begin quote from Kevin Gagel <[EMAIL PROTECTED]> written 2002-09-24:
Here's a sampling from my logs:
13:10:53 filter1 spamd[29471]: clean message (0/8) for (unknown):100 in 0
seconds.
Sep 3 13:10:59 filter1 spamd[29472]: clean message (0/8) for (unknown):100 in
0 seconds.
Sep 3 13:11:10 filter1 spamd[29473]: identified spam (8/8) for (unknown):100
in 1 seconds.
Sep 3 13:11:59 filter1 spamd[29475]: clean message (1/8) for (unknown):100 in
0 seconds.
Sep 3 13:12:02 filter1 spamd[29474]: clean message (5/8) for (unknown):100 in
6 seconds.
Sep 3 13:12:25 filter1 spamd[29476]: identified spam (15/8) for (unknown):100
in 1 seconds.
Sep 3 13:12:42 filter1 spamd[29477]: identified spam (11/8) for (unknown):100
in 1 seconds.
Sep 3 13:13:10 filter1 spamd[29478]: identified spam (11/8) for (unknown):100
in 0 seconds.
Sep 3 13:13:18 filter1 spamd[29479]: identified spam (13/8) for (unknown):100
in 1 seconds.
Sep 3 13:13:33 filter1 spamd[29480]: clean message (5/8) for (unknown):100 in
1 seconds.
Sep 3 13:13:36 filter1 spamd[29481]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:13:40 filter1 spamd[29482]: clean message (6/8) for (unknown):100 in
1 seconds.
Sep 3 13:13:45 filter1 spamd[29483]: identified spam (15/8) for (unknown):100
in 0 seconds.
Sep 3 13:14:23 filter1 spamd[29484]: identified spam (13/8) for (unknown):100
in 0 seconds.
Sep 3 13:14:58 filter1 spamd[29485]: clean message (1/8) for (unknown):100 in
1 seconds.
Sep 3 13:14:58 filter1 spamd[29486]: identified spam (14/8) for (unknown):100
in 0 seconds.
Sep 3 13:15:05 filter1 spamd[29489]: identified spam (15/8) for (unknown):100
in 0 seconds.
Sep 3 13:15:25 filter1 spamd[29490]: identified spam (15/8) for (unknown):100
in 1 seconds.
Sep 3 13:15:33 filter1 spamd[29491]: identified spam (14/8) for (unknown):100
in 7 seconds.
Sep 3 13:16:14 filter1 spamd[29492]: clean message (3/8) for (unknown):100 in
1 seconds.
Sep 3 13:16:15 filter1 spamd[29493]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:16:17 filter1 spamd[29494]: clean message (1/8) for (unknown):100 in
1 seconds.
Sep 3 13:16:30 filter1 spamd[29495]: clean message (0/8) for (unknown):100 in
0 seconds.
Sep 3 13:16:49 filter1 spamd[29496]: clean message (2/8) for (unknown):100 in
1 seconds.
Sep 3 13:17:08 filter1 spamd[29497]: clean message (3/8) for (unknown):100 in
1 seconds.
Sep 3 13:17:10 filter1 spamd[29498]: clean message (2/8) for (unknown):100 in
0 seconds.
Sep 3 13:17:30 filter1 spamd[29499]: clean message (-2/8) for (unknown):100
in 1 seconds.
Sep 3 13:17:35 filter1 spamd[29500]: clean message (2/8) for (unknown):100 in
0 seconds.
Sep 3 13:18:25 filter1 spamd[29501]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:18:25 filter1 spamd[29502]: identified spam (11/8) for (unknown):100
in 0 seconds.
Sep 3 13:18:31 filter1 spamd[29503]: clean message (-88/8) for (unknown):100
in 6 seconds.
Sep 3 13:18:42 filter1 spamd[29505]: identified spam (9/8) for (unknown):100
in 1 seconds.
Sep 3 13:18:45 filter1 spamd[29504]: clean message (-88/8) for (unknown):100
in 5 seconds.
Sep 3 13:18:51 filter1 spamd[29506]: identified spam (15/8) for (unknown):100
in 1 seconds.
Sep 3 13:19:24 filter1 spamd[29507]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:19:34 filter1 spamd[29508]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:19:39 filter1 spamd[29509]: identified spam (8/8) for (unknown):100
in 1 seconds.
Sep 3 13:20:03 filter1 spamd[29512]: identified spam (12/8) for (unknown):100
in 1 seconds.
Sep 3 13:20:10 filter1 spamd[29513]: clean message (4/8) for (unknown):100 in
1 seconds.
Sep 3 13:20:42 filter1 spamd[29515]: identified spam (14/8) for (unknown):100
in 4 seconds.
Sep 3 13:20:48 filter1 spamd[29514]: identified spam (9/8) for (unknown):100
in 12 seconds.
Sep 3 13:20:58 filter1 spamd[29516]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:21:12 filter1 spamd[29517]: clean message (2/8) for (unknown):100 in
1 seconds.
Sep 3 13:21:27 filter1 spamd[29518]: clean message (3/8) for (unknown):100 in
1 seconds.
Sep 3 13:21:56 filter1 spamd[29519]: identified spam (8/8) for (unknown):100
in 1 seconds.
Sep 3 13:22:03 filter1 spamd[29520]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:22:14 filter1 spamd[29521]: clean message (-1/8) for (unknown):100
in 1 seconds.
Sep 3 13:22:16 filter1 spamd[29522]: clean message (-1/8) for (unknown):100
in 1 seconds.
Sep 3 13:23:32 filter1 spamd[29523]: clean message (2/8) for (unknown):100 in
1 seconds.
Sep 3 13:23:44 filter1 spamd[29524]: skipped large message in 0 seconds.
Sep 3 13:23:51 filter1 spamd[29525]: clean message (0/8) for (unknown):100 in
0 seconds.
Sep 3 13:23:59 filter1 spamd[29526]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:24:09 filter1 spamd[29527]: clean message (-1/8) for (unknown):100
in 1 seconds.
Sep 3 13:24:19 filter1 spamd[29528]: identified spam (13/8) for (unknown):100
in 0 seconds.
Sep 3 13:24:32 filter1 spamd[29529]: clean message (2/8) for (unknown):100 in
1 seconds.
Sep 3 13:24:33 filter1 spamd[29530]: identified spam (9/8) for (unknown):100
in 1 seconds.
Sep 3 13:24:34 filter1 spamd[29531]: clean message (5/8) for (unknown):100 in
0 seconds.
Sep 3 13:24:36 filter1 spamd[29532]: clean message (2/8) for (unknown):100 in
1 seconds.
Sep 3 13:24:43 filter1 spamd[29533]: clean message (4/8) for (unknown):100 in
1 seconds.
Sep 3 13:24:52 filter1 spamd[29534]: clean message (1/8) for (unknown):100 in
1 seconds.
Sep 3 13:24:59 filter1 spamd[29535]: clean message (0/8) for (unknown):100 in
1 seconds.
Sep 3 13:25:23 filter1 spamd[29538]: clean message (4/8) for (unknown):100 in
2 seconds.
Sep 3 13:25:36 filter1 spamd[29539]: clean message (1/8) for (unknown):100 in
4 seconds.
