I'm running phplib 7.4rc1 with PHP-4.1.1 and Oracle 8.1.7 on Linux (--with-oci8), and have major problems with OCIPlogon() which refuses to clean up the sessions in the database after calling OCILogoff(). This again caused the 'number of open sessions' to hit the roof quite fast.
I know, this is not a phplib problem and I got around the problem by using OCINlogon() instead in the .../php/db_oci8.inc file.
My question is, however, if the usage of OCIPlogon was done intentionally, and if by only changing this to OCINlogon, I have opened up for some other trouble?
I can live with the performance penalty of multiple reconnections to the database, if the alternative is a 'deadlock'.
Cheers! :-)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm running phplib 7.4rc1 with PHP-4.1.1 and Oracle 8.1.7 on Linux (--with-oci8), and have major problems with OCIPlogon() which refuses to clean up the sessions in the database after calling OCILogoff(). This again caused the 'number of open sessions' to hit the roof quite fast.
I know, this is not a phplib problem and I got around the problem by using OCINlogon() instead in the .../php/db_oci8.inc file.
My question is, however, if the usage of OCIPlogon was done intentionally, and if by only changing this to OCINlogon, I have opened up for some other trouble?
I can live with the performance penalty of multiple reconnections to the database, if the alternative is a 'deadlock'.
Cheers! :-)