[Kolab-devel] Kontact reliability - revisited

[ITSEF Admin]

My apologies for the long "silence" - the joys of being a part-time admin: 
"real" work took over... :-} In any case thank you kindly for your response!


On Wednesday 18 July 2007 11:14, Bernhard Reiter wrote:
[...]
> > We have some 35 users, all using Kontact proko2 (by now 2.1.7) on SuSE
> > 10.0.
>
> For others: It is important to know that KK offers full professional
> support for Kontact on a set of systems. This largely depends on if we have
> enough business to fund tighter quality control, packaging and
> the occasional large effort for the really hard problems that are sometimes
> related to other libraries on the platform.
> Most stable currently is Debian Sarge/Proko2 branch.
> Debian Etch/Enterprise Branch is quite close behind this
> and will be the next platform.

Hm. Any opinion on (K)ubuntu?


> Back to Thomas problem:
[...]
> > - Some cases were due to issue1498 (appointments vanish from view, but
> > come back after a restart)
>
> I have pinged developers about this. As we do not see this on Debian,
> it might be related to the Suse KDE setup or libraries.

Ok... We're currently looking into moving to another distro anyway (support 
for SuSE 10.0 runs out in October) and we're not too attached to SuSE, so 
looking at something else is an option. The guys we have most of our IT 
support outsourced to seem to favour Kubuntu, hence my question above... :-)


[...]
> > - In several cases, events were created but they somehow vanished from
> > the server
>
> You need to find out if they have been on the server in the first place.
> Things to try:
> * Check the imapd logs, you should be able to follow created and deleted
>   emails in there.
> * do lists of the files more often for the users in question.

Right. I have on my ToDo to add an additional hourly backup of all groupware 
folders (should be Contacts, Calendar and Tasks as the most important ones 
(haven't heard much about people using Notes and Journal)) for at least a 
couple of days back. If I exclude the actual mail folders, this should be 
doable in an incremental manner without requiring excessive amounts of HD 
space. I reckon this will happen within the next two weeks.


> > - I had two cases where the agenda was emptied completely. As far as I
> > understand the whole system, this should only happen if for some reason
> > the local DIMAP cache is removed, right? In that case, I *might* tend to
> > put it down to [0].
>
> Also find out how many users have write access to the folder in question.
> Are you talking about single user folders?

In all cases that I have actually heard of: Yes. There aren't many shared 
groupware folders in use around here.


> With Agenda: Do you mean the objects of one calender folder?

Yes - from a user's point of view: "my agenda is empty" basically means that 
(s)he does not see any appointments in the calendar anymore, and in the cases 
I've seen, all events had indeed vanished from the IMAP server and needed to 
be restored from backup.


> Lists of the files in shorter periods might help.
> You could count the number of files on the server in the calender folders
> and examine any change to find out what might be going on and regain trust.

Interesting idea, I'll keep that in mind when setting up the extra back-up.


> > I myself am using the exact same setup, but have yet to loose a single
> > appointment, despite intense use - one of the reasons why this is so
> > difficult to handle... I still do not exclude user mistakes, to be
> > honest, however, right now I can't quite see how a user can accidentally,
> > without noticing, remove an appointment himself.
>
> Do they physically switch off machines?
> Unreported crashes?

To that extent, I agreed with $BOSS to come up with a short user enquete, so 
we finally get more reliable data (besides the few reports we actullally 
*did* get). All these rumours are pure poison when it comes to trust (throw 
some general "but it worked under Windows" into the mix and you get the 
idea...).


[...]
> We can only work with what we have
> and KDE / Kontact does not seem to be NFS safe.
> KDE 4 hopefully will change this, but it is far away.
> My guess is that your problems are not related to the filesystem,
> so I would examine other things first, because the problems might
> just stay if you move the AFS.

Good point.


> > As always, thanks in advance for all and any insights you might have for
> > me. I'm really scratching my head here, especially, as I see no other
> > reports about similar problems e.g. on kolab-users - so there is a chance
> > we're doing something stupid. If only I knew... :-}
>
> This is why I guess that the special versions you are using for KDE
>  libraries from Suse could be the problem. I do not think the combination
> is used very often.
>
> You could try to see if you have more success with Enterprise branch,
> this is in a testing/beta stage right now. It would be an alternative path
> which would enable you to use newer KDE versions for the libraries as well
> getting you closer to where other KDE developers are.

Well, what I'm currently planning is to prepare a test machine with Kubuntu 
7.04 (KDE 3.5.6 or even 3.5.7) and the proko2 client - and then give it to 
our user with the worst problems. If that machine works for that user, I'd be 
cautiously optimistic that we're out of the fire... ...unless you have some 
strong recommendations against Kubuntu. Reason I agree with the Kubuntu 
choice is that we want to stay as close as possible to what SuSE has with 
regard to user interface (if I dare to come up with some extensive UI changes 
some people will probably start to revolt...) and that seems easier with 
Kubuntu, as it is already KDE centered.

Talking about which: Would you recommend using the Debian binary of the proko2 
client or rather build from source?

Thanks again for your insight, I really appreciate it!

Cheerio,

Thomas