Ok, I'm not 100% sure, but I'm pretty sure this issue with the Error
code: sec_error_bad_signature is related to adding the line:
ProxyPass /principals/ http://127.0.0.1:20000/SOGo/dav/ interpolate
to my SOGo.conf for apache2. With that line, after ~1 week, I
intermittently get that error until I restart apache2 (t-bird or web
mail). Without out, I seem not to. When I have that line, I also get the
warning:
[warn] worker http://127.0.0.1:20000/SOGo already used by another worker
but from what I've ready that is just a warning because in general,
apache isn't expecting to have different aliases to the same proxy.
I've noticed that the latest documentation does not say to use that
ProxyPass line (except for Apple Addressbook on port 8800) while before
it was mentioned for iOS mobile calendar access. Any consequences to
just removing the line?
Also, could this be related to running apache2-mom-worker instead of
prefork? This is on debian 5.0 (Lenny)
Thanks,
Ben
On 4/9/2011 11:58 AM, Ben Luey wrote:
A little more news on this issue:
I don't think it is related to networking, as I have the same problem
if I tunnel the traffic through a VPN. Some days I get no errors,
others, every calendar refresh I get
(Error code: sec_error_bad_signature)
I get no errors on the server (either sogo or apache). Running T-bird
on the same machine on Windows (virtualbox) does not reproduce the
problem. Also, no problems when accessing calendars over caldav
(https) on evolution. I'm thinking this is a lightning on Linux
(64bit?) issue.
Any idea of what is going on? I wonder if this 'problem' is there on
windows as well, but windows doesn't display a error message and just
does a retry or something to that effect. I upgrade from 1.3.5a to
1.3.6 and the just-posted integrator and connector and it makes no
difference. Any help would be appreciated, it is very annoying to
click "ok" to 5 pop-ups every 10 minutes.
Thanks,
Ben
On 03/26/2011 09:53 AM, Ben Luey wrote:
I have SOGo setup and running on apache on https. Everything works
fine (t-bird, web interface) for my Windows clients. Webmail works
fine on Linux as well. I recently switched on a Linux machine (Ubuntu
x64 10.10) from evolution to t-bird with all the plugins. Now I
frequently get the error message:
An error occurred during a connection to mail.server.com:443.
Peer's certificate has an invalid signature.
(Error code: sec_error_bad_signature)
this happens to many of the calendar's that I'm subscribed to. Not
every calendar every time it makes a connection, but maybe half the
time t-bird does a refresh 1-4 calendars have this issues. Not all
the same calendars, and some times it is fine. My ssl certs are fine
-- this works from windows computers and web browser, etc.
The command line error message from t-bird is:
Warning: There has been an error reading data for calendar:
Purchasing. Error code: READ_FAILED. Description:
Warning: CalDAV: No response status doing webdav sync for calendar
Personal Calendar (User Name <[email protected]>)
Warning: CalDAV: Error doing webdav sync: undefined
Warning: There has been an error reading data for calendar: Personal
Calendar (User Name <[email protected]>). Error code: DAV_REPORT_ERROR.
Description: There has been an error reading data for calendar:
https://mail.server.com/SOGo/dav/user2/Calendar/user_personal/. It
has been disabled until it is safe to use it.
Warning: There has been an error reading data for calendar: Personal
Calendar (User Name<[email protected]>). Error code: READ_FAILED.
Description:
deliver mode: 7
The problem I'm having is only one location, so it could be some
packets getting mangled, but as I said I didn't have this problem
with evolution (using caldav to access calendars) or via a web
browser -- any ideas? The sogo and apache logs just show the dav
connections to
https://mail.server.com/SOGo/dav/user2/Calendar/user_personal/
with no errors.
Thanks,
Ben
--
[email protected]
https://inverse.ca/sogo/lists
