hobbitping latency reporting with multiple hosts
Hello, We are having some discrepancies with the latencies xymon reports in it's conn tests as opposed to what we were seeing in other monitoring programs. What we're seeing is that xymon reports a much higher latency for certain hosts than what we're used to. We are using hobbitping from xymon 4.2.2 on x64, novell suse linux enterprise server 10 sp2.
We think we've tracked it down to the following:
We are testing with 3 hosts: 194.109.6.92 - www.xs4all.nl 195.130.131.39 - www.telenet.be 212.27.48.10 - www.free.fr
These are used as an example, but our real world tests with internal hosts show the same issue. Anyway, when we ping each of these hosts we don't see any issue:
% ./hobbitping 194.109.6.92 194.109.6.92 is alive (6 ms) ./hobbitping 195.130.131.39 195.130.131.39 is alive (2 ms) % ./hobbitping 212.27.48.10 212.27.48.10 is alive (16 ms)
However, as soon as we start combining hosts we see strange results: ./hobbitping 195.130.131.39 194.109.6.92 195.130.131.39 is alive (22 ms) 194.109.6.92 is alive (6 ms) ./hobbitping 195.130.131.39 194.109.6.92 212.27.48.10 195.130.131.39 is alive (44 ms) 194.109.6.92 is alive (22 ms) 212.27.48.10 is alive (16 ms)
Even multiple tests on the same host show this: % ./hobbitping 195.130.131.39 195.130.131.39 195.130.131.39 195.130.131.39 is alive (44 ms) 195.130.131.39 is alive (22 ms) 195.130.131.39 is alive (2 ms)
We've found a reference to someone who might have the same issue, but there wasn't any follow up to that: http://www.hswn.dk/hobbiton/2008/09/msg00188.html
Is anyone else seeing this issue?
thx
// nick
On Wed, Feb 25, 2009 at 10:29:40AM +0100, nick.nauwelaerts at thomsonreuters.com wrote:
We are having some discrepancies with the latencies xymon reports in it's conn tests as opposed to what we were seeing in other monitoring programs. What we're seeing is that xymon reports a much higher latency for certain hosts than what we're used to. We are using hobbitping from xymon 4.2.2 on x64, novell suse linux enterprise server 10 sp2.
We think we've tracked it down to the following:
We are testing with 3 hosts: 194.109.6.92 - www.xs4all.nl 195.130.131.39 - www.telenet.be 212.27.48.10 - www.free.fr
hobbitping in the current version isn't really designed to do very accurate RTT measurements - it's primary purpose is to see if the server is responding at all, not how fast it responds.
What happens is that hobbitping is extremely sensitive to delays induced by ARP lookups and brief spikes in e.g. the load on a network switch or router passing the ping packets. That is because hobbitping will only ping each host once (if the host responds to the first ping within the timeout period), and the first ping will often vary a lot more in RTT than subsequent ones.
So I'd recommend using fping if you want some more accurate RTT measurements.
There is a new ping daemon in the works for a later Xymon release, and indications are that it performs much better in this respect.
Regards, Henrik
-----Original Message----- From: Henrik Størner [mailto:henrik at hswn.dk] Sent: Thursday, February 26, 2009 11:22 To: hobbit at hswn.dk Subject: Re: [hobbit] hobbitping latency reporting with multiple hosts
On Wed, Feb 25, 2009 at 10:29:40AM +0100, nick.nauwelaerts at thomsonreuters.com wrote:
We are having some discrepancies with the latencies xymon reports in it's conn tests as opposed to what we were seeing in other monitoring programs. What we're seeing is that xymon reports a much higher latency for certain hosts than what we're used to. We are using hobbitping from xymon 4.2.2 on x64, novell suse linux enterprise server 10 sp2.
We think we've tracked it down to the following:
We are testing with 3 hosts: 194.109.6.92 - www.xs4all.nl 195.130.131.39 - www.telenet.be 212.27.48.10 - www.free.fr
hobbitping in the current version isn't really designed to do very accurate RTT measurements - it's primary purpose is to see if the server is responding at all, not how fast it responds.
