Bug#484454: [: Re: Bug#484454: openoffice.org-filter-binfilter: hangs at ~98+% CPU trying to open .sdw docs from StarOffice 5.2]
[re-sending to proper Bug#, as suggested by owner@DBTS; somehow it was
addressed to #484453]
----- Forwarded message from -----
Date: Wed, 4 Jun 2008 15:29:29 +0200
To: 484453@bugs.debian.org
Subject: Re: Bug#484454: openoffice.org-filter-binfilter: hangs at ~98+% CPU trying to open .sdw docs from StarOffice 5.2
On Wed, Jun 04, 2008 at 10:29:07AM +0200, Rene Engelhard wrote:
> again. BPO is not in the scope of this BTS and the BTS does not know
hm. ok. Then I suggest the BTS automatically rejects, with proper msg to
reporter, reports for version it doesn't know of and/or isn't interested in.
> Try it in sid/lenny please before reporting a bug..
that'd mean I'd shut up till I eventually upgrade a machine to Lenny, which
would happen rather long after Lenny->stable. Sorry, BPO is the only source
of prepackaged .deb for Etch tracked by packages.debian.org afaik.
Above the stable 2.0.4 that is.
Testing and gathering infos for a bug report takes some effort, if you're
not going to consider bug.rep. against BPO, pls state so clearly - possibly
in the README.Debian - so that we save everybody's time (though I still
need to find a workaroud).
But perhaps that should go (also) on front pages on both pkg search and BTS
as a general statement?
> > this is part of Bug#471348 saga, only that I'm filing against the - hoeply -
> > proper sub-pkg.
>
> In which you didn't submit any info with which the author can debug it.
> And in this bug you didn't either.
wrong, check your mailbox:
!> Date: Sat, 22 Mar 2008 07:01:01 +0100
!> To: Rene Engelhard <rene@debian.org>
!> Subject: Re: Bug#471348: openoffice.org: Segmentation fault trying to open a .sdc
!> Message-ID: <20080322060101.GK20978@localhost>'
in which I attached one of those problematic files.
Assuming such msg didn't vanished midway, of course.
> > Like for the .sdc case there, OOo2.4.0-4~bpo40+1 hangs while trying to open
> > a .sdw made in SO5.2, taking away the CPU.
answering to your other msg: SEGV was the best/worst case: best since OOo2
died before taking the system down, worst since - well - it segv.
That was for OOo2.3.1, OOo2.4.0 worked on those files, though (Debian's) OOo2
takes much more CPU to do the job than OOo1.1.5 *and* official OOo's
OOo2.4.0 (which works just fine, so far).
I feel I'm seeing effects of the some bug, in SO import filters.
> > And like in the other case, OOo1.1.5 does the job without a glitch.
>
> No one even remotely cares about OOo 1.1.5 anymore.
guess so; though I'm not reporting a bug against OOo1x, the fact that OOo1
get it right may give some hint.
> > This is becoming a serious issue, as the doc base (here) is largely .sd* and
>
> Where have you been the last years?
doing something else than keeping on upgrade & break working systems.
> Then you should give out documents to get the authors debug it. Not just
> filing bugs which rot because no one can test. (like #471348).
see above. I do what+when I can - I don't own those docs.
> anyway that it just happens for some documents, not all.
correct (likely); looks like an heisenbug. Likely Debian-specific though, as
OOo2 from OOo (build OOH680_m12_native_packed-1_it.9286) seems to work just
fine (so far), on Woody, Sarge, Etch.
thanks
--
paolo
----- End forwarded message -----
--
paolo
GPG/PGP id:0x3A47DE45 - B5F9 AAA0 44BD 2B63 81E0 971F C6C0 0B87 3A47 DE45
- 9/11: the outrageous deception and ongoing coverup: http://911review.org -
Reply to: