[Desktop-packages] [Bug 1490293] [NEW] STARTTLS Handshake problem introduced in 1:31.8.0+build1

2015-08-30 Thread Fionn
Public bug reported:

I am running trusty. After the update from 1:31.7.0+build1 to 1:31.8.0+build1 i 
suddenly was no longer able to send mail via my provider.
The mail server is running exim and uses STARTTLS for connection encryption. 
When hitting the send button, Thunderbird informed me that it is connecting to 
the mail server, immediately followed by a popup informing me that sending 
failed due to an unknown error and that I should talk to my network admin.

I then made sure they had not changed anything in the exim server config
(for quite some time actually). When I tried to send mail, the server
would just have an entry in its log like this:

2015-08-30 15:35:04 TLS error on connection from (my.host.some.where)
[176.198.xxx.yyy] (gnutls_handshake): A TLS packet with unexpected
length was received.

Sending mail from Android or with Evolution still works.

I then grabbed a backup and downgraded thunderbird to the previous
version 1:31.7.0+build1. The problem went away immediately.

A simple test using

# openssl s_client -connect mail.my.provider.com:25 -starttls smtp

does not exhibit anything uncommon, the handshake goes through without a
hickup.

Since thunderbird is not totally uncommon (although noone else seems to
have complained yet), I could arrange a test login for anyone willing to
run an investigation on the issue. I just don't want to publish server
names or IPs here.

** Affects: thunderbird (Ubuntu)
 Importance: Undecided
 Status: New

** Description changed:

+ I am running trusty. After the update from 1:31.7.0+build1 to 1:31.8.0+build1 
i suddenly was no longer able to send mail via my provider.
+ The mail server is running exim and uses STARTTLS for connection encryption. 
When hitting the send button, Thunderbird informed me that it is connecting to 
the mail server, immediately followed by a popup informing me that sending 
failed due to an unknown error and that I should talk to my network admin.
  
- I am running trusty. After the update from 1:31.7.0+build1 to 1:31.8.0+build1 
i suddenly was no longer able to send mail via my provider.
- The mail server is running exim and uses STARTTLS for connection encryption. 
When hitting the send button, Thunderbird informed me that it is connecting to 
the mail server, immediately followed that sending failed due to an unknown 
error and that I should talk to my network admin.
- 
- I then made sure they had not changed anything in the exim config (for
- quite some time). Also, doing a STARTTLS handshake with  When I tried to
- send mail, the server would just have an entry in its log like this:
+ I then made sure they had not changed anything in the exim server config
+ (for quite some time actually). When I tried to send mail, the server
+ would just have an entry in its log like this:
  
  2015-08-30 15:35:04 TLS error on connection from (my.host.some.where)
  [176.198.xxx.yyy] (gnutls_handshake): A TLS packet with unexpected
  length was received.
+ 
+ Sending mail from Android or with Evolution still works.
  
  I then grabbed a backup and downgraded thunderbird to the previous
  version 1:31.7.0+build1. The problem went away immediately.
  
  A simple test using
  
  # openssl s_client -connect mail.my.provider.com:25 -starttls smtp
  
  does not exhibit anything uncommon, the handshake goes through without a
  hickup.
  
  Since thunderbird is not totally uncommon (although noone else seems to
  have complained yet), I could arrange a test login for anyone willing to
  run an investigation on the issue. I just don't want to publish server
  names or IPs here.

-- 
You received this bug notification because you are a member of Desktop
Packages, which is subscribed to thunderbird in Ubuntu.
https://bugs.launchpad.net/bugs/1490293

Title:
  STARTTLS Handshake problem introduced in 1:31.8.0+build1

Status in thunderbird package in Ubuntu:
  New

Bug description:
  I am running trusty. After the update from 1:31.7.0+build1 to 1:31.8.0+build1 
i suddenly was no longer able to send mail via my provider.
  The mail server is running exim and uses STARTTLS for connection encryption. 
When hitting the send button, Thunderbird informed me that it is connecting to 
the mail server, immediately followed by a popup informing me that sending 
failed due to an unknown error and that I should talk to my network admin.

  I then made sure they had not changed anything in the exim server
  config (for quite some time actually). When I tried to send mail, the
  server would just have an entry in its log like this:

  2015-08-30 15:35:04 TLS error on connection from (my.host.some.where)
  [176.198.xxx.yyy] (gnutls_handshake): A TLS packet with unexpected
  length was received.

  Sending mail from Android or with Evolution still works.

