items like "PL/SQL Varables/Parameters" and "Join Suggestions" will significantly increase the amount of internal work.
Michael
For Standalone Select and DML Stmts.
I'm using Toad for Oracle EBS, Previous version 12.1 is much quicker in displaying the table list and its associated columns, but in 12.10 there is a delay in the drop down list to show the tables and columns. I exported Application Data directory folder from 12.1 and imported to 12.10 to keep my Toad options as is, but still the results are same.
In 12.10
under toad insight objects; I selected (synonyms,tables,users and views) only.
Under Toad insight group box ; I selected Enable Code Insight Pick list, Cache code insight results, Show object types as text in pick list and delay popups as 50 milliseconds.
I heavily rely on toad for my day to day development due to slow code insight response I had to switch back to my older version. Any help is highly appreciated.
Thank you!
Hey Mark,
No, you're not missing anything in your configuration. Team Coding doesn't currently support Database Links, Materialized Views, and Jobs. It only supports the object types displayed in the Team Project settings of the configuration window. If you would like, feel free to add an entry to the Idea Pond with the features you'd like, and we can take a look at what might be involved in implementing them.
For your second issue, I would probably recommend looking into Toad Intelligence Central (TIC) Community Edition. If the 3rd party database is Oracle, but doesn't allow you to create custom schemas or objects other than triggers, you can store the Team Coding configuration within Toad Intelligence Central. Then, any time you connect to that 3rd party database (and you have a connection to the TIC configured), you can still do change control on the database since all the history is stored on the TIC server.
Let me know if that helps to solve what you need. If you have any questions, feel free to shoot me an e-mail offline and I'll see what I can do to help.
Thanks!
-John
I'm starting to use Team Coding in Toad 12.10 with an Oracle 12.1 database and TFS 2012. I've got it working for tables, indexes, constraints, etc. What I would like is to include the Database Links, Materialized Views, Jobs, and other things, but they don't seem available when I select what objects to put under control. Why is that? Am I missing something in my configuration?
I'd also like to put a set of objects from a different database under control because they are all related to the same software application, but I manage one database and don't have management access to the other (it's a 3rd party database that allowed us to create a handful of triggers) I was hoping I could add a "folder" and just dump the trigger scripts there, but I may have to go old school and simply use Visual Studio to manage it. Any ideas?
Thanks
Hi all,
Toad 12.11.0.40 (but last prod version) x64.
There is a small bug in script output.
Script is:
clear breaks
break on report
clear computes
compute sum of tot_blocks on report
compute sum of tot_rows on report
compute AVG of avg_rows on report
column avg_rows format 990.90
select
twentieth,
min(rows_per_block) min_rows,
max(rows_per_block) max_rows,
sum(block_ct) tot_blocks,
sum(row_total) tot_rows,
round(sum(row_total)/sum(block_ct),2) avg_rows
from
(
select
ntile(nvl(&&3,20)) over (order by rows_per_block) twentieth,
rows_per_block,
count(*) block_ct,
rows_per_block * count(*) row_total
from
(
select
fno, bno, count(*) rows_per_block
from
(
select
dbms_rowid.rowid_relative_fno(rowid) as fno,
dbms_rOwId.rowid_block_number(rowid) as bno
from
&&1..&&2
)
group by
fno, bno
)
group by
rows_per_block
order by
rows_per_block
)
group by
twentieth
order by
twentieth
;
Where last few lines from output in toad are:
3 15 21 1572 27081 17.23
4 22 28 223 5272 23.64
5 29 36 19 582 30.63
--------
avg 17.23
sum 22672 169615Elapsed: 00:00:04.85
Brg,
Damir
This exists in the beta but not 12.10.
In SB-Users, there is a RHS tab called "Container Data Restrictions"
Here are the newest results:
Switching off the Code Insight cache helped a little, but not for long. After some test cycles with my 500 KB package the "Virtual Size" (VS, measured with Sysinternals "Process Explorer") of Toad.exe sky-rockets to 2.3 GB, though I have only one connection and one editor with three tabs. With Toad 9.6 I worked on the same project, and the VS of this EXE never got over 450 MB. But 12.10 has far better Code Insight support, a more stable debugger and more improvements, so I really would prefer the new version.
I watched the VS of Toad 12.10 develop thru my work session, and it went like this:
- Starting Toad 12.10, with no DB connection open, the VS is about 500 MB
- After opening the connection, editor and loading the package source, the VS was about 700 MB, after doing some tests it slowly but surely climbed to 800 and little later to 1 GB.
- With time the VS steadily increases until it reaches 2+ GB, even 2.5 GB. Then, it eventually would freeze. But I have to admit, that there were two Oracle instances (11.2 and 12c) running on my computer (16 GB RAM, pagefile is now 17 GB!), and a SQL Server 2012 instance, all three together consuming about 35 GB Memory (!). This really was too much, so I shut down the Oracle 12c and the SQL Server instance, which I do not need for now (these alone occupied 30 GB VS !), and the freezing of Toad got significally better, at least it felt this way.
- After I rewrote the package and outsourced functional units into their own packages, I ended up with 4 packages (sizes 300, 130, 15 and 10 MB). At the moment, I see a VS of 1.6 GB, which is much, but not yet too much, I hope.
I feel that Code Insight is responsible for big parts of the VS, I tried and switched off all features I do not need. Especially global synonyms seem to play a role, but I'd like to have parameter support for SYS-packages like DBMS_LOB, UTL_FILE and some more (not many, there are 4 or 5, the rest just consumes memory). It would be fine, if you could specify the synonyms to load not in a global manner (all or nothing, though you can check "Do not load synonyms starting with "/""), but I'd like to configure Code Insight in this way: "Support only global synonyms starting with "DBMS_"", i.e. give a pattern for the global synonyms to support I think this could be an improvement in memory.
So it's still a special situation, Toad 12.10 has features I like very much (to name a few. refactoring, better code assistance, support of conditional pl/sql compilation, there may be many more), but for the moment it seems to take about 4 times the memory of Toad 9.6 working on the same project. I still have three weeks for evaluation, maybe someone can point out a way to keep memory consistently under 2 GB. I'll try and switch off global synonyms for code insight, the SYS-Package support is not that important.
I'll be back soon with more results ...
I just caught that you're looking at virtual memory and not the physical memory usage. Earlier my mem usage report was Toad's working set. When I enable virtual size column in process explorer I do see higher usage, but ~700MB for my 300KB test package. It fluctuates a little here and there, but nothing crazy. What you are seeing is quite a big jump. If you can narrow down steps to something in particular and zip and send me your User Files folder to michael<dot>staszewski<at>quest<dot>com I will run through them enabling advanced memory manager that will report on leaks.
Michael
We have found that Oracle now offers a 'service' called "Oracle Database Cloud Service" which is essentially a VM version of your cloud database. Without this subscription, you cannot access the database:
I'll leave this here in case anyone else has experience with this.
Toad does work with the Oracle Cloud Database Service, assuming the ports are open (something you have to do after database creation)
In Toad 12.10 and prior: Options -> Data Grids -> Visual. Click the "Edit Styles" button, then "Edit" in the next dialog. Find "Header" in the grid element list and change the font or color however you like, then click OK until you are all the way out.
This is much simpler in Toad 12.11 beta. Options -> Data Grids -> Visual. Click "Edit" and you can figure out the rest.
Is there any way how to change background color for grid header. Recently I migrated to windows 10 and header has same background color as records, i.e. white.
Thanks again for looking into ,my problem, Michael !
So what columns of process explorer's memory items shold one look at ? Working Set, Peak Working Set ? As I reported, I divided the big package into 1 main package (240 KB) and 3 support packages (130 KB, 7 KB and 15 KB). First things seemed to get better (actually, Toad seems to be more stable now, but this may be due to the fact, that I saved >30 GB of virtual memory by shutting down two of my three database instances). But now - after some heavy editing and debugging Process Explorer again shows 2.15 GB Virtual Size, 1.7 GB Working Set (WS) and 1.8 GB Peak Working Set (PWS). Of course, I now have 9 tabs open (4 * 2 for the headers an bodies of the four packages, 1 for the test cases), seemingly the refactoring had no impact on memory usage, but up to now, Toad has not frozen again,.
I'll try and find the moment, when memory increases from 700 or 800 MB VS to more than 2 GB. Then I'll send to you my zipped user folder, so you can modify the settings to activate the advanced memory manager.
[quote user="Michael Staszewski"]
I just caught that you're looking at virtual memory and not the physical memory usage. Earlier my mem usage report was Toad's working set. When I enable virtual size column inprocessexplorer I do see higher usage, but ~700MB for my 300KB test package. It fluctuates a little here and there, but nothing crazy. What you are seeing is quite a big jump. If you can narrow down steps to something in particular and zip and send me your User Files folder to michaelStaszewskiquestcom I will run through them enabling advanced memory manager that will report on leaks.
Michael
[/quote]