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

Re: [Testbed-admins] Interface Re-numbering Error



I wouldn't even worry about remapping.  We identify interfaces now by their
MAC address (not name), so it shouldn't be necessary to remap.  As you can
tell by the state of the code, we haven't done it in awhile!

On Tue, Jul 06, 2010 at 10:46:46AM -0500, Barry Trent wrote:
> I'm trying to add a new net of nodes to my Emulab. They have 7 Ethernet 
> interfaces and, unlike nodes I have previously added, FreeBSD and Linux 
> number these interfaces differently. Using the web interface page for 
> New Testbed Nodes, I tried to enter the required remapping of interfaces:
> 
> (FreeBSD:Linux)
> 0:5
> 1:6
> 2:0
> 3:1
> 4:3
> 5:4
> 6:2
> 
> When I select the button to re-number the interfaces, I get an error 
> message:
> 
> Could not continue. Please contact  Testbed Operations 
> (testbed-ops@emulab.atcorp.com)
> 
> And an email with the following error message is delivered to the 
> testbed-ops account:
> 
> In 
> /newnodes_list.php3?selected[]=65&newtype=&newprefix=&addnumber=&remap[0]=5&remap[1]=6&remap[2]=0&remap[3]=1&remap[4]=3&remap[5]=4&remap[6]=2&remap[6]=&renumber=Re-number 
> interfaces of selected nodes
> 
> DB Query failed - In /newnodes_list.php3
>    Query: UPDATE new_interfaces SET card = 6 WHERE card = 101 AND 
> (new_node_id=65)
>    Error: Duplicate entry '65-6' for key 1
> 
> Interesting that I left box 7 on the web page blank, but the last 2 
> "&remap[n]=" clauses in the form submission both seem to reference 
> "&remap[6]". Shouldn't those last two be "&remap[6]=2&remap[7]=" instead 
> of "&remap[6]=2&remap[6]=" ?
> 
> I just looked a bit deeper and it looks like a "cut-and-paste" bug in 
> the newnodes_list.php3 script to me:
> 
> <tr>
>      <td>6</td>
>      <td><input type="text" size=2 name="remap[6]"></td>
> </tr>
> <tr>
>      <td>7</td>
>      <td><input type="text" size=2 name="remap[6]"></td>
> </tr>
> 
> 
> Should I just edit the script and try again? Is this a known bug?
> _______________________________________________
> Testbed-admins mailing list
> Testbed-admins@flux.utah.edu
> http://www.flux.utah.edu/mailman/listinfo/testbed-admins