What happens is that hobbitping is extremely sensitive to delays induced by ARP lookups and brief spikes in e.g. the load on a network switch or router passing the ping packets. That is because hobbitping will only ping each host once (if the host responds to the first ping within the timeout period), and the first ping will often vary a lot more in RTT than subsequent ones.
So I'd recommend using fping if you want some more accurate RTT measurements.
There is a new ping daemon in the works for a later Xymon release, and indications are that it performs much better in this respect.
Even using hobbitping with the same host trice guives these strange results, so the arp entries over most of the link chain should already be cached. That being said, for the time being we'll migrate to fping because we do look at the RTT variations.
thx
// nick
Hello.
There are a few things I dont quite understand yet about hobbit and I cant find anything that explains it in the docs so i better ask :)
In the hobbit "display", the webpage, the hobbit server itself is listed with a few things i dont get on the other servers, like bbd, bbgen, bbtest and hobbitd. These seem to provide interesting things though. I have installed hobbit-client on the clients.
Regards, Benedikt
It is confusing, but Hobbit (now called XyMon, by the way) is derived from something called Big Brother. Big Brother abbreviates to 'BB', therefor many things in Hobbit/XyMon are still called 'bb<something>'
You will only see those processes monitored on the Hobbit server, since they (together) ARE the Hobbit server.
-----Oorspronkelijk bericht----- Van: Benedikt Kristinsson [mailto:benedikt at utrad.is] Verzonden: maandag 2 maart 2009 23:56 Aan: hobbit at hswn.dk Onderwerp: [hobbit] Understanding hobbit
Hello.
There are a few things I dont quite understand yet about hobbit and I cant find anything that explains it in the docs so i better ask :)
In the hobbit "display", the webpage, the hobbit server itself is listed with a few things i dont get on the other servers, like bbd, bbgen, bbtest and hobbitd. These seem to provide interesting things though. I have installed hobbit-client on the clients.
Regards, Benedikt
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
On Monday 02 March 2009, Benedikt Kristinsson wrote:
Hello.
There are a few things I dont quite understand yet about hobbit and I cant find anything that explains it in the docs so i better ask :)
In the hobbit "display", the webpage, the hobbit server itself is listed with a few things i dont get on the other servers, like bbd, bbgen, bbtest and hobbitd. These seem to provide interesting things though. I have installed hobbit-client on the clients. Like someone else explained, bb is/was used by BigBrother, Hobbit was 'BigBrother improved' and Xymon is the new name for Hobbit.
The xymon/hobbit server software running on your monitor server is reporting its status as bbd,bbgen, ... What I do is changing MACHINEDOTS in hobbitserver.cfg to <hostname>-hobbit and add a entry "<ip> <hostname>-hobbit # noconn bbd" in bb-hosts. So all the hobbit server checks are reported on a separate line.
Stef
Is there a question in there somewhere?
GLH
-----Original Message----- From: Benedikt Kristinsson [mailto:benedikt at utrad.is] Sent: Monday, March 02, 2009 4:56 PM To: hobbit at hswn.dk Subject: [hobbit] Understanding hobbit
Hello.
There are a few things I dont quite understand yet about hobbit and I cant find anything that explains it in the docs so i better ask :)
In the hobbit "display", the webpage, the hobbit server itself is listed with a few things i dont get on the other servers, like bbd, bbgen, bbtest and hobbitd. These seem to provide interesting things though. I have installed hobbit-client on the clients.
Regards, Benedikt
To unsubscribe from the hobbit list, send an e-mail to hobbit-unsubscribe at hswn.dk
participants (6)
-
benedikt＠utrad.is
-
d.kip＠gdi.minjus.nl
-
greg.hubbard＠eds.com
-
henrik＠hswn.dk
-
nick.nauwelaerts＠thomsonreuters.com
-
stef.coene＠docum.org