Please re-read the first post, Yes it is very technical, but...

There is no CSSU-Training, it is CSSU-Testing you are looking for.

The error seems to indicate that the necessary repositories might not be added for the kernel install, or the update of the packages list did not succeed. You need a capable kernel, either kernel-power or kernel-cssu, for thumb2-compiled binaries to run. You might have your repo settings checked by posting here the output of

It is called CSSU-Testing, not Training, and it is not an application you could fimd in the list but an operating system update. The best guide to install it is following the link in the top right corner of the screen, where it says Maemo community update.

CSSU requires an updated kernel, either its own or kernel-power. Your conflicting kernel errors indicate that you are not following the correct installation process. Please follow that link and try again.

Does the battery drain continue? As the only strange thing I see in your log are those "wlan0: driver reports beacon loss from" piles of messages, see https://bugs.maemo.org/show_bug.cgi?id=9101

maemo-kernel etc is kernel-power 24 or something IIRC, I think it would be way easier for him to reflash and install CSSU on top of a clean system, as he seems a bit clueless

"Core was generated by `update-mime-database /usr/share/mime'"

The same command doesn't segfault in my SB, and as shared-mime-info package (update-mime-database is part of) doesn't have -dbg package, I can't get a meaningful backtrace. However, it seems to me that something is broken in you SB installation

Installed and all is fine - even under heavy load n900 remains stable. Thanks FMG.

Being amongst the best of the linux geeks, you should have done your homework before polluting the thread with useless comments and emotions, i.e.:

24.11.2013 - 21.2011.38-1Tmaemo8.3+thumb3
* upgrade hildon-desktop to 1:2.2.153-2+thumb0
* Thumb2 compile cairo, libx11, libxfont

There is nothing in that update that could cause the severe regression you're reporting, as simple as that. *If* your problems had started after the previous update(21.2011.38-1Tmaemo8.3+thumb2), which upgraded glib that would have been another beer.

Honestly, I can't think of a reason why you refused to provide information about do you OC and what is your filesystems' state, but it is your right. As is my right to say "RTFM" and ignore your further non-informative posts. As much as I feel responsible for the quality of the SW I put in the cssu-thumb repo, despite the fact that I try to do it in the same way I handle the SW in my job (I am developer for more that 25 year, the last ~15 of them - professional), I simply refuse to continue to be involved in such a waste of time.

Thanks for the advices, but I *know* how to do my job. And no, I am not trying to help you, I know you're experienced enough to fix your problem, eh? What I am trying to do is to find if really there is a problem in cssu-thumb went unnoticed(no matter how small the chance is) and fix it for those that are clueless. IOW I was asking for your help, but it seems you CBA to give it. I am fine with that, after all you owe me nothing. Neither do I.

it doesn't look like the battery is still draining that much. i just posted the logs since there were errors at the bottom and i can't make heads or tails of them.
thanks!

if you would care to read my post(s) properly you would find all the answers to all your questions
but maybe a long post w/out code isn't worthwhile your attention

however that might be, my professional & home experience based on

• having had a N900 for the last 3½ yrs - more then 2½ of which w/ power kernel, running pretty stable (certainly without crashes during GSM calls )
• having had many crashes (including in the middle of GSM calls ) since 22nd of September when i did a fresh install with the configuration that run fine for months before except adding CSSU(-thumb) according to TFM.
  the previous config only run the most recent version software installed w/out CSSU whatsoever

tells me CSSU doesn't work (for me)
thus i'm going to scrap it and re-flash Fremantle and re-install the same packages as before 22nd of September
sorry if i'm not good enough for your perfect software
guess

says it all... user has to abide by your perfect code or can go to hell
nuff said

Let me see...

Question 1: Did you overclock?
Answer: "the device is just the same config (except for adding CSSU-thumb) then until 2 months ago..."

Err... does that mean yes or no?

Question 2: Did you check filesystems for corruption(you can check fsck logs in /var/log/ )?
Answer: "i'm a Linux user (professionally & privately) for over 15 yrs... i know how to take care of a Linux system, thanks for asking."

Err... does that mean yes or no?

Question 3 (implied): Did you check the battery?
Answer: "again, ceteris paribus..."

Err... does that mean yes or no?

Question 4: Could you install syslogd and sp-rich-core and provide coredump next time crash/reboot happens.
Answer: "i'm going to re-flash the other one without CSSU-* & think about whether i'm going to leave this one w/ CSSU-thumb
even IF i do, it is going to go back to the "alarm-clock" & prepay calls status... if it still crashes under those circumstances, reformat..."

At least this one is clear. I take it to mean no.

And yes, I have looked though the complete list of your posts, including in other threads and no, I have not found anything specific. Argumentum ex silentio comes to mind. I am sure you are familiar with it, since you like Latin phrases so much.

Ab imo pectore, vade in pace.