Re: Question about message numbers in news groups
Junior
Anonymous wrote:
> Anonymous wrote:
>
>> When the message numbers in a news group is determined to be
>> getting too high, does the ng admin simply reset it to zero?
>
> Those numbers... article numbers you can see in Xref headers for
> example... are generated by individual news servers. They're
> internal to that server, not specific to a news group globally.
>
> It's "rude" for a news server admin to reset article numbers
> because it messes up most news clients (they keep track of what
> you've already read using article numbers), and news servers "peer"
> by using them to move new articles about, but it can and *has*
> happened. Suddenly you see no traffic in the groups that were
> reset, and typically you have to do something like manually
> download X number of headers in each group to "reset" your article
> numbers so they sync with the server again.
For the INN server the number will max out at 9,999,999,999 so some
high volumn groups could run into that after several years. There
was some talk a few months back about the boneless newsgroup and a
server admin who had to deal with that. I don't remember if it was
mentioned what actually had to be done to fix that group.
Ray Banana
> For the INN server the number will max out at 9,999,999,999 so some
> high volumn groups could run into that after several years. There
> was some talk a few months back about the boneless newsgroup and a
> server admin who had to deal with that. I don't remember if it was
> mentioned what actually had to be done to fix that group.
http://www.giganews.com/news/article/64-bit-Usenet.html
--
Too many ingredients in the soup, no room for a spoon
http://www.eternal-september.org
Julien ÉLIE
> For the INN server the number will max out at 9,999,999,999 so some
> high volumn groups could run into that after several years.
2^31-1 in fact, that is to say 2,147,483,647 for INN (and probably
also some other news servers).
The fact that the active file uses 10 digits does not mean that
it can read 10^11-1.
According to RFC 3977:
Article numbers MUST lie between 1 and 2,147,483,647, inclusive. The
client and server MAY use leading zeroes in specifying article
numbers but MUST NOT use more than 16 digits. In some situations,
the value zero replaces an article number to show some special
situation.
Note that it is likely that the article number limit of 2,147,483,647
will be increased by a future revision or extension to this
specification. While servers MUST NOT send article numbers greater
than this current limit, client and server developers are advised to
use internal structures and datatypes capable of handling larger
values in anticipation of such a change.
--
Julien �LIE
� C'est la goutte d'eau qui fait d�border le vase et qui met
le feu aux poudres. �
Stray Cat
From http://en.wikipedia.org/wiki/Alt.binaries.boneless :
"On March 31, 2009, NewsBin sent out an alert to its users that the
alt.binaries.boneless post count is the first to break the 32-bit
threshold, having over 4,294,967,296 posts. Until then, all
newsgroups' post totals could be stored as a 32-bit number. This
occurrence might cause compatibility issues with binary news readers
that cannot address post counts stored as a 64-bit number."
Julien ÉLIE
>> For the INN server the number will max out at 9,999,999,999 so some
>> high volumn groups could run into that after several years.
>
> 2^31-1 in fact, that is to say 2,147,483,647 for INN (and probably
> also some other news servers).
> The fact that the active file uses 10 digits does not mean that
> it can read 10^11-1.
Of course, it was to be read 10^10-1.
--
Julien �LIE
� Les id�es fausses ne sont pas toujours mauvaises. � (Marcel Aym�)
Junior
> Correcting a miscount I did...
>
>>> For the INN server the number will max out at 9,999,999,999 so some
>>> high volumn groups could run into that after several years.
>>
>> 2^31-1 in fact, that is to say 2,147,483,647 for INN (and probably
>> also some other news servers).
>> The fact that the active file uses 10 digits does not mean that
>> it can read 10^11-1.
>
> Of course, it was to be read 10^10-1.
I miss Carl Billions N Billions Sagan too. ;)