A pl sql package is taking 3 hours to compile in a 10g database. There is no library cache load/pin lock waits as so one is working in the database while compiling. The wait event seen is db_file_sequential_read. Total number of line of code is around 4K and mainly contains lots of update statements on a table with 150 columns. The statement written is trying to update each field of the table with value based on calculation and this construct is repeated 8 to 10 times.If some of the update statements are removed then speed of the compilation increases (infact if all the update statements are removed then it is compiled under less than 3 secs.). I tried to find out what is running behind by quering v$sqlarea with the hash_value from v$session but there is no entry in that view.
The cache related parameters in the database are given below :-
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
Either somebody is executing the package or your session died.
Hi PAVB
Neither the session has died nor somebody is executing the package as i have already mentioned that nobody is working in the database when I am compiling the package
Disclaimer: Advice is provided to the best of my knowledge but no implicit or explicit warranties are provided. Since the advisor explicitly encourages testing any and all suggestions on a test non-production environment advisor should not held liable or responsible for any actions taken based on the given advice.
Bookmarks