Closed GoogleCodeExporter closed 9 years ago
Why did you decide that problem is in Firefox 5? Did you check it using other
browser? I have no any problems with Firefox 5.
Check /opt/ovz-web-panel/log/production.log for errors. At least it should not
be empty and should contain information about latest processed requests.
Original comment by sibprogrammer
on 6 Jul 2011 at 12:50
Here it works only on Internet Explorer. Dont work on firefox 3 (portable) and
dont work on firefox 5.
When i put admin and pass... it´s show blank webpage. But on Internet Explorer
works good. No problems with this.
I will try to see that log and post it here. Thank you.
Original comment by fcbi...@gmail.com
on 6 Jul 2011 at 10:57
I have the same problem with Firefox 5, by the way IE8 sometime crash too,
Chrome work great.
this is the log /opt/ovz-web-panel/log/production.log
Firefox 5
----------------------------------------------------
Processing SessionsController#new (for 4.8.7.1 at 2011-07-09 11:18:10) [GET]
Parameters: {"action"=>"new", "controller"=>"sessions"}
Redirected to controlleradmin/dashboard
Completed in 8ms (DB: 1) | 302 Found [http://6.3.6.8/]
Processing Admin::DashboardController#index (for 4.8.7.1 at 2011-07-09
11:18:10) [GET]
Parameters: {"action"=>"index", "controller"=>"admin/dashboard"}
Rendering template within layouts/admin
Rendering admin/dashboard/index
Completed in 169ms (View: 140, DB: 8) | 200 OK [http://6.3.6.8/admin/dashboard]
----------------------------------------------------
IE8
----------------------------------------------------
Processing SessionsController#new (for 4.8.7.1 at 2011-07-09 11:23:13) [GET]
Parameters: {"action"=>"new", "controller"=>"sessions"}
Rendering template within layouts/application
Rendering sessions/new
Completed in 19ms (View: 13, DB: 0) | 200 OK [http://6.3.6.8/]
----------------------------------------------------
Chrome
----------------------------------------------------
Processing SessionsController#new (for 4.8.7.1 at 2011-07-09 11:24:24) [GET]
Parameters: {"action"=>"new", "controller"=>"sessions"}
Rendering template within layouts/application
Rendering sessions/new
Completed in 20ms (View: 14, DB: 0) | 200 OK [http://6.3.6.8/]
-----------------------------------------------------
Original comment by dataclub...@gmail.com
on 9 Jul 2011 at 7:32
Your log shows that you passed login screen and should see dashboard. Something
wrong with dashboard? Can you give me access to your installation (privately
via mail)?
Original comment by sibprogrammer
on 10 Jul 2011 at 2:21
Sorry for this server i cen give you access, tomorrow i will try install to
another server CentOS 5.6 + ovz_panel and we will see..
so i see this error in a firebug, when i try connect to my http://server:3000
via Firefox 5
-----------------------------------------------
Console
-------
Ext is not defined
[Break On This Error] Ext.onReady(function() { ---- line 20
--------
500 Internal Server Error
1.1.1.1:3000
343 B
20ms
---------
Response Headersview source
Connection close
Date Sun, 10 Jul 2011 20:35:34 GMT
Content-Type text/html; charset=ISO-8859-1
Etag 2098946-2105f-4c9f00e6
Server WEBrick/1.3.1 (Ruby/1.8.5/2006-08-25)
Content-Length 343
-------------------
I have attached file, please see.
Original comment by dataclub...@gmail.com
on 10 Jul 2011 at 7:44
Attachments:
Guys, first of all to avoid mixing all problems into one you should add info
about your installation. Please report your Owp version number. You see 500
errors, this means that webserver crashes on some requests. This means that
production.log should contain errors. Please check the log for errors and post
them here.
Also check issues
http://code.google.com/p/ovz-web-panel/issues/detail?id=209
http://code.google.com/p/ovz-web-panel/issues/detail?id=79
Original comment by sibprogrammer
on 12 Jul 2011 at 3:20
I have the same issue but i resolving cleaning the cash browser all the time
Original comment by agsw...@gmail.com
on 12 Jul 2011 at 4:00
Attachments:
What about my questions from comment #6 ?
