Send netdisco-users mailing list submissions to
netdisco-users@lists.sourceforge.net
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.sourceforge.net/lists/listinfo/netdisco-users
or, via email, send a message with subject or body 'help' to
netdisco-users-requ...@lists.sourceforge.net
You can reach the person managing the list at
netdisco-users-ow...@lists.sourceforge.net
When replying, please edit your Subject line so it is more specific
than "Re: Contents of netdisco-users digest..."
Today's Topics:
1. Re: (no subject) (Christian Ramseyer)
2. Re: Issue with H3C/HP 5130 Switches (Jeroen van Ingen)
3. Re: Issue with H3C/HP 5130 Switches (Muris)
4. Re: Issue with H3C/HP 5130 Switches (Oliver Gorwits)
5. Netdisco2 Job Queue (Oliver B?hrig)
--- Begin Message ---
Hi Andy
You can join node_ip, node and device together like this to somewhat say
"client_ip was assigned to client_mac and client_mac was connected to
switch_port":
select
ni.ip as client_ip, ni.mac as client_mac,
n.vlan, n.port as switch_port, d.name as switch,
n.switch as switch_ip from node_ip ni
join node n on ni.mac = n.mac
join device d on n.switch = d.ip
where n.active is true and ni.active is true
I say somewhat because the Netdisco data model does only store
node_ip: arp entries (ip, mac) with time_first, time_last
node: mac address table entries (mac, port) with time first, time last
but not the exact occurrences. So when you join these together, for
clients that connected to a bunch of switches and had a bunch of ips,
and time_first/last intervals which span a longer time,
you'll create ip -> port rows that might not have existed in reality.
You can mitigate this problem a bit by only looking at short time
intervals or using the active flag (which marks "current" entry-tupples
in the tables from the last run), but fundamentally you can't say with
100% confidence "this IP was on this port" over a longer period of
time... even though that's what people often seem to expect.
We've tried some approaches with reports similar to what you want to do,
but it always ended in confused users.
Nowadays we resort to presenting this information in two steps: first,
look at what macs where used with a certain IP, then look at on which
ports these macs where seen. Interpretations of "this IP was on this
port" is left to the user.
Christian
On 29/05/16 04:50, Andy Sainsbury wrote:
> Hi Everybody,
>
> Just wondering if anybody can help me out with a custom report in
> Netdisco (which is totally awesome Oliver!), I've got so far but I'm a
> bit stuck as how to proceed.
>
> I'm trying to generate a list of all the end devices on the network (by
> ip, mac and dns), and map that to a switch port and vlan, very similar
> to the
> out of the box device report, but I want to dump a big list joining all
> of our network devices together. Hope that makes sense?
>
> The pseudo output would look something like this:
>
> node_ip, node_ip.mac, node_ip.dns, subnet, device, device_ip,
> device_dns, device_port.port, device_port.vlan
>
> I can extract the data into each component just I can't work out how to
> stick it all together:
>
> Node list which is a tuple of all the data discovered on a scan:
> select node_ip
> from node_ip
> order by node_ip;
>
> Device and IP list:
> SELECT device.name <http://device.name>,
> device.ip,
> device.dns
> FROM device
> ORDER BY device.name <http://device.name>;
>
> Device and Port/Vlan list:
> SELECT device.name <http://device.name>,
> device.ip,
> device.dns,
> device_port.port,
> device_port.vlan
> FROM device_port
> INNER JOIN device ON device.ip=device_port.ip
> ORDER BY device.name <http://device.name>;
>
> And a simple subnet lookup:
> SELECT subnets.nets
> FROM subnets
> ORDER BY nets
>
> I can live without the subnet part, as I can use a perl module to later
> slot the host ip into its subnet but my skill level has reached a ceiling.
>
> Any help is much appreciated.
>
> Regards
>
> Andy Sainsbury
--- End Message ---
--- Begin Message ---
Hi Muris,
Generally bulkwalk is preferred because it's more efficient, but in some
cases it doesn't work properly: bugs in the SNMP agent on the switch,
MTU issue between switch and SNMP client (responses to GETBULK requests
can be quite large, depending on parameters used), much slower response
to a GETBULK request than to a GETNEXT request because the switch has to
do more processing.
