[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

RE: Network interface testing utlity



>
>
>
>---- Original Message ----
>From: kushalkoolwal@hotmail.com
>To: owens@netptc.net, debian-user@lists.debian.org
>Subject: RE: Network interface testing utlity
>Date: Sun, 4 Oct 2009 13:57:20 -0700
>
>>
>>> Your monitor will indirectly monitor these errors if
>>> it measures throughput which will decrease as the errors increase.
>>I always thought that the fields "Error on Receiving/Transmission",
>for example in the package ethstatus, is for telling if there are any
>errors detected in the link layer? Isn't that the case? If not, then
>what does those error tell you about?
>> 
>>
>>Kushal Koolwal
>>
>>I do blog at http://blogs.koolwal.net/
>>
>>
I'm not familiar with ethstatus but most error monitoring these days
is done at the TCP/IP level in software
L
>>
>>
>>
>>
>>
>>----------------------------------------
>>> From: owens@netptc.net
>>> To: kushalkoolwal@hotmail.com; owens@netptc.net;
>debian-user@lists.debian.org
>>> Subject: RE: Network interface testing utlity
>>> Date: Sun, 4 Oct 2009 07:32:53 -0700
>>>
>>>>
>>>>
>>>>
>>>>---- Original Message ----
>>>>From: kushalkoolwal@hotmail.com
>>>>To: owens@netptc.net, debian-user@lists.debian.org
>>>>Subject: RE: Network interface testing utlity
>>>>Date: Sat, 3 Oct 2009 23:37:52 -0700
>>>>

>>>>>> OK then. The next question is how and what you intend to test.
>Are
>>>>>> you trying only to test a link (the media between your box and
>the
>>>>>> next box) or end-to end (where the two boxes are separated by
>many
>>>>>> links and many routers)?
>>>>>> Probably a block diagram of your proposed test setup would
>help.
>>>>BTW
>>>>>> what media types?
>>>>>
>>>>>The idea is to test between two boxes (connected directly with a
>>>>standard CAT5 cable) to eliminate/minimize any other network
>>>>interference caused by routers, switches, hubs, etc.
>>>>>
>>>>>On one box I will have my Gigabit Ethernet card (connected with
>>>>different media types) that I want to test and on the other end of
>>>>the box, I am going to have a standard known reference Gigabit
>>>>Ethernet.
>>>>>
>>>>>The media types are:
>>>>>a) Standard CAT5
>>>>>b) Custom twisted pair
>>>>>
>>>>>Kushal Koolwal
>>>>>
>>>>>I do blog at http://blogs.koolwal.net/
>>>>>
>>>>>
>>>>>
>>> OK then, here's my opintion
>>> All (?) the tools you mention including Wireshark which is the one
>I
>>> use will provide network statistics. Since they are software,
>>> however, they reside above (in the protocol sense) the link (e.g.
>>> Ethernet) which is where the majority of the errors are likely to
>>> occur (if a check sum error occurs your Ethernet card will merely
>>> discard it and wait for a retransmission so it will never be seen
>by
>>> your analyzer). Your monitor will indirectly monitor these errors
>if
>>> it measures throughput which will decrease as the errors increase.
>I
>>> would surmise that by installing and using one of the tools you
>will
>>> gain lots of insight and knowledge but not enough to definitively
>>> choose between one network card and another.
>>> Larry
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>_________________________________________________________________
>>>>>Hotmail: Trusted email with powerful SPAM protection.
>>>>>http://clk.atdmt.com/GBL/go/177141665/direct/01/
>>>>>
>>>>>--
>>>>>To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
>>>>>with a subject of "unsubscribe". Trouble? Contact
>listmaster@lists.d
>>>>ebian.org
>>>>>
>>>>>
>>>
>>>
>>>
>>> --
>>> To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
>>> with a subject of "unsubscribe". Trouble? Contact
>listmaster@lists.debian.org
>>>
>> 		 	   		  
>>_________________________________________________________________
>>Hotmail: Powerful Free email with security by Microsoft.
>>http://clk.atdmt.com/GBL/go/171222986/direct/01/



Reply to: