From: Peter Staubach <staubach@redhat.com>
Subject: Re: [PATCH] nfsd: fail module init on reply cache init	failure
Date: Fri, 16 Nov 2007 10:38:47 -0500
Message-ID: <473DB987.7090807@redhat.com>
References: <1195163823-24609-1-git-send-email-bfields@citi.umich.edu>
	<1195163823-24609-2-git-send-email-bfields@citi.umich.edu>
	<1195163823-24609-3-git-send-email-bfields@citi.umich.edu>
	<1195163823-24609-4-git-send-email-bfields@citi.umich.edu>
	<473DA941.4050101@redhat.com> <20071116153036.GG31958@fieldses.org>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Cc: Neil Brown <neilb@suse.de>, linux-nfs@vger.kernel.org,
	nfs@lists.sourceforge.net
To: "J. Bruce Fields" <bfields@fieldses.org>
In-Reply-To: <20071116153036.GG31958@fieldses.org>
Sender: nfs-bounces@lists.sourceforge.net
Errors-To: nfs-bounces@lists.sourceforge.net

J. Bruce Fields wrote:
> On Fri, Nov 16, 2007 at 09:29:21AM -0500, Peter Staubach wrote:
>   
>> J. Bruce Fields wrote:
>>     
>>> +	printk(KERN_ERR "nfsd: failed to allocate reply cache\n");
>>>   
>>>       
>> I was thinking that it might be nice to have something which
>> better explains what the ramifications of this failure might
>> be.  This explains _precisely_ what happened, but not what
>> will happen in the future.
>>     
>
> The module will fail to load (or, I suppose, the kernel will fail to
> boot?).  So the failure will be pretty obvious.
>
>   

The module will fail to load, but I suspect that that won't cause
the system to fail to boot.  The only way to notice that the module
didn't load is to run lsmod or some such and to look for the module.

The admin may notice that the NFS server fails to start, but I
think that it would be nice to better connect this memory allocation
failure to the NFS server not running.

>> Or can we get this documented in some fashion that admins
>> will know how to find and translate to what they need to
>> understand and do?
>>     
>
> Would a more explicit statement of the cause (like "nfsd: out of memory
> setting up reply cache") be helpful?
>
>   

No, I don't think that this adds anything that one couldn't reason
out of the first proposed message.

> Obviously this should only happen in a rather extreme low-memory
> condition.
>   

Yes, and may never happen in nature, but if we are preparing
for the possibility, we may as well make it as easy for the
unsuspecting admin to figure out.

Otherwise, the message is only useful for someone who understands
the implementation and what the ramification of this failure
might be.

    Thanx...

       ps

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs
_______________________________________________
Please note that nfs@lists.sourceforge.net is being discontinued.
Please subscribe to linux-nfs@vger.kernel.org instead.
    http://vger.kernel.org/vger-lists.html#linux-nfs