) On Tue, May 05, 1998 at 12:45:53PM +0200, Dave Rynne wrote:
) >
) > Has anyone else out there had similar problems and if so, what's the best
) > work around?
)
) Something like this occured on one of our machines.
)
) The mainboard turned out to be faulty. You should try to compile
) the Linux kernel e.g. 10 times - if the compile process stops
) with "signal 11", your hardware is to blame.
Am aware of this problem but have never seen it on this box. Am currently
running a process to do 100 kernel compiles and log each of them to a
seperate file. This should detect the linux sig11 problem if this box
has it. It probably won't be finished before end of work today but I'll
post if this is the problem.
)
) At our site Exim mostly crashed when forking. The child process
) that was launched to send a warning message to the sender after
) several days died.
)
) By that, messages never got bounced and stayed in the queue.
)
) You might mail the result of
)
) "strace -f exim -M message-id"
Tried this for a few of them. A couple get this far with most dying
before here.
open("/etc/resolv.conf", O_RDONLY) = 7
fstat(7, {st_mode=S_IFREG|0644, st_size=152, ...}) = 0
mmap(0, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x400e0000
read(7, "domain esat.net\nsearch"..., 4096) = 152
read(7, "", 4096) = 0
close(7) = 0
munmap(0x400e0000, 4096) = 0
socket(PF_INET, SOCK_DGRAM, IPPROTO_IP) = 7
connect(7, {sin_family=AF_INET, sin_port=htons(53),
sin_addr=inet_addr("192.111.39.1")}, 16) = 0
send(7, "\373\303\1\0\0\1\0\0\0\0\0\0\7ho"..., 28, 0) = 28
oldselect(8, [7], NULL, NULL, {5, 0}) = 1 (in [7], left {5, 0})
recvfrom(7, "\373\303\205\200\0\1\0\5\0\2\0\7"..., 1024, 0,
{sin_family=AF_INET, sin_port=htons(53), sin_addr=inet_addr("192.111.39.1")},
[16]) = 306
close(7) = 0
getpid() = 32337
time(NULL) = 894377398
socket(PF_INET, SOCK_DGRAM, IPPROTO_IP) = 7
ioctl(7, SIOCGIFCONF, 0xbfffd244) = 0
ioctl(7, SIOCGIFFLAGS, 0xbffff18c) = 0
ioctl(7, SIOCGIFADDR, 0xbffff18c) = 0
ioctl(7, SIOCGIFFLAGS, 0xbffff18c) = 0
ioctl(7, SIOCGIFADDR, 0xbffff18c) = 0
close(7) = 0
--- SIGSEGV (Segmentation fault) ---
+++ killed by SIGSEGV +++
)
) to me - however, another mystique about this phenomena was, that
) the crash sometimes did not appear when you called exim by strace.
)
) (This is a bit similiar to your findings).
)
)
) Greetings,
)
)
) Georg
)
) --
) *** Exim information can be found at
http://www.exim.org/ ***
)
)
Dave
--
*** Exim information can be found at
http://www.exim.org/ ***
