bug-hurd
[Top][All Lists]
Advanced

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

Re: Panic: zalloc!


From: Jon Arney
Subject: Re: Panic: zalloc!
Date: Thu, 28 Feb 2002 04:51:54 -0700

> Indeed.  There are some bugs, either resource leaks in the Hurd system
> or in the kernel (or even some other bug).  We are currently lacking data
> to give a good interpretation for the zalloc problems.

Good to know it's not just me!

> 
> When we use oskit-mach we can get better data with gdb over serial line.

If you will be switching uKernels, does this mean that the zalloc
problem
is not worth investigating since it will probably go away with gnumach,
or am I missing something.  Sorry if I'm being thick-headed.

> 
> > Going to the physical console, I noticed a message 'panic: zalloc....'
> > and before I could copy it down, the box re-booted on me.
> 
> As it should.  The binary packages are what comes closest to a
> "production system, and for such systems rebooting is a much better behaviour
> than spinning or halting.  Imagine a web server, or a remote-adminstrated
> machine.

I would tend to disagree somewhat.  Firstly, I would like to point out
as
a sometimes system administrator that there is nothing more
disconcerting
than an unexplained system outage.  I would rather go on-site to
discover
a problem than to have a system mysteriously reboot itself leaving NO
trace of what transpired (or how to prevent it in future).  While I
agree
that a production system should reboot itself, I would also assert that
a
production system should leave some trace of events which transpired
such as the core data you describe.

This does apply, by the way, to hurd translators as well.  If fatal or
serious translator conditions are detected, by assert() or some other
mechanism, I would like to see that some facility exist for the
collection and distribution of such information.  Think of syslogd
for translators or something like Linux's klogd if you would like
some prescedent.

If some form of logging facility/crash dump is possible
(and indeed this may be problematic for a kernel level event),
I would like to see work done in this area.  I am not opposed
to contributing to such efforts.  In fact, for my own purposes, I
have created a 'hurdlog' translator for collecting messages in
some of my 'toy' translators.

Ok. Rant ends. :)


Thanks,

Jonathan S. Arney
Software Engineer
jarney1@cox.net
------------------------------------------------------------------------



reply via email to

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