I'm not using razor:
grep "^Sep" /var/log/spamdlog | grep "seconds" | awk '{print $12}' | wc -l
206068 messages
grep "^Sep" /var/log/spamdlog | grep "seconds" | awk '{print $12}' | sort -rn | uniq
-c | sort -rn
# msgs|seconds
114550 1
42415 0
15978 6
10398 5
7556 2
4532 4
1857 3
1806 11
834 10
770 7
666 9
494 35
424 8
306 16
276 12
273 15
219 65
213 14
138 21
124 13
66 95
65 36
65 24
60 66
53 20
52 25
50 19
45 30
40 17
35 41
29 18
27 22
22 28
21 96
21 31
21 29
19 23
16 34
15 44
15 26
14 50
14 33
14 27
11 45
10 61
10 51
10 40
9 39
9 32
8 64
8 48
8 43
7 82
7 68
7 62
6 70
6 57
6 52
6 49
5 97
5 86
5 67
5 60
5 56
5 53
5 46
5 42
4 81
4 55
3 83
3 80
3 37
3 124
3 115
2 98
2 94
2 90
2 87
2 85
2 84
2 78
2 77
2 75
2 73
2 71
2 63
2 54
2 47
2 38
2 184
2 121
2 117
2 104
2 103
2 101
2 100
1 99
1 91
1 89
1 88
1 76
1 74
1 72
1 69
1 59
1 58
1 337
1 316
1 313
1 312
1 302
1 293
1 283
1 273
1 217
1 212
1 206
1 187
1 185
1 182
1 175
1 157
1 151
1 150
1 144
1 142
1 139
1 137
1 134
1 133
1 122
1 120
1 116
1 113
1 112
1 109
1 108
1 106
1 105
1 102
> Here is a random sampling from my server. One of the highest ones I saw in
> skimming over two days worth was 47 seconds and one of the lowest I saw was 12
> seconds.
> Sep 24 10:30:03 spam spamd[9868]: clean message (1.9/5.0) for root:99 in 27
> seconds, 4310 bytes.
> Sep 24 10:30:24 spam spamd[9885]: clean message (4.5/5.0) for root:99 in 24
> seconds, 479 bytes.
> Sep 24 10:31:10 spam spamd[9893]: clean message (3.4/5.0) for root:99 in 26
> seconds, 4868 bytes.
> Sep 24 10:31:16 spam spamd[9899]: clean message (4.8/5.0) for root:99 in 21
> seconds, 26161 bytes.
> Sep 24 10:31:23 spam spamd[9905]: clean message (2.6/5.0) for root:99 in 25
> seconds, 62680 bytes.
> Sep 24 10:31:28 spam spamd[9915]: identified spam (19.2/5.0) for root:99 in 26
> seconds, 1939 bytes.
> Sep 24 10:31:30 spam spamd[9917]: identified spam (15.6/5.0) for root:99 in 28
> seconds, 1353 bytes.
> Sep 24 10:32:02 spam spamd[9978]: clean message (-0.5/5.0) for root:99 in 22
> seconds, 4047 bytes.
> Sep 24 10:32:12 spam spamd[9984]: clean message (0.2/5.0) for root:99 in 25
> seconds, 1470 bytes.
> Sep 24 10:32:19 spam spamd[9990]: identified spam (7.1/5.0) for root:99 in 26
> seconds, 1754 bytes.
> Sep 24 10:32:52 spam spamd[9999]: identified spam (34.5/5.0) for root:99 in 25
> seconds, 2116 bytes.
> Sep 24 10:32:53 spam spamd[10005]: identified spam (11.5/5.0) for root:99 in 26
> seconds, 5274 bytes.
> Sep 24 10:33:00 spam spamd[10011]: identified spam (15.8/5.0) for root:99 in 27
> seconds, 25666 bytes.
> Sep 24 10:33:15 spam spamd[10017]: clean message (-2.8/5.0) for root:99 in 25
> seconds, 2909 bytes.
> Sep 24 10:33:21 spam spamd[10025]: clean message (-0.2/5.0) for root:99 in 26
> seconds, 3942 bytes.
> Sep 24 10:33:25 spam spamd[10032]: identified spam (15.8/5.0) for root:99 in 25
> seconds, 25670 bytes.
> Sep 24 10:33:35 spam spamd[10038]: identified spam (8.7/5.0) for root:99 in 22
> seconds, 45975 bytes.
> Sep 24 10:33:54 spam spamd[10047]: identified spam (15.8/5.0) for root:99 in 29
> seconds, 25667 bytes.
> Sep 24 10:34:00 spam spamd[10054]: identified spam (13.4/5.0) for root:99 in 24
> seconds, 4600 bytes.
> Sep 24 10:34:05 spam spamd[10060]: clean message (-4.0/5.0) for root:99 in 26
> seconds, 17417 bytes.
> Sep 24 10:34:15 spam spamd[10072]: identified spam (9.9/5.0) for root:99 in 26
> seconds, 5867 bytes.
> Sep 24 10:34:16 spam spamd[10066]: clean message (2.9/5.0) for root:99 in 29
> seconds, 4115 bytes.
> Sep 24 10:34:22 spam spamd[10079]: identified spam (15.8/5.0) for root:99 in 28
> seconds, 25666 bytes.
> Sep 24 10:34:33 spam spamd[10087]: identified spam (13.8/5.0) for root:99 in 27
> seconds, 4652 bytes.
>
>
> Kevin Gagel wrote:
> >
> > Sorry everyone I missed the original...
> > >From this message I see that processing should be around 10 seconds or less on a
> > average email if you are using Razor with SA. I am using razor and my average
> > seems to be around 20 - 30 seconds for each message. So I would like to know if
> > there is something that I can check to see what the problem might be, any ideas?
> >
> > Mariano Absatz wrote:
> > >
> > > El 24 Sep 2002 a las 10:29, Vivek Khera escribió:
> > >
> > > > >>>>> "MA" == Mariano Absatz <[EMAIL PROTECTED]> writes:
> > > >
> > > > MA> I know this is definitively a muddy item, but I have to come
> > > > MA> through it... We are planning to integrate SA into a border smtp
> > > > MA> gateway to process, tag and accept all mail coming from the
> > > > MA> Internet for a large ISP.
> > > >
> > > > For a "large ISP" you'll need a "large computer". You need to be more
> > > I know that... it'll probably be a bunch of intel or sparc machines with ample
>ram
> > > and fast scsi disks...
> > >
> > > > specific on how much email you intend to process, how bursty that
> > > > traffic is (ie, is it pretty much consistent or does a big amount come
> > > > in at any specific time.)
> > > What I want to be able to say is something like "a Pentium IV with 512Mb of RAM
>and
> > > a SCSI drive will be able to deliver 2 or 3 messages per second on average" and
>then
> > > decide if I need 2, 3 or 10 of this machines for the job...
> > >
> > > >
> > > > In any case, for my inbound mail, I have SA running as a content
> > > > filter via amavisd-new (so it is always there waiting to process
> > > > mail), and I don't do the RBL checks since my mail server does those
> > > > by itself. A typical message will take anywhere from 200 to 10000 ms
> > > > to process, depending on how long Razor takes to respond. Without
> > > > Razor, it would almost never go above 300ms per message.
> > >
> > > This is good info for me... this times are for SA processing (no AV) are
>they?... on
> > > what kind of hardware?
> > >
> > > Thanx a lot for the info.
> > >
> > > -------------------------------------------------------
> > > This sf.net email is sponsored by:ThinkGeek
> > > Welcome to geek heaven.
> > > http://thinkgeek.com/sf
> > > _______________________________________________
> > > Spamassassin-talk mailing list
> > > [EMAIL PROTECTED]
> > > https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
> >
> > --
> > ========================
> > Kevin W. Gagel
> > Network Administrator
> > College of New Caledonia
> > [EMAIL PROTECTED]
> > [EMAIL PROTECTED]
> > (250)562-2131 loc. 448
> > ========================
> >
> > --------------------------------
> > The College of New Caledonia
> > Visit us at http://www.cnc.bc.ca
> > --------------------------------
> >
> > -------------------------------------------------------
> > This sf.net email is sponsored by:ThinkGeek
> > Welcome to geek heaven.
> > http://thinkgeek.com/sf
> > _______________________________________________
> > Spamassassin-talk mailing list
> > [EMAIL PROTECTED]
> > https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
>
>
--
Chad Ziccardi, Professional Slacker [EMAIL PROTECTED]
"Some cause happiness wherever they go; others whenever they go."
-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk
