>
> at the moment I'm working on the session management. It works
> very well and I like the idea from the udat hash. But I've
> noticed some strange behaviors. Everytime I'm checking if a
> user is logged in the system creates a new session id in the
> dat
Hoi list,
at the moment I'm working on the session management. It works very well
and I like the idea from the udat hash. But I've noticed some strange
behaviors. Everytime I'm checking if a user is logged in the system
creates a new session id in the database and I muss delete
, 17 January 2006 4:01 PM
> To: [email protected]
> Subject: Embperl::Object with session management?
>
> Hi all,
> I'm new to the list, but have been using Embperl for about a year and
> absolutely love it. I've just started using Embperl::Object and
> disc
Hi all,
I'm new to the list, but have been using Embperl for about a year and
absolutely love it. I've just started using Embperl::Object and
discovered a subtle issue with session management in a project with
subdirectories.
I have a project in a directory called "/var/www/cms
At Fri, 28 May 2004 15:51:43 -0400, Farouk Khawaja wrote:
> OK Folks, I realize that sessions management is 'magic' but I really
> want to know how it works.
> | id| varchar(32) | | PRI | | |
> | a_session | text| YES | | NULL| |
> | givenName | var
Title: Session Management Magic
> When I load IN.html and then OUT.html a second time, I get the
value> for 'surName', but I'm not sure where it's coming from? Can
anyone> explain?
Not sure if I understand right what your problem
is? Is it that surname is lo
Title: Session Management Magic
OK Folks, I realize that sessions management is 'magic' but I really want to know how it works. First some background
I've got session working with mySQL. I can see the cookie in my browser and it matches up with a record in the sessions t
On Thu, Aug 22, 2002 at 11:53:32AM +0200, Gerald Richter - ecos gmbh wrote:
> Apache::Session uses a hash function to select a semaphore out of a array.
> It may cause the the deadlock if both %udat and %mdat use the same one. If I
> have it right in mind per default we have 32 semaphores, to whic
>> [+ $mdat{cnt}++ +]
>>
>> Does it work or do you get the same problems?
>Ah, ok, that works. But the problem returns as soon as I have
>both udat and mdat in the page. Maybe it is Apache::Session or
>Apache::SessionX getting confused about having two session id's for
>one page and only one semap
On Thu, Aug 22, 2002 at 06:52:45AM +0200, Gerald Richter wrote:
> Strange, %udat and %mdat are handled excatly the same, they just get a
> different session id.
>
> If you make a very simple page which just contains
>
> [+ $mdat{cnt}++ +]
>
> Does it work or do you get the same problems?
Ah, o
>>
>> Seems like a problem with semaphores inside of Apache::Session. You may
try
>> to remove the semaphore with ipcrm so it gets recreated (you can view it
>> with ipcs).
>
>Well, exactly that seems to be the case. As soon as I kill the
>semaphore with ipcrm the page is displayed correctly, even
On Tue, Aug 06, 2002 at 06:36:02PM +0200, Gerald Richter wrote:
> > I have Apache::SessionX installed and
> > Embperl_Session_Config "File Semaphore"
>
> You don't need this when you have made it the default during
> Apache::SessionX Makefile.PL run
>
> > as the only session configuration option
: Saturday, July 27, 2002 3:39 AM
> To: Gerald Richter
> Cc: [EMAIL PROTECTED]
> Subject: Re: Cookieless Session Management
>
>
> Here is what I have found... I went back and ran make test on
> Embperl.
> Most, but not all, of the mod_perl tests fail with the
> message
offline = 0
ERRORS detected! NOT all test have been passed successfully
It still remains that cookie based session management does seem to work
as it should, but cookieless sessions do not. Whatever the case may be,
they both behave differently to be sure. I can watch session
information bein
an Conley; [EMAIL PROTECTED]
> Subject: Re: Cookieless Session Management
>
>
> Sean and Justin,
>
>
>
> > I've done:
> >
> > EMBPERL_SESSION_MODE 2
> >
> > Do I need to do something special other
> > than that?
