Is there anything better than Net::Ping ? (need to calculate timeout)

[Anonymous Monk]

As simple question as possible... seeking good methods of calculating the ping timeout. No CPAN modules for this :(

Originally posted as a Categorized Question.

[ybiC]

Possibly related - *nix utility bing - "a point-to-point bandwidth measurement tool (hence the 'b'), based on ping." Available for Debian at http://www.debian.org/Packages/stable/net/bing.html, and I think RedHat.

If you roll your own solution, should probably run several iterations to increase accuracy of answer.

    cheers,
    ybiC

[ybiC]

bing isn't a Perl module, it's a *nix command-line utility. It doesn't exactly answer your question, but might provide relevant info. In my experience, it works fine for slower WAN and dialup links, but is unable to cope with LAN speeds.

Here's an example of bing between 2 hosts on opposite ends of a 256Kbps frame-relay PVC :

mail.here.dom:~# bing mail.there.dom
BING    mail.here.dom (10.20.30.1) and mail.there.dom (10.20.40.1)
        44 and 108 data bytes
1024 bits in 3.925ms: 260892bps, 0.003833ms per bit
1024 bits in 5.011ms: 204350bps, 0.004894ms per bit
1024 bits in 5.666ms: 180727bps, 0.005533ms per bit
1024 bits in 5.637ms: 181657bps, 0.005505ms per bit
1024 bits in 3.650ms: 280548bps, 0.003564ms per bit

--- mail.here.dom ---
bytes   out    in   dup  loss   rtt (ms): min       avg       max
   44    35    35          0%           0.236     0.257     0.399
  108    35    35          0%           0.234     0.239     0.263

--- mail.there.dom ---
bytes   out    in   dup  loss   rtt (ms): min       avg       max
   44    35    35          0%          42.366    44.905    71.570
  108    35    35          0%          46.016    56.398   166.923

--- estimated link characteristics ---
warning: rtt big host1 0.234ms < rtt small host2 0.236ms
estimated throughput 280548bps
minimum delay per packet 39.621ms (11115 bits)

average statistics (experimental) :
packet loss: small 0%, big 0%, total 0%
warning: rtt big host1 0.239ms < rtt small host2 0.257ms
average throughput 89105bps
average delay per packet 42.138ms (11822 bits)
weighted average throughput 89105bps

resetting after 35 samples.
mail.here.dom:~#

Let me know if I can be of any help.
    cheers,
    ybiC

[ybiC]

Not completely clear on what you're asking, but here goes:

You may be able to do this with Net::Ping by establishing a positive response from remote host, then reducing timeout until no response (similar to internal operation of traceroute/tracert).

    cheers,     ybiC

[lhoward]

Or by taking the current time before and after the successful ping and subtracting them.

[Anonymous Monk]

Yeah .. the "establishing a positive response from remote host, then reducing..." method could be used, but it's terribly resource/time consuming...
Still looking for better method/module...

Oh my! How would I use this 'bing' module?

Originally posted as a Categorized Answer.

[Anonymous Monk]

Re: Is there anything better than Net::Ping ? (need to calculate timeout)

Originally posted as a Categorized Answer.

[Anonymous Monk]

Well I just want what ping does for us in our company but try pathping just read about it.

Originally posted as a Categorized Answer.