list archives?
The Hermit Hacker
scrappy at hub.org
Thu Sep 9 04:39:41 UTC 1999
Okay...makehistory is running righ tnow, and inndf -o reports 8% used of
~1.5gig...now just have to wait for this to finish, but what does the
following mean?
buffindexed: all buffers are brandnew, unlink
'/usr/local/news/admin/db/group.index'
I understand the 'brandnew' part, but until I started makehistory, the
~/admin/db/group.index file didn't exist...why does it appear to be
telling me to unlink it, if it is what created it? *raised eyebrow*
On Thu, 9 Sep 1999, Katsuhiro Kondou wrote:
> In article <Pine.BSF.4.10.9909090054280.412-100000 at thelab.hub.org>,
> The Hermit Hacker <scrappy at hub.org> wrote;
>
> } makehistory -O -I -T /var/spool/news/tmp
> }
> } to get things back up and running?
>
> Right.
>
> } The part that I'm still confused about...the writes to the buffer are
> } sequential, right? Now, I fill the buffer, and it starts from the
>
> Partly yes. Buffindexed has its own block allocation algorithm
> (not cyclical) and blocks are allocated every 32 blocks sequentially.
> OVadd(buffindexed_add) will fail if the buffer is filled up.
> OVexpiregroup(buffindexed_expiregroup) which is called by
> expireover allocaltes blocks, recreates new overview data and
> index from existing one and frees old blocks.
>
> Allocated block belongs to each newsgroup not each overview
> data. So several overview data (of course, they are the same
> newsgroup) will be written into one block in most case.
> --
> Katsuhiro Kondou
>
Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy at hub.org secondary: scrappy@{freebsd|postgresql}.org
More information about the inn-workers
mailing list