> >
>
;Sean Conley" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Friday, July 26, 2002 9:41 PM
Subject: Cookieless Session Management
> I am using Embperl 2.0b8, and am having problems using cookieless
> session management, achieved by setting PerlSetEnv EMBPERL_SESSION_MODE
> -Original Message-
> From: Gerald Richter [mailto:[EMAIL PROTECTED]]
> Sent: Friday, July 26, 2002 5:24 PM
> To: Sean Conley; [EMAIL PROTECTED]
> Subject: Re: Cookieless Session Management
>
>
> Basicly you do the right thing and it should work. The only
>
riday, July 26, 2002 7:13 PM
> > To: Justin Harrison; Sean Conley; [EMAIL PROTECTED]
> > Subject: Re: Cookieless Session Management
> >
> >
> > Sean and Justin,
> >
> >
> >
> > > I've done:
> > >
> > > EMBPERL_SESSION_M
gt;
> PerlModule Embperl
>
> without that Embperl will not look at the configuration directives you set
> with PerlSetEnv.
>
> Gerald
>
> - Original Message -
> From: "Sean Conley" <[EMAIL PROTECTED]>
> To: <[EMAIL PROTECTED]>
> Sent:
Sean and Justin,
> I've done:
>
> EMBPERL_SESSION_MODE 2
>
> Do I need to do something special other
> than that?
>
Session management must be working at all. That means Apache::Session and
Apache::SessionX should be installed. If cookie based session management i
>
> How will the URL-based session-ids look like? IIRC there are three
> possibilities: path based (as suggested here), query string based (as
> usual) or hostname based (using wild card DNS records).
>
Embperl 1.3.4 will support path based and query string based session ids.
It's up to you which
> > It will even be possible that Embperl checks if cookies are
> > supported and use URL based id only when cookies doesn't work.
>
> how do you intend to do this?
>
The easiest way (I know) to check this, without haveing to modify any page,
is to issue an external redirect which contains a cook
Hi!
On Thu, Oct 11, 2001 at 08:01:58PM +0200, Gerald Richter wrote:
> > > Recently, I have noticed several threads of discussion around
> > > session management and I want to present an alternate mechanism
> > > for managing a user's session. I have built a s
Angus Lees wrote:
>
> On Thu, Oct 11, 2001 at 08:01:58PM +0200, Gerald Richter wrote:
> > It will even be possible that Embperl checks if cookies are
> > supported and use URL based id only when cookies doesn't work.
>
> how do you intend to do this?
Its fairly simple to check if cookies are su
On Thu, Oct 11, 2001 at 08:01:58PM +0200, Gerald Richter wrote:
> It will even be possible that Embperl checks if cookies are
> supported and use URL based id only when cookies doesn't work.
how do you intend to do this?
--
- Gus
---
ckley [mailto:[EMAIL PROTECTED]]
Sent: Thursday, October 11, 2001 11:12 AM
To: Eric Lowry
Cc: '[EMAIL PROTECTED]'
Subject: Re: Session Management: An Alternate Approach
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
At 12:50 PM -0700 10/10/01, Eric Lowry wrote:
>Hello fellow Embperl users!
> > Recently, I have noticed several threads of discussion around session
> >management and I want to present an alternate mechanism for managing a
> >user's session. I have built a simple alternative to using cookies or
URL
> >re-writing that seems SO simple that
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
At 12:50 PM -0700 10/10/01, Eric Lowry wrote:
>Hello fellow Embperl users!
>
> Recently, I have noticed several threads of discussion around session
>management and I want to present an alternate mechanism for managing a
>user
Hello fellow Embperl users!
Recently, I have noticed several threads of discussion around session
management and I want to present an alternate mechanism for managing a
user's session. I have built a simple alternative to using cookies or URL
re-writing that seems SO simple that there
> Oh my god. I implemented my own sessin management as a new package
> Embperl::Session. So I will have to rewrite it to Apache::SessionX? Are
> there many differences?
