headers
Edson Manners | 7 Nov 2003 20:50
Picon

Loss values in rrd

Did you ever get an answer to this?

<being snippet>
Hi Tobi

Can u please tell me how can i store absolute loss in the rrd's of 
smokeping instead of the average loss for a particular sample of pings 
which is stored here.
1045226400: nan 9.3333333333e-01 5.5770533333e-01 5.2717600000e-01 
5.2975066667e-01 5.3167266667e-01 5.3310866667e-01 5.4013600000e-01 
5.4086000000e-01 5.4166800000e-01 5.4320866667e-01 5.4799200000e-01 
5.5765866667e-01 5.5036000000e-01 5.5986666667e-01 5.6321933333e-01 
5.6855800000e-01 5.7563866667e-01 5.9634400000e-01 6.2447133333e-01 
6.5556133333e-01 6.6326200000e-01 7.4157400000e-01
In above example even when no packet is lost, its showing a loss of 9.333%.

I assume that i have to change some code in Smokeping.pm. When update 
rrd is called this code is there.
         RRDs::update ( $name.".rrd",
                           '--template',(join ":", "uptime", "loss", 
"median", map { "ping${_}" } 1..($cfg->{Database}{pings})),
                           "N:".$probes->{$probe}->rrdupdate_string($tree)
                         );
I am printing $probes and its showing as follows
        print $probes->{$probe}->rrdupdate_string($tree),"\n";

U:1:6.4884000000e-01:6.3117000000e-01:6.3392000000e-01:6.3793000000e-01:6.3853000000e-01:
6.4057000000e-01:6.4591000000e-01:6.4702000000e-01:6.4712000000e-01:6.4884000000e-01:
6.4884000000e-01:U:6.5059000000e-01:6.5254000000e-01:6.5447000000e-01:6.5451000000e-01:
6.6208000000e-01:6.6362000000e-01:6.6723000000e-01:7.0756000000e-01:7.3684000000e-01
(Continue reading)

Permalink | Reply | 2 comments |
headers
Edson Manners | 7 Nov 2003 21:16
Picon

Re: Loss values in rrd

I realized that I need to properly format this question for the list.
Sorry about the confusing one that I posted previously.

I have been through the archives & smokeping.pm source and would like
clarification on one thing. 

If the loss value in the rrd database is as Tobi said "the number of
lost packets in absolute numbers"
(http://www.ee.ethz.ch/~slist/smokeping-users/msg00489.html)
Then how come I have floating point numbers instead of integers?

uptime  loss
nan 	1.9366666667e+00
nan 	2.9333333333e+00
nan 	2.0666666667e+00
nan 	1.0666666667e+00
nan 	1.9333333333e+00

I have looked at the source and cannot understand why I would end up
with floats in that column that didn't have all zero's in the decimal
positions. Further more when I count the number of 'nan's in the ping
values on the same row they don't always correlate with the rounded or
truncated integer value in the loss section. What am I doing stupidly??

--
Unsubscribe mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/smokeping-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
(Continue reading)

Permalink | Reply | 1 comment |
headers
Dierlam, Andy David | 10 Nov 2003 16:08

CISCOrttMonJitter.PM ? ? ?

Hey all, 

Anyone useing smokeping to monitor Cisco's RTT Jitter capabilities ? 
I see there is a RTT Probe for DNS and ICMP and TCP.

Thanks,
Andy Dierlam

-----------------------------------------------
   Make a PrimusTalk PC to Phone call today!
  Go to http://www.iprimus.net to learn more. 
------------------------------------------------

--
Unsubscribe mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/smokeping-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
Permalink | Reply | 1 comment |
headers
Scott Schappell | 10 Nov 2003 16:36

Gaps in ping charts?


I'm noticing large gaps in my fping charts on all hosts (even the
ones on the LAN) which leads me to believe it may be a server issue.
I was just wondering if anyone else has noticed a similar problem and
if so how it was remedied.  The gaps range from 30 minutes to 4
hours.

It's not a huge issue, it just generates some false positives on my
alerts.
--
PGP Key: http://archon.silvertree.org/pgp.txt (also on various key
servers)
"Compassion and retribution are two sides of the same coin. Necessity
dictates on what side the coin will fall."
Permalink | Reply | 0 comments |
headers
Naveen Nathan | 11 Nov 2003 15:55

FastCGI

Does anyone know how to go about using FastCGI with Smokeping?
I'm using Zeus Web Server v4.0.

Any help would be appreciated.

Thanks,
Naveen Nathan

--
Unsubscribe mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/smokeping-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
Permalink | Reply | 0 comments |
headers
Sureshbabu Nair | 12 Nov 2003 21:24

Graphing Question

Hi,
I have just installed smokeping to graph latency to some of our 
offices.. I need info on how to scale the graph. I wanted the graph 
y-axis to be displayed in milliseconds(ms) instead of seconds.. though 
some of the graphs are showing it in milliseconds some are not..please 
help me out on thi

thanks

Sureshbabu Nair

--
Unsubscribe mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/smokeping-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
Permalink | Reply | 1 comment |
headers
Nipper, Arnold | 12 Nov 2003 23:36
Picon

Re: Graphing Question

I guess option

 -U|--unit value
           You could use unit to be displayed on y axis. It is wise to use
only short units on graph, how­
           ever.

should do. In your case "--unit -3".

