[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

exec server / fork bug

From: Neal H Walfield
Subject: exec server / fork bug
Date: Sat, 16 Jun 2001 12:50:14 -0500
User-agent: Mutt/1.3.18i

While compiling the C library, I got the following:

        make: *** make[1]: *** wait: Computer bought the farmmake: 
../sysdeps/mach/hurd/fork.c:469: __fork: Unexpected error: (ipc/mig)
        server died.
        wait: Computer bought the farm..  Stop.
        make[1]: *** Waiting for unfinished jobs....make: *** Waiting for 
unfinished jobs....

        make: *** make[1]: *** wait: Computer bought the farm.wait: Computer 
bought the farm  Stop.
        .  Stop.

In another terminal, I was running a loop that would execute vmstat
every 5 seconds.  Here is the last report:

        Sat Jun 16 12:16:33 EDT 2001
        pagesize:          4K
        size:            375M
        free:            299M
        active:         35.3M
        inactive:       31.9M
        wired:          8.34M
        zero filled:    14.3G
        reactivated:       0 
        pageins:        1.11G
        pageouts:       1.25G
        page faults: 27092762
        cow faults:   3262672
        memobj hit ratio: 97%
        swap size:       954M
        swap free:       954M

As can be seen from the output, there is plenty of ram available (in
fact, in the last hour, there were consistently 295-305 MB free).

In yet another terminal, I attached gdb to exec to see if I could get a
backtrace before a crash:

        neal@desdemona:~ (0)$ sudo gdb /hurd/exec 5
        GNU gdb 5.0
        [Switching to thread 5.1]
        (gdb) c
        warning: Can't wait for pid 5: No child processes

Nothing came up.  Am I doing this wrong (e.g. does init completely kill
exec and reboot the machine before gdb gets a chance to see the signal)?

Attachment: pgp0jXib4KbpG.pgp
Description: PGP signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]