Nothing larger then int8?

hrrmm ... ignore this ... I'm suspecting that what I did was copied in
sum() data from an old table that had bytes declared as int4, without
casting it to int8 before storing it to the new table ...

if anyone is interested, here is one days worth of http traffic for the
main PostgreSQL.Org server ... this doesn't include the traffic that the
mirror sites absorb:

1160643846 / ( 1024 * 1024 * 1024 )
1.08gig

On Thu, 18 Jan 2001, The Hermit Hacker wrote:

I'm logging traffic to a database, so that I can do analysis on usage and
whatnot, and I need something bigger then int8 :(

/tmp/psql.edit.70.79087: 6 lines, 222 characters.
ip | maxbytes | port | runtime
---------------+-------------+------+------------------------
216.126.84.28 | 2169898055 | 80 | 2001-01-16 00:00:00-05
216.126.84.28 | 160579228 | 873 | 2001-01-16 00:00:00-05
216.126.84.28 | 365270 | 20 | 2001-01-16 00:00:00-05
216.126.84.28 | 196256 | 21 | 2001-01-16 00:00:00-05
216.126.84.28 | 195238 | 22 | 2001-01-16 00:00:00-05
216.126.84.28 | 182492 | 1024 | 2001-01-16 00:00:00-05
216.126.84.28 | 171155 | 143 | 2001-01-16 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-13 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-04 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-05 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-06 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-07 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-08 00:00:00-05
216.126.84.28 | -1392384544 | 80 | 2001-01-14 00:00:00-05
216.126.84.28 | -1452855018 | 80 | 2001-01-15 00:00:00-05
216.126.84.28 | -1452855018 | 80 | 2001-01-10 00:00:00-05
216.126.84.28 | -1452855018 | 80 | 2001-01-09 00:00:00-05
216.126.84.28 | -1513325492 | 80 | 2001-01-03 00:00:00-05
216.126.84.28 | -1694736914 | 80 | 2001-01-12 00:00:00-05
216.126.84.28 | -1815677862 | 80 | 2001-01-11 00:00:00-05

hub_traf_stats=# \d daily_stats
Table "daily_stats"
Attribute | Type | Modifier
-----------+-----------+----------
ip | inet |
port | integer |
bytes | bigint |
runtime | timestamp |

do we have anything larger to work with? I've checked docs, but that
looks like about it :(

Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org

Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org

The Hermit Hacker <scrappy@hub.org> writes:

I'm logging traffic to a database, so that I can do analysis on usage and
whatnot, and I need something bigger then int8 :(

Those "maxbytes" values shure look like they're only int4. How are
you calculating 'em, exactly?

regards, tom lane

The Hermit Hacker wrote:

if anyone is interested, here is one days worth of http traffic for the
main PostgreSQL.Org server ... this doesn't include the traffic that the
mirror sites absorb:

1160643846 / ( 1024 * 1024 * 1024 )
1.08gig

Not a bad day.

I've seen 100MB per day out of my http (backed by PostgreSQL since late
1997!), but the 2.5GB a day out the RealServer is the big hit....
--
Lamar Owen
WGCR Internet Radio
1 Peter 4:11

On Thu, 18 Jan 2001, Lamar Owen wrote:

The Hermit Hacker wrote:

if anyone is interested, here is one days worth of http traffic for the
main PostgreSQL.Org server ... this doesn't include the traffic that the
mirror sites absorb:

1160643846 / ( 1024 * 1024 * 1024 )
1.08gig

Not a bad day.

I've seen 100MB per day out of my http (backed by PostgreSQL since late
1997!), but the 2.5GB a day out the RealServer is the big hit....

my *big* site:

11395533772/ ( 1024 * 1024 * 1024 )
10.61gig/day :)

bytes | port
-------------+------
11298475398 | 81
94925095 | 80
1982130 | 20
122043 | 21
26766 | 22
2340 | 137

just a small site :)

To answer your question, wouldn't numeric(30,0) be the correct?

-alex
On Thu, 18 Jan 2001, The Hermit Hacker wrote: