[PROXYPER] Options for version 314?

Dan Oetting oetting at gldmutt.cr.usgs.gov
Mon Feb 14 21:25:18 EST 2000


At 10:49 -0700 2/14/2000, Jeff Lawson wrote:
>On Mon, 14 Feb 2000, Phillip wrote:
>
>> Before the last time (about a week or so ago), I had in place a sytem where
>> I had feeder PProxies and I rotate the in buffers so there's no stale
>>blocks.
>> Of course, I can't speak for other people, but I'd personally prefer the
>>idea
>> of my clients' clients being able to flush their blocks in case of a
>>prolonged
>> problem with the main keyservers again.
>
>Well, you still have not given a reasonable rationale for why flushing to
>an empty proxy might be preferable, other than a personal preference.

I'll throw in one reason: Diskless clients. Though you could still argue
that letting my desktop machines run cool and quiet over night is a
personal preference.

>One might argue that "distributed storage" (by allowing the clients to
>continue to hold their work until they have a verified operational
>upstream proxy) would even be more reliable, by eliminating single point
>of failure.

A proxy that is unable to unload blocks upstream is not a failure unless it
never gets corrected.

-- Dan Oetting <oetting at ghtmail.cr.usgs.gov>


--
To unsubscribe, send 'unsubscribe proxyper' to majordomo at lists.distributed.net



More information about the proxyper mailing list