Closed Bug 156850 Opened 22 years ago Closed 22 years ago

Renders two images instead of one

Categories

(Core :: Networking: HTTP, defect)

defect
Not set
normal

Tracking

()

VERIFIED DUPLICATE of bug 121084

People

(Reporter: magus, Unassigned)

References

()

Details

When I load that in Mozilla, I see an initial image, then it quickly changes to
another.
confirming using build 2002071008 on Win2k. weird.
The first GET (with referred from bugzilla) received a code 200 OK, but Mozilla
immediately sends a second GET (identical but without referrer). Hence the
reason why the server send 2 images and Mozilla will display them one after another.
I'm not using any proxy, nor HTTP Pipelining.
Sev -> Blocker it seems quite strange that Mozilla needs to send a HTTP request
twice.
Assignee: pavlov → new-network-bugs
Severity: normal → blocker
Component: Image: GFX → Networking
Keywords: regression
QA Contact: tpreston → benc
To reproduce the problem:
 - Launch a sniffer,
 - Load any URL with Mozilla,
 - see 2 initial consecutive HTTP GET requests (one with referrer, the other one
without any referrer).
I'll this behaviour with Mozilla 1.0.
Status: UNCONFIRMED → NEW
Ever confirmed: true
Summary: Renders two images instead of one. → Renders two images instead of one (Mozilla sends 2 initial HTTP GET instead of 1)
Sev back to Normal. Mozilla 1.0, Netscape 4.76 and IE6 have the same behaviour.
I now even wonder if this is a bug, sorry for the false alarm.
Severity: blocker → normal
Keywords: regression
Modified Summary too, back to the original one.
Summary: Renders two images instead of one (Mozilla sends 2 initial HTTP GET instead of 1) → Renders two images instead of one
IE6 does not exhibit this behavior. This is a snippet from the apache access log.


my ip - - [11/Jul/2002:00:57:37 -0700] "GET /cgi/rndimg.cgi HTTP/1.1" 200 6225
"-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.1a) Gecko/20020611"
my ip - - [11/Jul/2002:00:57:38 -0700] "GET /cgi/rndimg.cgi HTTP/1.1" 200 5349
"-" "Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.1a) Gecko/20020611"

my ip - - [11/Jul/2002:00:57:53 -0700] "GET /cgi/rndimg.cgi HTTP/1.1" 200 5211
"-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.0.3705)"


The first two lines are from one hit by Mozilla, the third is a single hit from IE6.
Hrmm, its possible this is a win32 bug only.

I've been checking the log, and I just got a hit from someone using Mozilla 1.1a
on Linux, and it shows a single hit.

address removed - - [11/Jul/2002:01:01:02 -0700] "GET /cgi/rndimg.cgi HTTP/1.1"
200 3706 "http://bugzilla.mozilla.org/show_bug.cgi?id=156850" "Mozilla/5.0 (X11;
U; Linux i686; en-US; rv:1.1a+) Gecko/20020710"
Also happens on linux... 20020709 (..and a very long time before..)
-> OS / PLATFORM: ALL

I also noticed this while i was writing a small webserver..

Mozilla seems to do this always when you view the image.. with or withoud HTTP
Pipelining..

..But when the image is inside an HTML File, mozilla just sends one GET request
to the server...
OS: Windows XP → All
Hardware: PC → All
Another single... This one on NetBSD... weird

address removed - - [11/Jul/2002:01:23:55 -0700] "GET /cgi/rndimg.cgi HTTP/1.0"
200 3879 "http://bugzilla.mozilla.org/show_bug.cgi?id=156850" "Mozilla/5.0 (X11;
U; NetBSD i386; en-US; rv:1.0.0) Gecko/20020617"
OS: All → Windows XP
Hardware: All → PC
Collin:

Does this also happen when you use a network sniffer like ethereal?
.. i don't trust the apache in this case:

I saw 2 get requests form mozilla in ethereal, but only one appeared in the
apache log..
OS: Windows XP → All
Hardware: PC → All
No sniffer here. I suppose its possible Apache isnt logging everything.
On Mac os X (20020626) mozilla also sends two GET requests..

IE 5.12 on OSX just sends one GET request..

*** This bug has been marked as a duplicate of 121084 ***
Status: NEW → RESOLVED
Closed: 22 years ago
Resolution: --- → DUPLICATE
Component: Networking → Networking: HTTP
QA Contact: benc → tever
Verified dupe.
Status: RESOLVED → VERIFIED
QA Contact: tever → junruh
You need to log in before you can comment on or make changes to this bug.