[rc5] Statistics Problem

Rebecca and Rowland rebecca at astrid.u-net.com
Tue Jul 15 23:52:13 EDT 1997


>Date: Mon, 14 Jul 1997 18:06:38 -0500 (CDT)
>From: Stephen Langasek <vorlon at dodds.net>
>Subject: Re: [rc5] Statistics Problem
>
>On Mon, 14 Jul 1997, Rebecca and Rowland wrote:
>
>> I found that my Performa 475 running the v2 client on a dial-up link wasn't
>> reporting its blocks back.  I've emailled this list and the Mac developer
>> about the problem, but I've not had a response on what the problem is or
>> how to fix it.
>>
>> I've stopped running the client - clearly, there's no point until this
>> problem is fixed.
>
>I, for one, would be more than willing to help you solve whatever problems
>you're having with the client, if only to reduce the number of repetitive,
>complaint-filled messages I see piling up in my inbox.  But I don't have
>access to any Macs (at least not any running MacOS), and if your claims
>about the lack of documentation are to be believed, then the new Mac
>client must have no features whatsoever that I would recognize from my
>experience with the windows and unix versions.  If this were any of the
>other clients, a simple flush would quickly reveal whether the client was
>having problems getting to the server.

Flushing is not possible with the v2 Mac client - the Mac client does
indeed lack the features of the other v2 clients, which is why it's so
awkward to either find out what it's doing, or affect its behaviour in any
way.

>
>Moreover, in all your posts to the mailing list, the one thing I haven't
>seen yet is the .ini file

No such animal under MacOS.

>-- or other description of how you've configured
>the client -- that Duncan has requested be sent with any bug report.

I've included all the relevant information I can think of - my
configuration took the form of entering my email address and telling the
client to fetch 4 blocks at a time.  What other configuration is possible?

> Even
>though your posts have since degenerated from a bug report to habitual
>whining, this still applies to you.  From what you've said, it wasn't even
>clear whether you were buffering keys,

What does that phrase mean exactly?  I can see two buffer files - which one
are you referring to?

>or whether you had even processed
>enough keyblocks for it to be *time* for the client to send the completed
>blocks back to the server--the simplest and (as anyone who has ever worked
>tech support will tell you) most likely answer.

I've checked everything I can, but I don't know how to tell how many blocks
the client's processed, or how many blocks are in either the input buffer
or the output buffer.  There is nothing in the documentation to tell me how
to do this.

>
>My guess is, your willingness to give up on the client -- and your
>UNwillingness to make any effort to diagnose the problem yourself

I have tried very hard to solve the problem - I have read all the
documentation I could find, and tried running the client under various
circumstances, and after some days of fiddling, I managed to conclude that
the client hung if it couldn't connect after finishing checking all input
buffered blocks.  I have no way of telling what the clients doing - it
appeared to be checking random blocks after I restarted it - there seemed
to be no communication with a keyserver, yet it began checking blocks.
There is no documentation on how to interpret the log file, so I can't be
sure exactly what it's doing.

Given that I can't tell the client to flush the output buffer, it seems
that there's nothing to be done about this whole mess.  I have asked for
explanations of how to check whether or not my client has actually sent
back its checked blocks (which isn't documented).  Having discovered that
my client *wasn't* doing this, I emailled the developers with a clear bug
report and log file.  I was ignored.  I mailed the list, wondering if
anyone had any ideas what could be done - given that the Mac client can't
do what the other v2 clients can do, and is virtually unconfigurable (I've
actually been told that it works fine out of the box and there's no need to
know what it's doing - clearly a false statement, since it's not reporting
blocks back on my Mac and something needs to be done to deal with this
problem).

>-- is
>probably the chief reason no one's stepped forward to help you.  This
>isn't Microsoft; nobody here is getting paid to cheerfully walk spoiled
>users through the installation process.

I'm not sure on what basis you feel you can refer to me as `a spoilt user'
- you haven't a clue what I'm like, or how much effort I've put in to
trying to figure out what's going on.  I have, in any case, had more help
that I'd expect from Microsoft.

As for being walked through the installation process - well, that's easy.
The problem is that the software doesn't work as specified, nor was there
any apparent way round the problem.

Please note that several people have emailled the list suffering similar
problems to me - I am not a lone whinger, just a loud one.

Rowland.



----
To unsubscribe, send email to majordomo at llamas.net with 'unsubscribe rc5' in the body.



More information about the rc5 mailing list