On Fri, Apr 04, 2003 at 01:56:13PM -0500, Brian Cuttler wrote:
>
> Ok, getting smarter...
>
> following your suggestion
>
> # truss amtape ninfo newtonl
>
> Tail of the output (I can script if if you want more)
>
> amtapewrite(2, " a m t a p e", 6) = 6
> : scanning all write(2, " : s c a n n i n g a".., 15) = 15
> 9write(2, " 9", 1) = 1
> slots in tape-changer rack:
> write(2, " s l o t s i n t a".., 29) = 29
> so_socket(PF_UNIX, SOCK_STREAM, 0, "", 1) = 4
> so_socket(PF_UNIX, SOCK_STREAM, 0, "", 1) = 5
> so_socketpair(0xFFBFFB48) = 0
> close(4) = 0
> fork() = 21854
> close(5) = 0
> read(6, " 2 / d e v / r m t / 1".., 1024) = 15
> wait() = 21854 [0x0000]
> close(6) = 0
> write(10, " c h a n g e r : g o t".., 41) = 41
> open64("/dev/rmt/1cn", O_RDONLY) = 4
> brk(0x0002FEC8) = 0
> brk(0x00035EC8) = 0
> ioctl(4, MNTIOC_NMNTS, 0xFFBFF258) = 0
> read(4, " A M A N D A : T A P E".., 32768) = 32768
> close(4) = 0
> slot write(2, " s l o t ", 5) = 5
> 2write(2, " 2", 1) = 1
> : date write(2, " : d a t e ", 7) = 7
> 20030403write(2, " 2 0 0 3 0 4 0 3", 8) = 8
> label write(2, " l a b e l ", 7) = 7
> NEWTONL02write(2, " N E W T O N L 0 2", 9) = 9
>
> write(2, "\n", 1) = 1
> write(10, " c h a n g e r _ l a b e".., 36) = 36
> so_socket(PF_UNIX, SOCK_STREAM, 0, "", 1) = 4
> so_socket(PF_UNIX, SOCK_STREAM, 0, "", 1) = 5
> so_socketpair(0xFFBFFA50) = 0
> close(4) = 0
> fork() = 21962
> close(5) = 0
> read(6, " 2 9 1 1\n", 1024) = 8
> wait() = 21962 [0x0000]
> close(6) = 0
> write(10, " c h a n g e r : g o t".., 34) = 34
> write(10, " c h a n g e r _ q u e r".., 40) = 40
> write(10, " c h a n g e r _ q u e r".., 30) = 30
> write(10, " c h a n g e r _ l a b e".., 48) = 48
> so_socket(PF_UNIX, SOCK_STREAM, 0, "", 1) = 4
> so_socket(PF_UNIX, SOCK_STREAM, 0, "", 1) = 5
> so_socketpair(0xFFBFFAC8) = 0
> close(4) = 0
> fork() = 22058
> close(5) = 0
> read(6, " 2 / d e v / r m t / 1".., 1024) = 15
> wait() = 22058 [0x0000]
> close(6) = 0
> write(10, " c h a n g e r : g o t".., 41) = 41
> Incurred fault #5, FLTACCESS %pc = 0xFF0C37F4
> siginfo: SIGBUS BUS_ADRALN addr=0x000248F2
> Received signal #10, SIGBUS [default]
> siginfo: SIGBUS BUS_ADRALN addr=0x000248F2
>
>
OK, of course that is only system calls. There is probably
a "printf" in the .c file that corresponds to that write.
It is writing to file descriptor 10, you can find above an
"open" that returns a 10 ( = 10) to see the file name but
it probably is /tmp/amanda/???debug???.
The write succeeded, tried to write 41 bytes and the return
( = 41) says it did so successfully.
So, the trick now is to find the C code that caused the write
and figure what happened next. That is likely where the occured.
--
Jon H. LaBadie jon AT jgcomp DOT com
JG Computing
4455 Province Line Road (609) 252-0159
Princeton, NJ 08540-4322 (609) 683-7220 (fax)
|
