[OpenAFS] Windows cache rehashed...

[Jeffrey Altman]

This is a cryptographically signed message in MIME format.

--------------ms060703020106040702010301
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Rodney:

We have been talking about this for several weeks now.  When you 
originally brought this
up I asked you to report on the amount of RAM and number of handles 
which were allocated
under different scenarios.  You gave us a report which indicated that 
when the AFSCache
was filled the memory allocated to afsd_service.exe was slightly above 
the size of the
cache file and that the number of handles were equal to <cache size> / 
<block size> plus
a few.  The e-mail was dated 12/1 and the actual numbers were 209,608K 
and 140 handles.

You also provided a scenario which I tested on my systems using the 
daily builds.  The
behavior was in-line with that which you reported.  If you have new data 
to report,
please provide scenarios and numbers which can be replicated in a 
controlled environment.
Please add this data to the Request (#2628).  Be sure to specify exactly 
which build
of OpenAFS you tested against. (Preferably something very recent due to 
all of the memory
and handle allocation bugs which have been recently fixed.)

If there is a bug I would like to find it, if there is not I will 
explain to you why the
behavior is what it is. 

Jeffrey Altman


Rodney M Dyer wrote:

> Jeffrey,
>
> Ah, tedious to the end eh?  No, you are incorrect.  The only error 
> made here is that the units that I used in my previous email where 
> wrong.  Everywhere I used "Meg", I should have said "K".  Using that 
> reasoning...
>
> 8192 = 8.192 Meg
>
> So,
>
> 8192K / 32K = 256 Handles...Not many at all.
>
> So please explain why the handle count rises into the multiple thousands?
>
> If you had even tried to look at my mail a little closer you should 
> have realized my mistake.  An 8192 Meg cache would be 8 Gig.  It 
> should be obvious to anyone who has used the Windows version that you 
> can't make a cache over 2 Gig in size.  So I was obviously intending 
> the 8 Meg cache.  You could have just asked.  We've only been 
> discussing this for a couple weeks now, you should have seen that I 
> was meaning a smaller cache than normal.
>
> The bug still stands, unless you can reason a way out of this...
>
> Rodney
>

--------------ms060703020106040702010301
Content-Type: application/x-pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
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--------------ms060703020106040702010301--