Some of use IE6, others Firefox, others IE7. I haven't verified with
everyone, but several people I /have/ asked all maintain position. I
myself see it maintain position in both IE6 and Firefox 2.0.0.8
Some pieces of our apache config I thought were relevant:
SSLSessionCache shmcb:/var/lib/apache2/ssl_scache
NameVirtualHost *:80
NameVirtualHost *:443
Listen 443
<VirtualHost *:80>
ServerAdmin user-220aee41b4ef@xymon.invalid
DocumentRoot /path/to/root
ServerName our.webaddress.com
RedirectMatch permanent (/.*) https://our.webaddress.com //$1
</VirtualHost>
<VirtualHost *:443>
ServerAdmin user-220aee41b4ef@xymon.invalid
DocumentRoot /srv/www/htdocs
ServerName monitor.directpointe.net
SSLEngine on
SSLCertificateFile /etc/apache2/ssl.crt/monitor.crt
SSLCertificateKeyFile /etc/apache2/ssl.key/monitorssl.key
</VirtualHost>
# This file is for Apache 1.3.x and Apache 2.0.x
#
# Add this to your Apache configuration, it makes
# the Hobbit webpages and cgi-scripts available in the
# "/monitor" and "/monitor-cgi" URLs.
# NB: The "Alias" line below must NOT be used if you have
# the Hobbit webfiles as the root URL. In that case,
# you should instead set this:
#
# DocumentRoot /home/hobbit/server/www
<Directory "/srv/www/htdocs">
Options Indexes FollowSymLinks Includes MultiViews
Order allow,deny
Allow from all
SSLRequireSSL
AuthUserFile /path/to/hobbitpasswd
AuthGroupFile /path/to/hobbitgroups
AuthType Basic
AuthName "monitor"
Require user user1 user2
</Directory>
Alias /monitor/ "/home/hobbit/server/www/"
<Directory "/home/hobbit/server/www">
Options Indexes FollowSymLinks Includes MultiViews
Order allow,deny
Allow from all
SSLRequireSSL
AuthUserFile /path/to/hobbitpasswd
AuthGroupFile /path/to/hobbitgroups
AuthType Basic
AuthName "monitor"
Require user user1 user2
</Directory>
I can't see anything about caching, which I think would be more relevant
to the discussion. Then again, I'm no apache admin guru.
Tod Hansmann
Network Engineer
-----Original Message-----
From: Scott Wilson [mailto:user-999746b4662e@xymon.invalid]
Sent: Monday, October 29, 2007 5:25 PM
To: user-ae9b8668bcde@xymon.invalid
Subject: RE: [hobbit] Strange "problem" with refreshes
What web browser do you use? Since I tried it with meany sites (not all
apache) and had the same behavior, I think the "ssl pages jump to top on
refresh" is more likely to be a browser based issue.
Scott Wilson Lead System Administrator
user-999746b4662e@xymon.invalid NSIT - DCS - SeaSol
On Mon, 29 Oct 2007, Tod Hansmann wrote:
We use SSL exclusively with ours and it maintains its position. I'm
thinking this would be in your apache configs somewhere. Have you
asked
an Apache list about this?
Tod Hansmann
Network Engineer
-----Original Message-----
From: Scott Wilson [mailto:user-999746b4662e@xymon.invalid]
Sent: Friday, October 26, 2007 1:54 PM
To: user-ae9b8668bcde@xymon.invalid
Subject: Re: [hobbit] Strange "problem" with refreshes
Yup, it seems to be a fundamental property of how Firefox handles
refreshes. If you refresh a SSL page, it bounces to the top. A
non-SSL
page stays where it is.
I've tried it with other web pages, and they all function that way.
Scott Wilson Lead System Administrator
user-999746b4662e@xymon.invalid NSIT - DCS - SeaSol
On Fri, 26 Oct 2007, Josh Luthman wrote:
So you're telling us that the only variable you changed was SSL and
it
worked perfectly?
Very odd...
On 10/26/07, Scott Wilson <user-999746b4662e@xymon.invalid> wrote:
I found the answer. It's pretty simple really...
I was using SSL with my hobbit server. With SSL'ed web pages,
refreshes
force you back to the top of the page. NonSSL'ed web pages (which I
had
totally disabled) work fine.
Now I just need to hack up the header files so the secure areas
point
to
SSL even when I'm looking at the nonSSL pages. But as I already ran
into
the header files while debugging this, that should be trivial.
Thanks for the help!
Scott Wilson Lead System Administrator
user-999746b4662e@xymon.invalid NSIT - DCS - SeaSol
On Fri, 26 Oct 2007, Josh Luthman wrote:
I am using this for an Apache server:
Server version: Apache/2.2.3
Server built: Jun 26 2007 19:26:32
Latest and greatest package from the default repo's. Using CentOS
5
(which
is RHEL based).
I honestly have no idea how the page stays position. It has to do
something
along the lines of cache, but that is a client side thing so...my
ideas
stop
here =(
Please let us know what you come up with. I know I would like to
know
the
resolution to this.
Josh
On 10/26/07, Scott Wilson <user-999746b4662e@xymon.invalid> wrote:
Brilliant thought....
The demo does maintain position exactly as I want. So it
definitely
isn't
hobbit in general. My best guess now is it somehow lies in my
Apache.
That's just RedHat's stock Apache 2.0.52, but it may be related to
a
module (I shutoff the mods I wasn't using).... at least I have
something
to poke at now....
Thanks,
Scott Wilson Lead System Administrator
user-999746b4662e@xymon.invalid NSIT - DCS - SeaSol
On Fri, 26 Oct 2007, Josh Luthman wrote:
Troubleshooting technique!
Use the Hobbitmon demo on the project page. This was the only
page
long
enough (for me - 1680x1050 resolution) for me to scroll down and
maintain my
position.
http://www.hswn.dk/hobbit/services/services.html
Does this maintain position for you, Scott?
My client: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
rv:1.8.1.8)
Gecko/20071008 Firefox/2.0.0.8
On 10/26/07, Hobbit User <user-24d6f8323faa@xymon.invalid> wrote:
On Fri, October 26, 2007 11:48, Scott Wilson wrote:
What is the behavior for others? Do others not get forced to
the
top
upon
refresh. (Automatic or manual has the same behavior.)
Firefox 1.5.0.12 on Centos EL5 maintains position on refresh.
Help-->About Firefox: Mozilla/5.0 (X11; U; Linux i686; en-US;
rv:
1.8.0.12)
Gecko/20070719 CentOS/1.5.0.12-3.el5.centos
--
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
Those who don't understand UNIX are condemned to reinvent it,
poorly.
--- Henry Spencer
--
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
Those who don't understand UNIX are condemned to reinvent it,
poorly.
--- Henry Spencer
--
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencer