headers
Sebastian Wiesinger | 2 Feb 2005 01:38
Picon

FPing6 localhost probe

Hi,

the IPv6 probe does the following to check the output:

fping6 -C 1 localhost 2>&1

On my Debian system localhost has only the ipv4 address assigned
(127.0.0.1) so smokeping hangs at startup [::1 is ip6-localhost here].

Perhaps your could change

$_ = `$self->{properties}{binary} -C 1 localhost 2>&1`;

to

$_ = `$self->{properties}{binary} -C 1 ::1 2>&1`;

in FPing6.pm if there are no reasons against it.

Regards,

Sebastian

--

-- 
GPG Key-ID: 0x76B79F20 (0x1B6034F476B79F20)
Wehret den Anfaengen: http://odem.org/informationsfreiheit/
Thunder rolled. ... It rolled a six.
  --Terry Pratchett, Guards! Guards!

--
(Continue reading)

Permalink | Reply | 0 comments |
headers
Niko Tyni | 5 Feb 2005 12:48
Picon
Picon

Re: packet-loss changes

On Thu, Jan 20, 2005 at 11:52:39AM +0000, Chris Wilson wrote:

> I have noticed problems with using smokeping to monitor large numbers of
> hosts, where fping would not wait long enough for replies to come back
> from the slowest servers. This probably depends on the order in which
> they are pinged. You could try modifying smokeping/lib/probes/FPing.pm,
> to add an extra parameter to the fping command time, to set the timeout.

> In this case I have added '-t20000' to set the timeout to 20 seconds.
> Probably a 1 second timeout (-t1000) would work better for you.
> 
> To support my number of hosts, I had to hack fping.pm to reduce the
> select time. I can explain more if anyone is interested.

Hi Chris,

please do explain. I'm working on a patch that changes a lot of things
with the probes, and it will include the '-t' fping option. How many
hosts do you have and what's the select time problem?

Cheers,
--

-- 
niko

--
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
Niko Tyni | 5 Feb 2005 13:53
Picon
Picon

Re: RRDTool Filling In Unknown Values For Missing Pings

Hi Kennedy,

I'm quoting the longish mail; see the comments below.  (BTW, your
mailer is broken, it's using 8-bit quote characters but claiming it's
7-bit ASCII.)

In article <20050106160149.54654.qmail <at> web54607.mail.yahoo.com>, 
hkclark <at> yahoo.com wrote:

> # Create a simplified Smokeping-like RRD
> rrdtool create test_01.rrd --start 1000000000 --step
> 300   \
>   DS:loss:GAUGE:600:0:20 DS:ping1:GAUGE:600:0:180  \
>   RRA:AVERAGE:0.5:1:1008 RRA:AVERAGE:0.5:12:4320  \
>   RRA:MIN:0.5:12:4320 RRA:MAX:0.5:12:4320  \
>   RRA:AVERAGE:0.5:144:720 RRA:MAX:0.5:144:720 \
>   RRA:MIN:0.5:144:720 
> # 
> # Load in some dummy data
> rrdtool update test_01.rrd 1000000200:4:5
> rrdtool update test_01.rrd 1000000521:4:5
> rrdtool update test_01.rrd 1000000821:8:9
> rrdtool update test_01.rrd 1000001121:U:5
> rrdtool update test_01.rrd 1000001421:U:5
> # Dump
> rrdtool dump test_01.rrd > test_01.xml
> 
> Note that it's putting in 3 values followed by 2 "U's"
> for the DS "loss".
>
(Continue reading)

Permalink | Reply | 1 comment |
headers
hkclark@yahoo.com | 7 Feb 2005 05:07
Picon
Favicon

Re: RRDTool Filling In Unknown Values For Missing Pings

Hi Niko,

Many thanks for taking a look at this!  Also, thanks
for all of your excellent contributions to Smokeping!

I like your idea about storing things at the average
value so the graph "looks right".  I guess the ideal
solution would be a flag in RRDtool that would force a
more strict interpretation of NaNs, but it seems the
"average trick" would probably be the next best thing.

Regards,
Kennedy

PS -- Sorry about the mailer issue -- I'm using Yahoo
mail for this list in case some spammer bums are
mining addresses off the list -- so I'm not sure if I
will be able to get Yahoo to fix it. :-)

--- Niko Tyni
<ntyni+smokeping-users <at> mappi.helsinki.fi> wrote:

> Hi Kennedy,
> 
> I'm quoting the longish mail; see the comments
> below.  (BTW, your
> mailer is broken, it's using 8-bit quote characters
> but claiming it's
> 7-bit ASCII.)
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Niko Tyni | 7 Feb 2005 12:38
Picon
Picon

Re: patch candidate for FPing.pm probe

(I'm pretty sure I sent a followup on this yesterday, but it seems
it got lost. Apologies for possible duplicates anyway.)

In article <F1C891E8-4B93-11D9-ADC4-003065FB84C2 <at> xmlhq.com>, cwitte wrote:

> I modified FPing.pm to calculate the delay between ping packets based 
> on the configured period length, so that the pings would be sent over 
> the duration of the period.

This is oldish, but I think it's a good idea. It should just be made
configurable, with an option like "spread". I'll look into it.

It would be nice to have something like this for the other probes too,
but doing it in a generic way would mean quite big changes. Hmm...

