This ticket is no longer relevant.
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
All Stories
Mar 22 2019
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This seems like low-priority aesthetics to me more so than an actual problem.
No longer relevant.
No longer relevant.
We do not actively support the database backend driver, so please bear in mind this has a very low priority for us.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
The application can't address having run out of file descriptors.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
This ticket is no longer relevant.
In a broader sense, many such form fields in the webclient require input confirmation by pressing Enter, including those that initially use a "display filter", like the one able to search listed calendars vs. searching the server-side for all possible options. The exception to the rule seems to be auto-complete.
eimap is going the way of the dodo, but I'll comply with your request for at least a license.
Folders should generally not have a trailing space, so this should be fixed at the input / form level.
It seems to me this doesn't have the highest of priorities.
I fail to understand how this is not the expected result. When an input field is selected (presumably the keyboard pops up?), then the cursor is supposed to stick to that input field despite any scrolling, no? The cursor only disappears when a click event happens outside the input field?
There is nothing...
I've disabled registration.
In the web client, what does Settings > ActiveSync show, and/or what does it show for the "device" or "application" listed?
This fundamentally flawed version of a build system has needed to be patched again, but this crap should now be resolved.
Thanks for your reply. I've just checked and the issue persist. I've tried with my user and I can confirm I logged first in webmail before adding exchange account.
Mar 21 2019
If only things were that easy. :-/
We have no standard definition for "slow", and you're not telling us anything about the system symptoms. High CPU load? High I/O? Memory consumption? Entropy?
What modifications exist against pykolab?
Wallace on CentOS 7 for kolab 16 uses systemd, not sysvinit, so I don't understand what I'm being told here.
Sending invitations uses the same routines as composing a new email message and sending that. As far as we're aware, compose+send under GSSAPI works, bearing in mind Postfix will of course need to be configured for it.
Please update roundcubemail-plugins-kolab.
Your problems shouldn't have occurred if you choose "Internet configuration" for postfix.
This isn't a Kolab bug, nor a bug in the master branch of kolab-plugins that provides kolab_2fa, that closely follows the roundcube upstream master branch.
Destroyed.
Could it be the user has not yet logged in to webmail before attempting to use ActiveSync? Should be resolved in the latest round of updates.
Was, at any point, some calendar imported?
In T5148#72927, @machniak wrote:This has been fixed in 5834133fadb9e. Not yet packaged.
Either remove the content from the files, or remove the module from the configured list of modules.
The aspell-fi and aspell-pl dependencies have been dropped from kolab-webclient.
It seems the signatures are valid:
Hi,
this is covered in my feature request:
https://git.kolab.org/T5239
This isn't going to be a discussion about whichever level people may think they require for checksummed payload obtained over a secure channel, but should instead revolve around how something like the OBS starts failing to sign packages/repositories, and seemingly for APT repositories only.
Mar 19 2019
- Use python UUID to convert binary objectGUID to string. This way dont need to chage cache db format. It doesn't look like value of uniqueid returned by from cache is used anywhere (needs to double check), so only forward conversion is needed.