>
The API is the same, so you will not have to change anything. At least not
to use the current functionality. You may need to s
>Apache::SessionX is an extented version of Apache::Session (and still
>requires Apache::Session to work). So the X simply stands for eXtened. It
>replaces HTML::Embperl::Session in the current Embperl version. The reason
>for the name change from HTML::Embperl::Session to Apache::SessionX is,
tha
> >Apache::SessionX is available for testing. That fixes some seesion
problems
> >and should ease configuration. The cookie-less session management is
still
> >in a state, that I have to work on first ...
> Is it`s name final? Why 'X'? Maybe you co
>Apache::SessionX is available for testing. That fixes some seesion problems
>and should ease configuration. The cookie-less session management is still
>in a state, that I have to work on first ...
Is it`s name final? Why 'X'? Maybe you could think about some othe
ms
and should ease configuration. The cookie-less session management is still
in a state, that I have to work on first ...
Gerald
-
Gerald Richterecos electronic communication services gmbh
Internetconnect * Webserver/-design
"Gerald Richter" <[EMAIL PROTECTED]> writes:
> > My idea is to make all mysite links relative and just
> > add session-id to the begining. When the link is absolute,
> > I will not add session-id.
> >
> > Is it possible to do this with embperl?
> >
>
> Embperl 1.3.4 will support this, it's under
David Hajek wrote:
>
> Hello,
>
> I would like to use some sort of session management on my
> site. I do not like cookie based solution, since not all
> clients support cookies. I like the URL rewriting (mangling)
> technique. So I'd like to rewrite each url in m
>
> I would like to use some sort of session management on my
> site. I do not like cookie based solution, since not all
> clients support cookies. I like the URL rewriting (mangling)
> technique. So I'd like to rewrite each url in my site to
> something like:
>
>
Hello,
I would like to use some sort of session management on my
site. I do not like cookie based solution, since not all
clients support cookies. I like the URL rewriting (mangling)
technique. So I'd like to rewrite each url in my site to
something like:
http://mysite.com/session-id/so
>[21384]Reading /usr/local/apache/htdocs/iportal/test1.html as input using
PerlIO ...
>[21384]SES: Received Cookie ID: 49c784ae7385ebaf7abde4a203f3bac8 New
Cookie ID: 43903564a7b36f84b785a0882afb3974 Session data is NOT modified
The session id is received by test1.html, but it generate a new
Hi again,
Here are the logs concerning session handling, maybe the are of some help,
they seem ok to me, and the cookies set are actually stored to Postgres.
The test.html file simple sets [- $udat{say} = "something" -]
and test1.html should write it back [+ $udat{say} +], but it doesnt..
Than
Hi everybody,
Afther a few tries installing Embperl with Postgres as its session
management
database I came out with the following configuration directives:
PerlSetEnv EMBPERL_SESSION_CLASSES "Postgres Semaphore Base64"
PerlSetEnv EMBPERL_SESSION_ARGS
"DataSource=dbi:Pg(AutoCo
Hi,
you could add more variables (or better to say hash keys). There maybe
another problem with your code.
If you are using Embperl, it's easier to use Embperl buildin session
management (which also uses Apache::Session), but provides the setup for
you. If you don't use Embperl, you
Mod Perl has nothing to do with Session Management.
Cheers
Ruben
> Hello there...
>
> I have a doubt regarding the session management in perl.
> The entries that I saw in the mailing list were speaking of mod_perl
> which I did
Hello there...
I have a doubt regarding the session management in perl.
The entries that I saw in the mailing list were speaking of mod_perl
which I did not understand.
I am not using mod_perl. I use the Apache::Session module and tie a hash to
it and add session variables to the hash as soon
>
> Yes, but only when I restart apache using apachectl stop/start not
> with graceful, is this normal?
>
Yes
Gerald
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
Hi Gerald,
"Gerald Richter" <[EMAIL PROTECTED]> writes:
> When Apache starts, you should see a message "Embperl session management
> enabled", did you?
Yes, but only when I restart apache using apachectl stop/start not
with graceful, is this normal?