P.S. Looks like issue title is completely misleading.
Original comment by sibprogrammer
on 12 Jul 2011 at 4:06
I haven't 500 error but 302
Steps to reproduce:
New install of webpanel 2.0
Actual result: Open http://IP:3000 - Login with admin admin
Expected result: Blank page (dont show nothing)
Version of the product: (latest version, downloaded today).
Server OS: Centos 5.6 x64
Browser: IE6
Processing SessionsController#new (for xxx.xxx.xxx.xxx at 2011-07-12 20:00:07)
[GET]
Parameters: {"action"=>"new", "controller"=>"sessions"}
User admin was updated.
Cookie set: auth_token=829a82829776aafcd0c0d825d2e73fd18dd9ac02; path=/;
expires=Tue, 26 Jul 2011 17:54:21 GMT
Redirected to controlleradmin/dashboard
Completed in 191ms (DB: 1) | 302 Found [http://64.120.133.130/]
Original comment by agsw...@gmail.com
on 12 Jul 2011 at 4:49
I have the same issue on comment 3 with firefox 5
the production log continue to (sorry i have cut)
Processing Admin::DashboardController#index (for xxx.xxx.xxx.xxx at 2011-07-12
20:00:07) [GET]
Parameters: {"action"=>"index", "controller"=>"admin/dashboard"}
Rendering template within layouts/admin
Rendering admin/dashboard/index
Completed in 111ms (View: 104, DB: 2) | 200 OK
[http://64.120.133.130/admin/dashboard]
Original comment by agsw...@gmail.com
on 12 Jul 2011 at 4:58
(sorry for my numerous post, isn't here the possibility to edit, but i thing
this can help you)
Also the previous error message is:
Original comment by agsw...@gmail.com
on 12 Jul 2011 at 5:09
Attachments:
IE6 is not supported. Please use another browser.
Original comment by sibprogrammer
on 13 Jul 2011 at 12:45
The issue is not related to the browser, also this occurs if you have succesful
logged in and you do F5 (refresh page), here the IE8 error
Original comment by agsw...@gmail.com
on 13 Jul 2011 at 8:36
Attachments:
Can anybody give me access (via mail) to problematic server?
Original comment by sibprogrammer
on 13 Jul 2011 at 1:54
the issue occurs only with centos 5.6 x86_64, i have installed new server
debian 6 squeeze x86_64 and there is no issue.
sibprogrammer write me in private your email i send to you the credential.
Original comment by agsw...@gmail.com
on 13 Jul 2011 at 3:05
i think the problem is that
centos running ruby 1.8.5 (2006-08-25)
debian running ruby 1.8.7 (2010-08-16)
Original comment by agsw...@gmail.com
on 13 Jul 2011 at 3:32
Ok, found the 'problem'. It's not just x64, it affects i686 as well. I've tried
this on 3 different system, sometimes it worked in Firefox 5 other times it
didn't. I narrowed it down to it not working on FF5 on the systems I did a yum
update to. The systems where FF5 worked was running ruby-1.8.5-5.el5_4.8 but
the systems where FF5 would not work, near identical installation, they were
running ruby-1.8.5-19.el5_6.1
So still 1.8.5 but they must have made a change or a patch that broke things.
Still worked in IE though.
So I installed Ruby 1.8.6 via their repository (http://krnjevic.com/wp/?p=75)
and now it works in FireFox 5 again. So to make a long story short, either run
Ruby 1.8.5-5.el5_4.8 (not recommended) or upgrade to 1.8.6.111-1
Original comment by glitch...@gmail.com
on 13 Jul 2011 at 10:12
Well i'm right but i don't want to install from any unofficial repo
than i recompile ruby from myself here my script
rm -rf /usr/local/lib/ruby
rm -rf /usr/local/src/ruby-1.8.7-p352
rm -rf /usr/local/src/rubygems-1.3.5
yum -y remove ruby
yum -y install gcc openssl-devel zlib-devel
mkdir /usr/local/src
cd /usr/local/src
wget ftp://ftp.ruby-lang.org/pub/ruby/1.8/ruby-1.8.7-p352.tar.gz
tar xzvf ruby-1.8.7-p352.tar.gz
cd ruby-1.8.7-p352
./configure --enable-shared --enable-pthread
make
make install
cd ext/zlib
ruby extconf.rb –with-zlib-include=/usr/include –with-zlib-lib=/usr/lib64
cd ../../
make
make install
cd ext/openssl
ruby extconf.rb
cd ../..
