[Orca-users] orca segfaults on start

[Rob Joost]

Hello, (my first post, I read on tested all week, please be patient)
The install went well, I can start_orcallator (although I get a ps: read()
on /proc/9415/as: Value too large for defined data type message, it still
starts??) 
BUT I can Not start orca

root at dev3:/usr/local/bin# ps -ef |grep orc
    root  9415     1  0 17:47:44 pts/2    0:02
/opt/RICHPse/bin/se.sparcv9.5.8 -DWATCH_OS /usr/local/lib/orcallator.se

BUT, when I try to start orca, ( ./orca -v /usr/local/lib/orcallator.cfg )

I get the following:

Orca version 0.265 using RRDs version 1.000401 at Wed Feb 26 18:39:35 2003.
./orca: warning: cannot open state file
`/orca/rrd/rrd/orcallator/orcallator/orca.state' for reading: No such file
or directory
Creating orca_logo.gif.
Creating rrdtool_logo.gif.
Creating rothschild_image_logo.png.
Finding files and setting up data structures at Wed Feb 26 18:39:36 2003.
 RSS  VSZ %MEM        TIME     USER   PID COMMAND
5288 5944  0.7        0:02     root  9568 /usr/local/bin/perl
Current running time is 0:04 minutes.
Loading new data from group orcallator for dev3.
  Read     9 data points from
`/usr/local/var/orca/orcallator/dev3/orcallator-2003-02-26-002'.
  Read    10 data points from
`/usr/local/var/orca/orcallator/dev3/orcallator-2003-02-26-000.bz2'.
  Read     1 data point from
`/usr/local/var/orca/orcallator/dev3/orcallator-2003-02-26-001.bz2'.
Flushing new data and updating PNGs from dev3.
Segmentation Fault (core dumped)


Right before core is dumped, it takes about 20 seconds....
so, I truss it right before core dumps...and get ==>

close(5)                                        = 0
time()                                          = 1046310015
open("/orca/rrd/rrd/orcallator/orcallator/o_dev3/gauge_1024_X_freememK.rrd",
O_RDWR) = 5
fstat64(5, 0xFFBEF798)                          = 0
ioctl(5, TCGETA, 0xFFBEF724)                    Err#25 ENOTTY
read(5, " R R D\0 0 0 0 1\0\0\0\0".., 8192)     = 8192
llseek(5, 0, SEEK_CUR)                          = 8192
lseek(5, 0xFFFFE474, SEEK_CUR)                  = 1140
fcntl(5, F_SETLK, 0xFFBEF928)                   = 0
llseek(5, 0, SEEK_CUR)                          = 1140
lseek(5, 1316, SEEK_SET)                        = 1316
write(5, " AB8 D 6F5C28F \", 8)                 = 8
lseek(5, 688, SEEK_SET)                         = 688
write(5, " > ] k G U N K N\0\0\0\0".., 452)     = 452
close(5)                                        = 0
time()                                          = 1046310015
time()                                          = 1046310015
time()                                          = 1046310015
time()                                          = 1046310015
time()                                          = 1046310015
open("/orca/rrd/rrd/orcallator/orcallator/o_dev3/gauge_1024_X_freememK.rrd",
O_RDONLY) = 5
fstat64(5, 0xFFBEED30)                          = 0
ioctl(5, TCGETA, 0xFFBEECBC)                    Err#25 ENOTTY
read(5, " R R D\0 0 0 0 1\0\0\0\0".., 8192)     = 8192
llseek(5, 0, SEEK_CUR)                          = 8192
lseek(5, 20372, SEEK_SET)                       = 20372
lseek(5, 1172, SEEK_SET)                        = 1172
read(5, " AACB8 (\0\0\0\0 AAD16F0".., 8192)     = 8192
llseek(5, 0xFFFFFFFFFFFFE098, SEEK_CUR)         = 1324
close(5)                                        = 0
    Incurred fault #6, FLTBOUNDS  %pc = 0xFF07291C
      siginfo: SIGSEGV SEGV_MAPERR addr=0xFC5FB848
    Received signal #11, SIGSEGV [default]
      siginfo: SIGSEGV SEGV_MAPERR addr=0xFC5FB848
        *** process killed ***


I realize orca.state is not created, and that it is supposed to get created
on first start, but
orca.state Never gets created.  

HERE is what I run on:

root at dev3:/usr/local# cat /etc/release
                       Solaris 8 10/01 s28s_u6wos_08a SPARC
           Copyright 2001 Sun Microsystems, Inc.  All Rights Reserved.
                           Assembled 12 September 2001

Generic_108528-16 sun4u sparc SUNW,Ultra-5_10