Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 8 Jan 2016 17:02:36 +0200
From:      Daniel Braniss <danny@cs.huji.ac.il>
To:        Eugene Grosbein <eugen@grosbein.net>
Cc:        Freebsd hackers list <freebsd-hackers@freebsd.org>
Subject:   Re: watchdog not rebooting
Message-ID:  <C7B88BA6-6958-4760-B742-55FBC6F47562@cs.huji.ac.il>
In-Reply-To: <568F99E0.5080008@grosbein.net>
References:  <95D6F157-70B1-48F3-B512-9A4B7D6F9EA4@cs.huji.ac.il> <568F9916.2000802@grosbein.net> <568F99E0.5080008@grosbein.net>

next in thread | previous in thread | raw e-mail | index | archive | help

> On 8 Jan 2016, at 1:13 PM, Eugene Grosbein <eugen@grosbein.net> wrote:
>=20
>>> on real hardware, with hardware watchdog, it does the right thing, =
panics and reboots,
>>> but on vmware, it enters the debugger, and =E2=80=98cont=E2=80=99 =
will resume.
>>> the kernel in both cases(it=E2=80=99s the same) has option KDB and =
ALT_BREAK_TO_DEBUGGER
>>>=20
>>> so how can I get the vm to reboot?
>>=20
>> debug.debugger_on_panic=3D0 in the /boot/loader.conf and/or options =
KDB_UNATTENDED in the kernel config.
>=20
> Hmm, no - in case of watchdog, only kernel options KDB_UNATTENDED will =
help,
> debug.debugger_on_panic will not.
>=20
I just tried an older kernel, 10.1, and it works as expected, with =
debugger_on_panic =3D 0, it reboots.
so something changed.  I actually tried this on different hardware, and =
so far it fails on 10.2
and works ok on 10.1=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?C7B88BA6-6958-4760-B742-55FBC6F47562>