[ale] Ethernet info/question

Greg Freemyer greg.freemyer at gmail.com
Mon May 10 16:32:34 EDT 2010


David,

I disconnected the suspicious switch last Wed, and all is still running fine.

In fact my Linux desktop is running the best it has run in months.  I
strongly suspect the suspicious switch has been playing havoc with
samba type traffic for months or even years.

Its nice to have things working smooth again.

Greg

On Thu, May 6, 2010 at 7:21 PM, David Tomaschik <david at tuxteam.com> wrote:
> You've discovered the purpose of the spanning tree protocol.[1]  Any
> modern managed switch should support this with no problem, and it is not
> vendor-specific if implemented properly.  There are some VLAN-related
> extensions to STP that make the situation a little cloudy in some cases,
> and this can lead to problems.  For example, (to my understanding) it's
> possible to connect 2 switches with access-mode ports on separate VLANs
> that are then leading to a loop.
>
> Knowing more about your situation would help to give any input on what
> might have been occurring.  Generally speaking, network loops or other
> MAJOR networking issues will bring a host/interface/network down
> entirely rather than just causing mild intermittent problems.
>
> --David
>
>
> [1] http://en.wikipedia.org/wiki/Spanning_tree_protocol
>
>
> On 05/06/2010 09:05 AM, Greg Freemyer wrote:
>> All,
>>
>> I've been sporadically fighting a series of sporadic issues with my 3
>> samba file servers.  Installing latest patches, updated one OS, etc.
>>
>> I came to the conclusion it had to be a network issue, not a server
>> issue and got serious about troubleshooting it and I think I found the
>> culprit to be a 24-port switch.  It will take a couple days to be
>> sure.
>>
>> ==> Info
>> I found a 48-port managed distribution switch at Amazon for $320 or
>> so.  (Cisco SRW248G4 48-port 10/100 + 4-port Gigabit Switch - WebView;
>> on sale for a couple more days)
>>
>> I did not realize you could get a managed switch that cheap, so I
>> thought others may find it useful to know.
>>
>> ==> Question
>> In my troubleshooting it seemed that I had a Ethernet loop somehow.
>> ie. switch1 -- switch2 -- switch3 -- switch1
>>
>> How bad is that?  Could it have be the real source of my problems, and
>> not my existing distribution switch like it seems to me?
>>
>> How can I easily figure out where the loop is?  (ie. I have a
>> relatively small setup, but still about 75-100 end-points and 10 or so
>> switches, with over half of those being in labs or under peoples desk,
>> etc.  I doubt I even know where they all are.
>>
>> Thanks
>> Greg
>>
>
>
> --
> David Tomaschik, RHCE
> Moderator, LinuxQuestions.org
> http://www.tuxteam.com
> david at tuxteam.com [GPG: 0x6D428695]
>
>
>
> _______________________________________________
> Ale mailing list
> Ale at ale.org
> http://mail.ale.org/mailman/listinfo/ale
> See JOBS, ANNOUNCE and SCHOOLS lists at
> http://mail.ale.org/mailman/listinfo
>
>



-- 
Greg Freemyer
Head of EDD Tape Extraction and Processing team
Litigation Triage Solutions Specialist
http://www.linkedin.com/in/gregfreemyer
CNN/TruTV Aired Forensic Imaging Demo -
   http://insession.blogs.cnn.com/2010/03/23/how-computer-evidence-gets-retrieved/

The Norcross Group
The Intersection of Evidence & Technology
http://www.norcrossgroup.com



More information about the Ale mailing list