Em Fri, 05 Dec 2003 14:19:15 -0200, Leandro GuimarÃes Faria Corsetti
Dutra escreveu:

> Em Fri, 05 Dec 2003 11:58:39 -0200, Christoph Simon escreveu:
> 
>> Try the -f option of strace to get the trace of the kids.
> 
>       It goes forever repeating the following lines without end:
> 
> [pid 12551] poll([{fd=22, events=POLLIN}], 1, 2000) = 0 [pid 12551]
> getppid()                   = 12542 [pid 12551] poll(

        Found something else interesting too, before the above lines:

[pid 12945] time(NULL)                  = 1070642167
[pid 12945] time(NULL)                  = 1070642167
[pid 12945] read(29, "time=\"1068594145\" visits=\"1\"/>\n "..., 4096) = 4096
[pid 12945] time(NULL)                  = 1070642167
[pid 12945] time(NULL)                  = 1070642167
[pid 12945] --- SIGSEGV (Segmentation fault) @ 0 (0) ---
[pid 12945] unlink("/home/leandro/.galeon/mozilla/galeon/lock") = 0
[pid 12945] rt_sigprocmask(SIG_SETMASK, NULL, ~[KILL STOP], 8) = 0
[pid 12945] rt_sigsuspend(~[KILL STOP RTMIN] <unfinished ...>
[pid 12953] <... poll resumed> [{fd=22, events=POLLIN}], 1, 2000) = 0
[pid 12953] getppid()                   = 12945
[pid 12953] poll([{fd=22, events=POLLIN}], 1, 2000) = 0
[pid 12953] getppid()                   = 12945
[pid 12953] poll([{fd=22, events=POLLIN}], 1, 2000) = 0


        So it seems it takes some session information from somewhere else 
than ~/.galeon/session/, and then it tries to do something that causes a
segmentation fault, but not to the main processes as these continue to
run... but it is something drastic enough to prevent any windows from
opening, and to prevent it from dying entirely at SIGTERM.

        Would deleting the contents of ~./galeon/mozilla/galeon/Cache
be likely to make things worse or better?

        Any other ideas?
 

-- 
Leandro GuimarÃes Faria Corsetti Dutra <[EMAIL PROTECTED]>
                                                +55 (11) 9406 7191
Belo Horizonte, Londrina, SÃo Paulo             +55 (11) 5686 9607
http://br.geocities.com./lgcdutra/              +55 (11) 5685 2219



-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]

Reply via email to