make
make install
cd /usr/local/src
wget http://rubyforge.org/frs/download.php/60718/rubygems-1.3.5.tgz
tar xzvf rubygems-1.3.5.tgz
cd rubygems-1.3.5
ruby setup.rb
gem install rake
gem install sqlite3-ruby --version=1.2.4
All work fine again
Original comment by agsw...@gmail.com
on 14 Jul 2011 at 5:16
and do
cp /usr/local/bin/ruby /usr/bin/
Original comment by agsw...@gmail.com
on 15 Jul 2011 at 6:22
I'm upgraded ruby to version 1.8.6, but this bug is still present.
Original comment by an...@hhut.ru
on 28 Jul 2011 at 10:18
I see no other way, than to mark this bug as invalid. 5 people posting comments
without details or details about similar but not the same problems are making
no opportunity to help.
Please don't waste my time. If you want to get help, always provide as much
details as possible. If you not sure fill separate reports. Don't post "I have
the same bug" comments w/o details. What are the details? Your OS, exact panel
version, used browser, steps to reproduce, errors from logs and etc. Always
re-read your comments and ask yourself "can anybody understand me/did I give
enough information"?
P.S. Ruby 1.8.6 is more than four years old! Use 1.8.7 if possible.
Original comment by sibprogrammer
on 28 Jul 2011 at 1:55
Greetings everyone - I believe I might have found the source of the blank
pages. I had the same issue and finally tracked it down to a call to a
'private' function. This allowed Google to finger webrick as being the source
of the problem.
Please review http://www.ruby-forum.com/topic/206225 for further information.
Long stroy short : switch to mongrel or modify htmlutils.rb to force returning
a string instead of a class.
Original comment by pduboi...@gmail.com
on 12 Aug 2011 at 9:06
Yes, comment #22 is localize problem - this is a webrick problem. Now i'm moved
owp to use the nginx passenger and all works fine.
to sibprogrammer:
OS: CentOS 5.6
Panel version: 2.0
Browser: Firefox 5.0
Steps to reproduce: spontaneus blank screens, when click to any links in
interface
Log: In log all ok, not found any errors
Original comment by an...@hhut.ru
on 16 Aug 2011 at 9:38
Furthermore, there already exists a workaround in the ai.sh installation script
:
--
if [ "$DISTRIB_ID" = "Ubuntu" -a "`lsb_release -cs`" = "lucid" ]; then
puts "Patching broken WEBrick for Ubuntu 10.04"
sed -i'' 's/string.dup/string.to_s.dup/' /usr/lib/ruby/1.8/webrick/htmlutils.rb
fi
--
This shows the issue is known - but the workaround leaves the installed system
open to failure upon any update to the associated ruby packages.
With this in mind, the proper fix would be to move away from webrick to
mongrel. This requires a slight modification to the init script and a
dependency on the mongrel package.
sibprogrammer; what is your opinion on this fix ?
Original comment by pduboi...@gmail.com
on 16 Aug 2011 at 3:01
Posted a new ticket :
http://code.google.com/p/ovz-web-panel/issues/detail?id=330
Original comment by pduboi...@gmail.com
on 16 Aug 2011 at 3:39
Installed v2 today:
Platform: CentSO5.7
Browser: Firefox 7 and midori and Chromium 17.0.931.0 (Developer Build 108803
Linux)
In Firefox 7, it still gives a blank page after login whereas in Midori it
shows up the dashboard. But when I tried to change the admin password, Midori
also got a blank page :-(
But with Chromium 17.x it works fine.
Any workout?
Original comment by garbytr...@gmail.com
on 8 Nov 2011 at 11:06
Original issue reported on code.google.com by
fcbi...@gmail.com
on 5 Jul 2011 at 8:23