C
Hi Jon,
"Jon Brisbin" <[EMAIL PROTECTED]> writes:
> Here's what I did: I switched to using file-based session storage in a tmp
> directory under /cgi-bin, which does not allow a malicious user, even if
> they knew a session id, to view any of the session information. Here's what
> I put in my
Eric Knauel wrote:
> Hi,
>
> I have some trouble getting session management to work. I'm using
> Apache Session 1.03, Embperl 1.3b4 with mod_perl on Apache 1.3.12,
> here what I put in the httpd.conf:
>
> SetEnv EMBPERL_DEBUG 2375713
> SetEnv EMBPERL_VIRTLO
Hi,
>
> I have some trouble getting session management to work. I'm using
> Apache Session 1.03, Embperl 1.3b4 with mod_perl on Apache 1.3.12,
> here what I put in the httpd.conf:
>
> SetEnv EMBPERL_DEBUG 2375713
> SetEnv EMBPERL_VIRTLOG /embperl/log
> Per
Jon Brisbin
www.jbrisbin.net
[EMAIL PROTECTED]
- Original Message -
From: Eric Knauel <[EMAIL PROTECTED]>
To: Embeded PERL ML <[EMAIL PROTECTED]>
Sent: Wednesday, August 02, 2000 8:24 AM
Subject: Problems with session management
>
> Hi,
>
> I have some trouble getting s
Hi,
I have some trouble getting session management to work. I'm using
Apache Session 1.03, Embperl 1.3b4 with mod_perl on Apache 1.3.12,
here what I put in the httpd.conf:
SetEnv EMBPERL_DEBUG 2375713
SetEnv EMBPERL_VIRTLOG /embperl/log
PerlSetEnv EMBPERL_COOKIE_D
>
> PerlSetEnv EMBPERL_SESSION_CLASSES "DBIStore SysVSemaphoreLocker"
> PerlSetEnv EMBPERL_SESSION_ARGS "object_store=DBIStore
> DataSource=DBI:mysql:session"
>
You don't need the "object_store=DBIStore" in SESSION_ARGS, because it's
already in SESSION_CLASSES, but it doesn't hurt here.
> [-
>
Hi everybody,
We plan to develop a new site using Embperl.
Our current installed program :
Apache: 1.3.12 mysql: 3.22.32
Embperl: 1.3b2 Apache::Session: 1.03
etc.
--
Hi,
>
> >I suggest tring DBIFileStore first because it less complicated. If this
> >works then we can try again DBIStore
>
> FileStore just works. :)
>
> Oracle wasn't committing updates, so DBIStore wasn't getting a look in.
> AutoCommit was on, but obviously not effective.
>
> I'll be discussing
Hi,
>I suggest tring DBIFileStore first because it less complicated. If this
>works then we can try again DBIStore
FileStore just works. :)
Oracle wasn't committing updates, so DBIStore wasn't getting a look in.
AutoCommit was on, but obviously not effective.
I'll be discussing this with our
Hi,
>Does it really create a new record for each update? It should only create a
>new record for a new user. So as long as you just testing with on browser,
>there should only be one record.
I am using one browser for testing. How about this for an idea, the DB is not
comitting? Maybe the multi
>
> I've checked the database, and the cookie info gets written to
> the DB. I notice
> it creates a new record each time a change occurs.
Does it really create a new record for each update? It should only create a
new record for a new user. So as long as you just testing with on browser,
there s
Hi,
>I don't have much ideas now anymore. Everythings seems to be really good.
>The only thing I can imagine, is that you see different session. Are you
>really sure that the session id you see in the log is the same for every
>page?
>
>The other possibility may be that the data goes never to the
> >That look very good. If the %udat is tied to the right object
> and the cookie
> >is the same, then I don't see any reason why is shouldn't work.
> Have tried
> >to print out all elements of %udat? Are there other values in it?
>
> [9515]SRC: Line 2: [$foreach $blah (keys(%udat)) $]
> [9515]CMD
>That look very good. If the %udat is tied to the right object and the cookie
>is the same, then I don't see any reason why is shouldn't work. Have tried
>to print out all elements of %udat? Are there other values in it?