  I then grabbed a backup and downgraded thunderbird to the previous
  version 1:31.7.0+build1. The problem went away immediately.

  A simple test using

  # openssl s_client -connect 

[Desktop-packages] [Bug 976032] Re: Place plugin problem with panel in fullscreen and gnome classic

2015-06-08 Thread Fionn
I'd like to mention that I just had exactly this problem with ubuntu-
mate 14.04 (trusty) compiz version
1:0.9.11.3+14.04.20150313-0ubuntu1.1

Disabling the place plugin healed the problem instantly.

-- 
You received this bug notification because you are a member of Desktop
Packages, which is subscribed to compiz in Ubuntu.
https://bugs.launchpad.net/bugs/976032

Title:
  Place plugin problem with panel in fullscreen and gnome classic

Status in Compiz:
  Fix Released
Status in Compiz Core:
  Fix Released
Status in compiz package in Ubuntu:
  Fix Released
Status in compiz source package in Precise:
  Fix Released

Bug description:
  [Impact]

  Flash video fullscreen problem with GNOME panel.

  [Test Case]

  In precise, when I want watch a flash video in fullscreen (like youtube or 
other flash site), the gnome panel doesn't dissapear.
  If I disable the place plugin with CCSM it works well.

  I use the last version of compiz : 1:0.9.7.4-0ubuntu3.

  [Regression Potential]

  Window placement in general, should be tested by manual/checkbox
  tests. The same commit has been well tested in Ubuntu 12.10.

To manage notifications about this bug go to:
https://bugs.launchpad.net/compiz/+bug/976032/+subscriptions

-- 
Mailing list: https://launchpad.net/~desktop-packages
Post to : desktop-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~desktop-packages
More help   : https://help.launchpad.net/ListHelp


[Desktop-packages] [Bug 918200] [NEW] local calendars inaccessible after upgrade from natty (workaround included)

2012-01-18 Thread Fionn
Public bug reported:


Upgrading Evolution from 2.3x (natty) to 3.2 (oneiric), seemingly came with a 
lot of changes of internal data storage formats.

Unfortunately the conversion process leaves room for improvement.

After the upgrade, calendar kept crashing on me often and randomly. Also, I 
could not accept new meeting requests. Instead, evolution showed the following 
error message:
Could not load calendar XYZ (No backend factory for 'file' of 'VEVENT')

Also, the mentioned calendar can neither be opened nor deleted.

According to this source:
https://bugzilla.redhat.com/show_bug.cgi?format=multipleid=716448

It helps to edit some rather complicated xml values in gconf-editor
which apparenty became flawed during the upgrade.

In my case, fixing the base_uri helped somewhat; the above mentioned
error message was replaced by a different one, for which I'll probably
open another bug after some closer examination.

Anyway, it may be worth checking whether this is the only case of a
malformed base_uri and if the upgrade process could be changed to
double-check for them.

** Affects: evolution (Ubuntu)
 Importance: Undecided
 Status: New

-- 
You received this bug notification because you are a member of Desktop
Packages, which is subscribed to evolution in Ubuntu.
https://bugs.launchpad.net/bugs/918200

Title:
  local calendars inaccessible after upgrade from natty (workaround
  included)

Status in “evolution” package in Ubuntu:
  New

Bug description:
  
  Upgrading Evolution from 2.3x (natty) to 3.2 (oneiric), seemingly came with a 
lot of changes of internal data storage formats.

  Unfortunately the conversion process leaves room for improvement.

  After the upgrade, calendar kept crashing on me often and randomly. Also, I 
could not accept new meeting requests. Instead, evolution showed the following 
error message:
  Could not load calendar XYZ (No backend factory for 'file' of 'VEVENT')

  Also, the mentioned calendar can neither be opened nor deleted.

  According to this source:
  https://bugzilla.redhat.com/show_bug.cgi?format=multipleid=716448

  It helps to edit some rather complicated xml values in gconf-editor
  which apparenty became flawed during the upgrade.

  In my case, fixing the base_uri helped somewhat; the above mentioned
  error message was replaced by a different one, for which I'll probably
  open another bug after some closer examination.

  Anyway, it may be worth checking whether this is the only case of a
  malformed base_uri and if the upgrade process could be changed to
  double-check for them.

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/evolution/+bug/918200/+subscriptions

-- 
Mailing list: https://launchpad.net/~desktop-packages
Post to : desktop-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~desktop-packages
More help   : https://help.launchpad.net/ListHelp