[GNUnet-developers] segmentation fault

[Jan Marco Alkema]

Hello Developers of gnunet,

Mine first goal is a reliable gnunet (24 hours a day and 365 days a year).
People won't use gnunet if the system goes down very often (segmentation
errors, etc).

I see that Chistian is not available until 11/10/02, maybe someone else can
solve the segmentation error:

Nov  1 19:37:43 DEBUG: local lookup did not find a result for
5EBDC0F5F571FA48EC467D457876DEBB755B30BD
Nov  1 19:37:43 DEBUG: local lookup did not find a result for
2A0393CA5C6E6751566BF745A042449D77D2F6FD
Nov  1 19:37:43 DEBUG: local lookup did not find a result for
F27FCAA1B9E48FD933532C7829395B33425A788D
Nov  1 19:37:43 INFO: Adding host A2BA21B8314DF1B95B74099850D6C8671AA8E344
to the connection table.
Segmentation fault

Nov  2 12:34:26 DEBUG: verified UDP helo from 12.247.254.37:6339
Nov  2 12:34:26 INFO: binding address of node
A2BA21B8314DF1B95B74099850D6C8671AA8E344.17
Nov  2 12:34:26 DEBUG: Received sessionkey from host
A2BA21B8314DF1B95B74099850D6C8671AA8E344.
Nov  2 12:34:26 DEBUG: scanning directory /usr/local/gnunet/var/data/hosts/
for peer identity
Nov  2 12:34:27 INFO: Adding host A2BA21B8314DF1B95B74099850D6C8671AA8E344
to the connection table.

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 2051 (LWP 30365)]
transportDisconnect (tsession=0x46236e1a) at transport.c:309
309       if (tsession->ttype >= tapis_count) {
(gdb) ba
#0  transportDisconnect (tsession=0x46236e1a) at transport.c:309
#1  0x080501e4 in acceptSessionKey (sender=0x8068154, tsession=0x0,
msg=0x8109424) at sessionkey.c:414
#2  0x0804a951 in handlePlaintext (msg=0x81091f0 "\0024", tsession=0x0,
sender=0x8068154, size=1084,
    crc=-353306659) at handler.c:197
#3  0x0804aa5f in handleMessage (tsession=0x0, sender=0x8068154,
msg=0x81091f0, size=1084, isEncrypted=0,
    crc=-353306659) at handler.c:262
#4  0x0804f588 in threadMain (id=-1073743280) at core.c:151
#5  0x4007afef in pthread_start_thread () from /lib/i686/libpthread.so.0
#6  0x4007b0df in pthread_start_thread_event () from
/lib/i686/libpthread.so.0

(gdb) backtrace
#0  transportDisconnect (tsession=0x46236e1a) at transport.c:309
#1  0x080501e4 in acceptSessionKey (sender=0x8068154, tsession=0x0,
msg=0x8109424) at sessionkey.c:414
#2  0x0804a951 in handlePlaintext (msg=0x81091f0 "\0024", tsession=0x0,
sender=0x8068154, size=1084,
    crc=-353306659) at handler.c:197
#3  0x0804aa5f in handleMessage (tsession=0x0, sender=0x8068154,
msg=0x81091f0, size=1084, isEncrypted=0,
    crc=-353306659) at handler.c:262
#4  0x0804f588 in threadMain (id=-1073743280) at core.c:151
#5  0x4007afef in pthread_start_thread () from /lib/i686/libpthread.so.0
#6  0x4007b0df in pthread_start_thread_event () from
/lib/i686/libpthread.so.0

I could be related to adding host in a inproper way.

Greetings Jan Marco

P.S. I use gdb ./gnunetd

I see strange 'processes' (gnunetd <defunct> in the tasks list appear if I
execute ./gnunet-stats

14963 root      15   0  4808 4688  1132 S     0.0  7.5   0:00 gnunetd
14964 root      15   0  4808 4688  1132 S     0.0  7.5   0:00 gnunetd
14971 root      17   0  1292 1032   948 S     0.0  1.6   0:00 bash
14982 root      16   0     0    0     0 Z     0.0  0.0   8:04 gnunetd
<defunct>
15024 root      16   0     0    0     0 Z     0.0  0.0   0:00 gnunetd
<defunct>
15031 root      16   0     0    0     0 Z     0.0  0.0   0:00 gnunetd
<defunct>