From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0rc2) Gecko/20020510 BuildID: 2002051013 We use a cookie as a key to information on our service. When the cookie is lost, autentication is also lost, along with other info to make the experience more user-friendly. We use php-skripts to serve locked images, and it is when using pages that contains these that the cookie is lost. My guess is that this build (RC1 also had the problem) dont always send the cookie information. <img src="somescript.php?bid=123412"> is my guess. The pre prerealease builds where nearly 100% on www.eurofoto.co.uk. Reproducible: Always Steps to Reproduce: 1. http://www.eurofoto.co.uk/login.php 2. Check for auth (text in the blue field on top changes to you are signed in as, and also on the blue field to the left). 3. http://www.eurofoto.co.uk/public_albums.php 4. Select an album 5. Reload original page, and you are no longer signed in wich means that the cookie is lost. To reproduce even better, upload a few images, and the browse one of your own images. Actual Results: Cookie is lost (to be more precise, a new cookie is set), user is signed of. Expected Results: Always send the cookie to the server. This bug prevents me to test our pages on Mozilla. Its also bad since I cannot recomend this browser to our customers. Which leaves the only browser to IE, and thats BAD.
Can you send a working Username/password to the developer or add it to this bug ?
Working user/password: username: firstname.lastname@example.org password: testuser Shortcut: http://www.eurofoto.co.uk/albumlist.php
Is your cookie-acceptance pref set to originating-server-only? If so, try setting it to accept-all and see if that makes a difference.
Although the symptoms are different, the patch just posted to bug 137079 will probably fix this bug as well. Can someone who can reproduce this bug please apply that patch and see if it takes care of this problem.
Being marked fixed since the patch for 137079 is checked in on trunk. However it is not yet fixed on the branch.
yes the patch was checked into the branch according to a comment in bug# 137079 verified on the branch - 05/23/02 builds, winNT4, linux rh6, mac osX