Pid wrote:
...
After a cursory look through the server.xml, (cursory because of the
trauma of wading through comments), I note:
<Listener className="org.apache.jk.config.ApacheConfig"
modJk="/opt/zimbra/httpd/modules/mod_jk.so" />
The OP made reference to the jvmRoute="jvmAlfresco1", so I think we need
to understand what's going on there to find a resolution.
I saw that too, but I believe this is just a reference to nod_jk's "auto-configuration"
capability. It does not necessarily mean that the Op /is/ accessing his Tomcat's through
Apache/mod_jk. And as far as I know, I don't think it would matter anyway.
There is another simple test of which I did not think before :
- login to instance A, verify it works
- check in the browser for any JSESSIONID cookie, note the beginning of the
value
- login to instance B
- check again the cookies for JSESSIONID
If there is only one, and it has changed, then there is the collision.
I am willing to be convinced, but I am not sure by the way that the cookie path is part of
what identifies a separate cookie.
In other words, can the browser really hold two distinct cookies where the cookie name and
hostname are the same, and only the path differs ?
---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org
