Dear Rene, thank you for the quick reply and checks. I think it can be closed, see below.
On 01/30/2017 12:09 PM, Rene Engelhard wrote: > reassign 853149 libreoffice-core > tag 853149 + unreproducible > tag 853149 + moreinfo > thanks > > On Mon, Jan 30, 2017 at 11:00:07AM +0100, Tjeerd Pinkert wrote: >> Package: libreoffice-base > > Obviously not. It's not a bug in the Database application, is it? > Please report against correct packages (and in this case the package > description > of libreoffice-base is clear what it is) I'm sorry, this is my fault, realised only after sending it should have been another package, would have chosen libreoffice-common or -core I think, still wrong, -base sounds like it (in Dutch) but is not :~ >> Version: 4.3.3-2+deb8u5 >> >> After an update I found that LibreOffice did not start anymore. It gave >> an error dialog saying: > > When and what and how? I restarted my machine last friday after a longer time with multiple updates done in the mean time (it partially acts as a server), but not as long ago as mid-2016. > 1:4.3.3-2+deb8u5 is there since months (mid-2016), as is > libreoffice-lightproof-en (from jessie release onwards) > (which would be the correct package to file it in if it was a bug in th > packages' file permissions) > >> After searching the web I found that this might happen if the user >> profile is defunct (that was not the case) or the Extention Manager has >> no access to some of the shared library locations: >> >> I found that including all rX on the following directories solved the bug. >> >> user@pudding:/usr/lib/libreoffice/share/extensions/lightproof_en/pythonpath$ >> ls -l >> total 44 >> -rw-r--r-- 1 root root 4078 May 15 2014 >> lightproof_handler_lightproof_en.py >> -rw-r--r-- 1 root root 14966 May 15 2014 lightproof_impl_lightproof_en.py >> -rw-r--r-- 1 root root 13965 May 15 2014 lightproof_lightproof_en.py >> -rw-r--r-- 1 root root 260 May 15 2014 lightproof_opts_lightproof_en.py >> drwxr-x--- 2 root root 4096 Jan 24 10:58 __pycache__ >> user@pudding:/usr/lib/libreoffice/share/extensions/lightproof_en/pythonpath$ >> sudo chmod a+rX __pycache__/ >> user@pudding:/usr/lib/libreoffice/share/extensions/lightproof_en/pythonpath$ >> cd __pycache__/ >> user@pudding:/usr/lib/libreoffice/share/extensions/lightproof_en/pythonpath/__pycache__$ >> ls -l >> total 28 >> -rw-r----- 1 root root 4938 Jan 24 10:58 >> lightproof_handler_lightproof_en.cpython-34.pyc >> -rw-r----- 1 root root 14624 Jan 24 10:58 >> lightproof_impl_lightproof_en.cpython-34.pyc >> -rw-r----- 1 root root 435 Jan 24 10:58 >> lightproof_opts_lightproof_en.cpython-34.pyc >> user@pudding:/usr/lib/libreoffice/share/extensions/lightproof_en/pythonpath/__pycache__$ >> sudo chmod a+rX * > > After installing libreoffice-lightproof-en and libreoffice-writer in a clean > jessie and proofchecking a simple wrong sentence and closing I get: > > drwxr-xr-x 2 root root 4096 Jan 30 11:04 __pycache__ > > So it's ok. (And normal files don't need to be +x) X is not x. Interesting that the clean install is OK. >> After this it still did not work, I also changed: >> >> user@pudding:/var/spool/libreoffice/uno_packages/cache$ ls -l >> total 4 >> drwxr-x--- 2 root root 4096 Jan 24 10:58 uno_packages >> user@pudding:/var/spool/libreoffice/uno_packages/cache$ sudo chmod a+rX * > > (cowbuilder-stable)root@frodo:/var/spool/libreoffice/uno_packages/cache# ls -l > total 4 > drwxr-xr-x 2 root root 4096 Jan 30 11:04 uno_packages Interesting... Looks like something on my machine went wrong, but what... > without doing anything. Did you loose directories and recreated them on your > own? > fs problems? No directories lost as far as I know. The latter might have been the case although I had a clean fs on reboot as far as I know/have seen? I also had some other old trouble back today, so it might have been an unclean fs restored to some previous state... Let's keep it on that. > In any case, I can't see any bug in any of the involving packages. OK, I accept your verdict, thank you for the efforts, my excuses for the false report. I hope then this will at the least serve as a lead to other users getting this error (since SSE is not the problem here, but that is what many searches refer to). Best regards, Tjeerd
