Conn Test Times
list Michael A. Price
Hello, Just curious if anyone knows the answer to this... I am monitoring two hosts on the same switch/network from the hobbit server. When I look at the ping times on the hobbit server it says, it's around 20 ms for one and 40 ms for the other. The trending graphs for the conn test is reflecting the same thing. But when I actually get on the CLI on the hobbit server and ping the two hosts, they are around .7 ms ping response times. Whats the different between the application pinging the hosts and me doing it at the CLI.???? Thanks, michael
list Martin Flemming
Hi !
I would use the procs-function to count several software versions & programms to see which version will be heavily used or nobody cares about it ...
(our software developer want's to know it ) ...
So, the first shot was to include this in the PROCS test and it runs really good with rrd-files and graphs ... but the procs-test should be only my test for important system-processes ...
So i want to have the same test but with another flag like "sfw" or something else ...
Is there a solution to use PROCS exactly for another/additional test ?
.. or got anybody a solution or hint for it ?
Thanks,
Martin
list Josh Luthman
Michael, Click the icon on the bb.html and confirm it is the correct IP address. When I had this "problem" I noticed it was resolving the hostname and ping'ing the incorrect IP address. If you do notice it is ping'ing the wrong address add the testip arguement after the last comment. Josh
▸
On 10/12/07, Michael A. Price <user-d7d653acf808@xymon.invalid> wrote:Hello, Just curious if anyone knows the answer to this… I am monitoring two hosts on the same switch/network from the hobbit server. When I look at the ping times on the hobbit server it says, it's around 20 ms for one and 40 ms for the other. The trending graphs for the conn test is reflecting the same thing. But when I actually get on the CLI on the hobbit server and ping the two hosts, they are around .7 ms ping response times. Whats the different between the application pinging the hosts and me doing it at the CLI.???? Thanks, michael
--
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencer
list Joost van den Broek
I don't think this has anything to do with ping'ing the wrong hosts. In my case it has afaik to do with hobbitping. Switching to fping would solve this issue, but I personally don't care (using smokeping for RTT stats). If you check Henkrik's demo page, you'll see that his local hosts (172.16.x, assuming these are not in a VPN) also have very high response times. Don't know why that is, maybe it will be resolved in the 4.3 final? -Joost Josh Luthman schreef:
▸
Michael,
Click the icon on the bb.html and confirm it is the correct IP address. When I had this "problem" I noticed it was resolving the hostname and
ping'ing the incorrect IP address. If you do notice it is ping'ing the
wrong address add the testip arguement after the last comment.
Josh
On 10/12/07, *Michael A. Price* <user-d7d653acf808@xymon.invalid
<mailto:user-d7d653acf808@xymon.invalid>> wrote:
Hello,
Just curious if anyone knows the answer to this…
I am monitoring two hosts on the same switch/network from the hobbit
server. When I look at the ping times on the hobbit server it says,
it's around 20 ms for one and 40 ms for the other. The trending
graphs for the conn test is reflecting the same thing.
But when I actually get on the CLI on the hobbit server and ping the
two hosts, they are around .7 ms ping response times.
Whats the different between the application pinging the hosts and me
doing it at the CLI.????
Thanks, michael
--
Josh Luthman
Office: XXX-XXX-XXXX
Direct: XXX-XXX-XXXX
XXXX Wayne St
Suite XXXX
Troy, OH XXXXX
Those who don't understand UNIX are condemned to reinvent it, poorly.
--- Henry Spencerlist Josh Luthman
That is another thing I forgot to ask - are you using ping for both tests? Hobbitping for both tests? Try Joost's (let me know if that's wrong) suggestion and use fping.
▸
On 10/12/07, Joost van den Broek <user-5bf70bf7662e@xymon.invalid> wrote:I don't think this has anything to do with ping'ing the wrong hosts. In my case it has afaik to do with hobbitping. Switching to fping would solve this issue, but I personally don't care (using smokeping for RTT stats). If you check Henkrik's demo page, you'll see that his local hosts (172.16.x, assuming these are not in a VPN) also have very high response times. Don't know why that is, maybe it will be resolved in the 4.3 final? -Joost Josh Luthman schreef:Michael, Click the icon on the bb.html and confirm it is the correct IP address. When I had this "problem" I noticed it was resolving the hostname and ping'ing the incorrect IP address. If you do notice it is ping'ing the wrong address add the testip arguement after the last comment. Josh On 10/12/07, *Michael A. Price* <user-d7d653acf808@xymon.invalid <mailto:user-d7d653acf808@xymon.invalid>> wrote: Hello, Just curious if anyone knows the answer to this… I am monitoring two hosts on the same switch/network from the hobbit server. When I look at the ping times on the hobbit server it says, it's around 20 ms for one and 40 ms for the other. The trending graphs for the conn test is reflecting the same thing. But when I actually get on the CLI on the hobbit server and ping the two hosts, they are around .7 ms ping response times. Whats the different between the application pinging the hosts and me doing it at the CLI.???? Thanks, michael -- Josh Luthman Office: XXX-XXX-XXXX Direct: XXX-XXX-XXXX XXXX Wayne St Suite XXXX Troy, OH XXXXX Those who don't understand UNIX are condemned to reinvent it, poorly. --- Henry Spencer
-- Josh Luthman Office: XXX-XXX-XXXX Direct: XXX-XXX-XXXX XXXX Wayne St Suite XXXX Troy, OH XXXXX Those who don't understand UNIX are condemned to reinvent it, poorly. --- Henry Spencer