Skip Menu |

This queue is for tickets about the Net-Ping CPAN distribution.

Report information
The Basics
Id: 27151
Status: stalled
Priority: 0/
Queue: Net-Ping

People
Owner: Nobody in particular
Requestors: ang [...] nmc-m.dtag.de
tom [...] stonehenge.com
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: (no value)



CC: <bug-Net-Ping [...] rt.cpan.org>, <bbb [...] cpan.org>
Subject: Net::Ping Bug found?
Date: Wed, 16 May 2007 15:01:00 +0200
To: <beginners [...] perl.org>
From: "Angerstein" <ang [...] nmc-m.dtag.de>
Hi there! I am using Net::Ping on AIX (here 5.2) and I noticed a strange behaviour concerning the ICMP Payload Data Field. If I do a typical default ping with $p = new Net::Ping('icmp', $ping_timeout); I will get a EthernetII-IP-ICMP-Package (so far so good), but this package does not contain a Payload field (I would normaly expect), therefor it contains a Ethernet II Trailer (Wireshark). (I verified this by using iptrace v2.0 on aix and wireshark on my win2k desktop box.) If I do a ping with a manuell specified payload (like $p = new Net::Ping('icmp', $ping_timeout, 18);) the package looks (for me as a halfaway network geek) good. When is this a problem? It looks like some Router (by some Vendors) do not send icmp echo replies on such kind of requests. Sometimes it might look like that your router is down, but it isn´t. (And If you have like me over 1000 multivendor routers/switche/etc to manage, it´s not funny.) What could be the Problem: The package building process in Net::Ping is filthy-> I guess the IP-Total Length field is not set correctly (the payload lenght is not added to it). So Please: Could somebody verify this on her/his own System? If anybody feels able or have time to patch this, please patch the multithreadsupport patch right with it. (https://rt.cpan.org/Public/Bug/Display.html?id=4170) If you need to verify the fix feel free to ask me. Bastian Angerstein
CC: beginners [...] perl.org, bug-Net-Ping [...] rt.cpan.org, bbb [...] cpan.org
Subject: Re: Net::Ping Bug found?
Date: Wed, 16 May 2007 09:37:57 -0700
To: Angerstein <ang [...] nmc-m.dtag.de>
From: "Tom Phoenix" <tom [...] stonehenge.com>
On 5/16/07, Angerstein <ang@nmc-m.dtag.de> wrote: Show quoted text
> So Please: > Could somebody verify this on her/his own System?
I'm sure that many people will be glad to help you. Could you please supply a small program that testers could run and send you the output? Good luck with your project! --Tom Phoenix Stonehenge Perl Training
On Wed May 16 09:01:33 2007, ang@nmc-m.dtag.de wrote: Show quoted text
> Hi there! > > I am using Net::Ping on AIX (here 5.2) and I noticed a strange behaviour > concerning the ICMP Payload Data Field. > > If I do a typical default ping with > $p = new Net::Ping('icmp', $ping_timeout); > I will get a EthernetII-IP-ICMP-Package (so far so good), > but this package does not contain a Payload field (I would normaly > expect), > therefor it contains a Ethernet II Trailer (Wireshark). > > (I verified this by using iptrace v2.0 on aix and wireshark on my win2k > desktop box.) > > If I do a ping with a manuell specified payload > (like $p = new Net::Ping('icmp', $ping_timeout, 18);) > the package looks (for me as a halfaway network geek) good. > > When is this a problem? > It looks like some Router (by some Vendors) do not send icmp echo > replies on such kind of requests. > Sometimes it might look like that your router is down, but it isn´t. > (And If you have like me over 1000 multivendor routers/switche/etc to > manage, > it´s not funny.) > > What could be the Problem: > The package building process in Net::Ping is filthy-> I guess > the IP-Total Length field is not set correctly (the payload lenght is > not added to it). > > So Please: > Could somebody verify this on her/his own System? > > If anybody feels able or have time to patch this, please patch the > multithreadsupport patch right with it. > (https://rt.cpan.org/Public/Bug/Display.html?id=4170) > > If you need to verify the fix feel free to ask me. > > Bastian Angerstein >
Can you please provide a code sample that demonstrates the problem? Without it, its very difficult to guess what the problem might be.