> -- Attached file removed by Ecartis and put at URL below --
> -- Type: application/octet-stream
> -- Size: 4k (4651 bytes)
> -- URL : http://www.ee.ethz.ch/~slist/p/02-FPing_delay.pm

Cheers,
--

-- 
niko

--
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 | 3 comments |
headers
Tobias Oetiker | 7 Feb 2005 12:51
Picon
Picon
Favicon

Re: patch candidate for FPing.pm probe

Today Niko Tyni wrote:

> (I'm pretty sure I sent a followup on this yesterday, but it seems
> it got lost. Apologies for possible duplicates anyway.)
>
> In article <F1C891E8-4B93-11D9-ADC4-003065FB84C2 <at> xmlhq.com>, cwitte wrote:
>
> > I modified FPing.pm to calculate the delay between ping packets based
> > on the configured period length, so that the pings would be sent over
> > the duration of the period.
>
> This is oldish, but I think it's a good idea. It should just be made
> configurable, with an option like "spread". I'll look into it.

I also wonder what the implications are as to what we measure ...
at the moment fping sends the icmp packets back-to-back
intentionally, to get a educated snapshot of what the link looked
at measurement time ... spreading the packets over the measurement
periode looses that advantag and goes into the direction of 'I want
to measure more often' this is already configurable by lowering the
step size so I am not sure this is a good idea ...

cheers
tobi

--

-- 
 ______    __   _
/_  __/_  / /  (_) Oetiker  <at>  ISG.EE, ETL F24.2, ETH, CH-8092 Zurich
 / // _ \/ _ \/ /  System Manager, Time Lord, Coder, Designer, Coach
/_/ \.__/_.__/_/   http://people.ee.ethz.ch/oetiker +41(0)44-632-5286
(Continue reading)

Permalink | Reply | 2 comments |
headers
Niko Tyni | 7 Feb 2005 13:23
Picon
Picon

Re: patch candidate for FPing.pm probe

On Mon, Feb 07, 2005 at 12:51:00PM +0100, Tobias Oetiker wrote:

> I also wonder what the implications are as to what we measure ...
> at the moment fping sends the icmp packets back-to-back
> intentionally, to get a educated snapshot of what the link looked
> at measurement time ... spreading the packets over the measurement
> periode looses that advantag and goes into the direction of 'I want
> to measure more often' this is already configurable by lowering the
> step size so I am not sure this is a good idea ...

You're right, I didn't think this through.

Reducing the step size has the disadvantage of making the measurement
harder on the network, unless we simultaneously reduce the # of ping
packets sent.

In fact, we get the very same behaviour by setting pings = 1 and step =
15, so there's really not much point having a new variable for it
anyway.

Cheers,
--

-- 
niko

--
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
Tobias Oetiker | 7 Feb 2005 13:30
Picon
Picon
Favicon

Re: patch candidate for FPing.pm probe

Today Niko Tyni wrote:

> On Mon, Feb 07, 2005 at 12:51:00PM +0100, Tobias Oetiker wrote:
>
> > I also wonder what the implications are as to what we measure ...
> > at the moment fping sends the icmp packets back-to-back
> > intentionally, to get a educated snapshot of what the link looked
> > at measurement time ... spreading the packets over the measurement
> > periode looses that advantag and goes into the direction of 'I want
> > to measure more often' this is already configurable by lowering the
> > step size so I am not sure this is a good idea ...
>
> You're right, I didn't think this through.
>
> Reducing the step size has the disadvantage of making the measurement
> harder on the network, unless we simultaneously reduce the # of ping
> packets sent.

sure this would have to be done ...

> In fact, we get the very same behaviour by setting pings = 1 and step =
> 15, so there's really not much point having a new variable for it
> anyway.

there you go :-)

tobi
>
> Cheers,
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bill Earle | 9 Feb 2005 22:31
Favicon

Possible modification to EchoPingSmtp


Hey folks,

I have submitted a patch for echoping to allow alternate
SMTP ports to be monitored. I have a modified version of
EchoPingSmtp that uses this feature.

Here is a diff for anyone interested:

*** EchoPingSmtp.pm     2003-03-20 17:06:15.000000000 -0500
--- EchoPingSmtpPort.pm 2005-02-09 15:01:58.000000000 -0500
***************
*** 1,8 ****
! package probes::EchoPingSmtp;

  =head1 NAME

! probes::EchoPingSmtp - an echoping(1) probe for SmokePing

  =head1 OVERVIEW

--- 1,8 ----
! package probes::EchoPingSmtpPort;

  =head1 NAME

! probes::EchoPingSmtpPort - an echoping(1) probe for SmokePing

  =head1 OVERVIEW
(Continue reading)

Permalink | Reply | 3 comments |
headers
Niko Tyni | 10 Feb 2005 11:19
Picon
Picon

Re: Possible modification to EchoPingSmtp

On Wed, Feb 09, 2005 at 04:31:55PM -0500, Bill Earle wrote:

> I have submitted a patch for echoping to allow alternate
> SMTP ports to be monitored. I have a modified version of
> EchoPingSmtp that uses this feature.

Hi,

what version of echoping are you using? My 5.2.0 just says

% echoping -S localhost:587 
echoping: The syntax hostname:port is only for HTTP or ICP

Perhaps this feature is only in the 6.0 beta version? I
haven't tried it at all and don't know if it's otherwise
command-line compatible with 5.2.

Cheers,
--

-- 
niko

--
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 | 2 comments |

Gmane