[9515]SRC: Line 2: [$foreach $blah (keys(%udat)) $]
[9515]CMD: Cmd = 'forea
et the values,
are they still there?
> That debugging stuff is very useful.
>
> >P.S. Please send answers also to the list, so other people can
> participate
>
> Just trying to keep my ignorance less widely known :)
>
Oh, there are more people who have problems with session
Hi,
>Do you run under mod_perl and have build mod_perl as DSO? If yes try either
>linking mod_perl staticly into Apaache or upgrade to mod_perl 1.22. If this
>doesn't help, include a [+ ref (tied(%udat)) +] in your pages, what does it
>show?
I have mod_perl staticly linked. This is what happens
Hi,
>
> HTTP_COOKIE has the same value on each page, just as I would expect.
> The $udat{'username'} is undefined at each new page, even though
> it is given a
> value several times.
>
> The database currently has 8 records in the sessions table using
> the HTTP_COOKIE
> value as the id. They appe
Hello,
>
> I have ran into a problem with setting up embperl and mod_perl.
>
> I am using Session Tracking, and mod_perl.
>
> The problem is : when I have :
> in the httpd.conf
> PerlModule HTML::Embperl
>
> I don't session tracking doesn't work and MBPERL_UID cookie doesn't get
> set on my
Hello,
I have ran into a problem with setting up embperl and mod_perl.
I am using Session Tracking, and mod_perl.
The problem is : when I have :
in the httpd.conf
PerlModule HTML::Embperl
I don't session tracking doesn't work and MBPERL_UID cookie doesn't get
set on my browser.
when I
i use the redirect via location with embperl and %udat
everyday without a problem. something else must be wrong
with your configuration.
cliff rayman
genwax.com
indrek siitan wrote:
> Hi,
>
> > But my problem of the other page where the session management
> > just doesn'
Hi,
> But my problem of the other page where the session management
> just doesn't work at all is still with me. I put data into
> $udat{'username'} and then go to another page with
> $http_headers_out{'Location'}. When I get there I check to see
> if $
Hi,
>
>
> But my problem of the other page where the session management
> just doesn't work
> at all is still with me. I put data into $udat{'username'} and
> then go to
> another page with $http_headers_out{'Location'}. When I get there
> I c
>
> > > What should I set the EMBPERL_COOKIE_PATH to?
> > > Any recommended path?
> > > How does this effect the cookie versus the default?
> > > What doc/code can I read to help me understand what is going on?
> > >
> > > Aaron Johnso
is effect the cookie versus the default?
> > What doc/code can I read to help me understand what is going on?
> >
> > Aaron Johnson
> >
> > Gerald Richter wrote:
> >
> > > Hi,
> > > >
> > > > I am having trouble with session manage
cookie versus the default?
> What doc/code can I read to help me understand what is going on?
>
> Aaron Johnson
>
> Gerald Richter wrote:
>
> > Hi,
> > >
> > > I am having trouble with session management. It seems to be
> > > inconsistent on my
Gerald,
What should I set the EMBPERL_COOKIE_PATH to?
Any recommended path?
How does this effect the cookie versus the default?
What doc/code can I read to help me understand what is going on?
Aaron Johnson
Gerald Richter wrote:
> Hi,
> >
> > I am having trouble with session
Hi,
>
> I am having trouble with session management. It seems to be
> inconsistent on my
> system. Sometimes it works, and sometimes it doesn't.
>
> I am using Oracle 8i on RH6.0 with Apache/1.3.12 (Unix) mod_perl/1.21.
> Apache is set up with DBIStore SysVSemaphoreLo
Hi,
I am having trouble with session management. It seems to be inconsistent on my
system. Sometimes it works, and sometimes it doesn't.
I am using Oracle 8i on RH6.0 with Apache/1.3.12 (Unix) mod_perl/1.21.
Apache is set up with DBIStore SysVSemaphoreLocker.
When I set $udat{'user
74 matches
Mail list logo