HTH, Arnold

On Wednesday, November 12, 2003 9:24 PM, Sureshbabu Nair
<suresh.babu <at> webhelp.com> wrote:
> Hi,
> I have just installed smokeping to graph latency to some of our
> offices.. I need info on how to scale the graph. I wanted the graph
> y-axis to be displayed in milliseconds(ms) instead of seconds.. though
> some of the graphs are showing it in milliseconds some are not..please
> help me out on thi
>
> thanks
>
> Sureshbabu Nair

--
Unsubscribe mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=unsubscribe
Help        mailto:smokeping-users-request <at> list.ee.ethz.ch?subject=help
Archive     http://www.ee.ethz.ch/~slist/smokeping-users
WebAdmin    http://www.ee.ethz.ch/~slist/lsg2.cgi
(Continue reading)

Permalink | Reply | 0 comments |
headers
Tobias Oetiker | 13 Nov 2003 16:57
Picon
Picon
Favicon

Re: Loss values in rrd

Nov 7 Edson Manners wrote:

> I realized that I need to properly format this question for the list.
> Sorry about the confusing one that I posted previously.
>
> I have been through the archives & smokeping.pm source and would like
> clarification on one thing.
>
> If the loss value in the rrd database is as Tobi said "the number of
> lost packets in absolute numbers"
> (http://www.ee.ethz.ch/~slist/smokeping-users/msg00489.html)
> Then how come I have floating point numbers instead of integers?
>
> uptime  loss
> nan 	1.9366666667e+00
> nan 	2.9333333333e+00
> nan 	2.0666666667e+00
> nan 	1.0666666667e+00
> nan 	1.9333333333e+00
>
> I have looked at the source and cannot understand why I would end up
> with floats in that column that didn't have all zero's in the decimal
> positions. Further more when I count the number of 'nan's in the ping
> values on the same row they don't always correlate with the rounded or
> truncated integer value in the loss section. What am I doing stupidly??

Note that the RRD is updated in real time, the database stores the
data every 'step' by interpolating the data it gets ... it is
normal to NOT get whole numbers
tobi
(Continue reading)

Permalink | Reply | 0 comments |
headers
Itamar Belisha | 13 Nov 2003 22:29

Alerts Bug!!!

Hello,
The smokeping was set and working for a while in my network.
I tried to add Alerts to it.
I have tested it but for some reason I don't get any messages.
Here under is the alert configuration:

 
+NOCDelay
type = rtt
pattern = >39
comment = Delay Over 39ms/10min

 
I have also added the alert to the computers in the config file.
When I run smokeping in Debug mode I get in the begging of the output
message that I am not sure but might be an error regarding the alerts.
Any good advice? Someone?
Thanks for the help and the effort,
Itamar

 

root <at> jmmftp1 bin]# ./smokeping --debug
### fping seems to report in 1 miliseconds
### Compiling
 sub { my $d = shift; my $y = $d->{rtt};
return 0 unless 2 <= scalar  <at> $y ;
return 0 unless defined $y->[-2] and $y->[-2] < 0.039;
return 0 unless defined $y->[-1] and $y->[-1] > 0.039;
return 1; }
(Continue reading)

Permalink | Reply | 1 comment |
headers
Manoj Kumar Neelapareddy | 14 Nov 2003 07:56
Picon
Favicon

Help with smokeping.cgi

Hi all,

i just installed smoke ping on Redhat linux 9
i am able to start and run the smokeping, but the
problem is with accesiing the smokeping.cgi with
webbrowser.

i am getting the error "500 Internal server erroor"

i found following lines in the error log of apache.

[Fri Nov 14 12:25:50 2003] [error] [client 10.37.4.3]
Premature end of script headers: smokeping.cgi
[Fri Nov 14 12:25:50 2003] [error] [client 10.37.4.3]
Can't locate Smokeping.pm in  <at> INC ( <at> INC contains:
/root/soft/smokeping/lib
/usr/local/rrdtool-1.0.45/lib/perl
/usr/lib/perl5/5.8.0/i386-linux-thread-multi
/usr/lib/perl5/5.8.0
/usr/lib/perl5/site_perl/5.8.0/i386-linux-thread-multi
/usr/lib/perl5/site_perl/5.8.0
/usr/lib/perl5/site_perl
/usr/lib/perl5/vendor_perl/5.8.0/i386-linux-thread-multi
/usr/lib/perl5/vendor_perl/5.8.0
/usr/lib/perl5/vendor_perl .) at
/var/www/cgi-bin/smokeping.cgi line 7.
[Fri Nov 14 12:25:50 2003] [error] [client 10.37.4.3]
BEGIN failed--compilation aborted at
/var/www/cgi-bin/smokeping.cgi line 7.
[Fri Nov 14 12:25:50 2003] [error] [client 10.37.4.3]
(Continue reading)

Permalink | Reply | 0 comments |

Gmane