[prev in list] [next in list] [prev in thread] [next in thread] 

List:       evolution
Subject:    Re: [Evolution] Crash on startup
From:       Fred RISS <Fred.Riss () iname ! com>
Date:       2001-03-31 14:41:43
Message-ID: 986049704.9799.0.camel () tothkiller
[Download RAW message or body]

Le 31 Mar 2001 22:42:23 +0930, Not Zed a écrit :
> On 30 Mar 2001 17:25:24 +0200, Fred RISS wrote:
> > Hello.
> > 
> > I don't know if this issue has already been reported, since I didn't
> > receive any mails from the list during 3 days (if one of the ximian
guys
> > knows what happened, I'd be glad to know...).
> > I'm experiencing a startup crash which seems to be an indexing
problem,
> > because when I delete the mbox.ibex and mbox.ev-summary files of one
of
> > my maildirs, evolution starts well (but slowly, that's normal). Here
is
> 
> Does it only work once, and then not again?  Or does it happen every
now
> and then?  What sort of things are you doing to make it happen, is
lots
> of mail coming and going, etc?
> 
> The backtrace isnt much help, but if you run evolution-mail in another
> window first, you should get some output that indicates if the indexer
> is crashing (it normally hits assertions when it does).
> 

I did some tests. Here's what I know :
- I get not more information when I launch evolution-mail in another
window before a crash; evolution-mail crashes during 'returning
description for file : <folder>' , before it has processed all the
maildirs
- Evolution seems to crash randomly. Even if I do nothing in evo (just
launch it and then close it), it mostly crashes the next time I try to
start it up.
- Sometimes there is no crash when I do exactly the same as above
- After a crash, sometimes, I can launch evolution by simply doing
killev and then lauching it (so it may finaly not be an indexing bug,
but removing mbox.* seems to work all the time, whereas just killev
doesn't)
- I realize that 'sometimes' is not precise, but I did half an hour of
tests, and I couldn't find any rational law about that crash... sorry

Regards

-- 
Fred




[prev in list] [next in list] [prev in thread] [next in thread]