Re: [rt-users] RT 3.8 and FastCGI out of memory via http but not https *SOLVED*

2009-02-25 Thread Ruslan Zakirov
You MUST read UPGRADING.mysql On Wed, Feb 25, 2009 at 1:33 AM, Nick Kartsioukas change+lists...@nightwind.net wrote: Changing the a_sessions table from longtext to longblob fixed it. Thanks, Steven!  Now that I know the answer I've found the other threads relating to this, but I was apparently

Re: [rt-users] RT 3.8 and FastCGI out of memory via http but not https *SOLVED*

2009-02-25 Thread change+lists . rt
On Wed, 25 Feb 2009 16:26:49 +0300, Ruslan Zakirov ruslan.zaki...@gmail.com said: You MUST read UPGRADING.mysql This was not obvious from the wiki documentation or output from make upgrade or the rt-upgrade-database script. ___

Re: [rt-users] RT 3.8 and FastCGI out of memory via http but not https *SOLVED*

2009-02-25 Thread Kevin Falcone
On Feb 25, 2009, at 11:21 AM, change+lists...@nightwind.net wrote: On Wed, 25 Feb 2009 16:26:49 +0300, Ruslan Zakirov ruslan.zaki...@gmail.com said: You MUST read UPGRADING.mysql This was not obvious from the wiki documentation or output from make upgrade or the rt-upgrade-database script.

Re: [rt-users] RT 3.8 and FastCGI out of memory via http but not https *SOLVED*

2009-02-24 Thread Nick Kartsioukas
Changing the a_sessions table from longtext to longblob fixed it. Thanks, Steven! Now that I know the answer I've found the other threads relating to this, but I was apparently using poor search terms before :) ___

Re: [rt-users] RT 3.8 and FastCGI out of memory via http but not https *SOLVED*

2009-02-24 Thread Tom Lahti
Changing the a_sessions table from longtext to longblob fixed it. Shouldn't the upgrade scripts for rt-setup-database --action upgrade take care of this? Seems weird that it doesn't. -- -- Tom Lahti BIT Statement LLC (425)251-0833 x 117

Re: [rt-users] RT 3.8 and FastCGI out of memory via http but not https *SOLVED*

2009-02-24 Thread Jesse Vincent
On Tue, Feb 24, 2009 at 03:09:44PM -0800, Tom Lahti wrote: Changing the a_sessions table from longtext to longblob fixed it. Shouldn't the upgrade scripts for rt-setup-database --action upgrade take care of this? Seems weird that it doesn't. The schema changes necessary for this major