If bulkwalk works for your gear when you increase the timeout a bit, I'd
use that; otherwise turn bulkwalk off for the problematic devices so
they'll be polled with GETNEXT requests.
Not sure yet about the other issues re LLDP etc. Am I correct that you
see two separate issues here?
1 - On the Cisco ME3600 you get "Port for IID:<x> not resolved"? Is that
during macsuck or during another operation?
2 - On the HP 5130, ports are marked as STP blocking while they are not
really STP blocking?
Regards,
Jeroen van Ingen
ICT Service Centre
University of Twente, P.O.Box 217, 7500 AE Enschede, The Netherlands
On 05/23/2016 12:41 PM, Muris wrote:
Thanks Jeroen, I will try that.
What i actually managed to do is set bulk_walk off to true, to use the
GETNEXT process to pull the data out instead of BULKWALK. This worked..
and I was able to pull data out, i think because went slower to pull out
the data..
Do you recommend using BULKWALK instead and increasing the snmp timeout
Value?
Also on another note, these 5130s in stack mode, are connected to Cisco
ME3600 distribution switches, everything between them has LLDP enabled,
but on netdisco it cant get a mapping correctly from the Cisco.
When i discover the Cisco ME3600, i get these errors
Port for IID:2 not resolved.
Port for IID:26 not resolved.
Port for IID:25 not resolved.
And these are the ports where i have the H3C/Comware switches on.
Is there a issue with Cisco ME3600 mapping to H3C/Comware switches via
LLDP? Does something need to be tweaked to get it working and map it
correctly.
When i look at the 5130, it is able to accurately get the interfaces
from the Cisco via LLDP , however it marks the ports in blue color,
which is blocking. But really it shouldnt be in blue as they are not
blocking ports.
Thanks again for your assistance.
--- End Message ---
--- Begin Message ---
Hi Jeroen, thanks again for your help.
Increasing the timeout didnt help with the bulkwalk. The HP 5130s when they
are stacked simply dont like it, and they seem to prefer the the other
method GETNEXT.
The issue is the network is 2500+ devices, turning of bulkwalk on specific
ones would be administrative nightmare specifying hostnames or IPs.
Is it possible to use the vendor model name instead in the config file to
use for GETNEXT only? For example the 5130 are listed under vendor name as
hp513024GSFP4SFPPlusEI and hp513048GPoEPlus4SFPPlus370WEI . If its possible
to use vendor model name, can you let me know how i would include the
parameter in the config file? This way when it sees this model it only does
a GETNEXT, but the rest use bulkwalk. This is for netdisco v1.
And with the ME3600, yes when i do a discover and do a debug on it, it
lists "port for IID" not resolved. Not a macsuck or arpnip, but a discover
command on it. I think its getting some kind of invalid IID recognising the
5130 on those ports.
On the 5130 yes the uplinks get marked as blocking marked in blue. Like I
have 2 uplinks, in a port channel going to the ME3600. PortChannel on the
HP they call it BridgeAggregation. The individual uplinks are marked as
blue, but the Bridge Aggregation is normal colour. BridgeAggregation doesnt
show the remote connected port channel, its just blank. However the
individual interfaces the 2 uplinks show the remote Cisco correctly and
ports, highlighted in blue however.
Thanks again for your work contributing to this magnificent tool.
On Mon, May 23, 2016 at 8:11 PM, Muris <alcat...@gmail.com> wrote:
> Thanks Jeroen, I will try that.
> What i actually managed to do is set bulk_walk off to true, to use the
> GETNEXT process to pull the data out instead of BULKWALK. This worked.. and
> I was able to pull data out, i think because went slower to pull out the
> data..
>
> Do you recommend using BULKWALK instead and increasing the snmp timeout
> Value?
>
> Also on another note, these 5130s in stack mode, are connected to Cisco
> ME3600 distribution switches, everything between them has LLDP enabled, but
> on netdisco it cant get a mapping correctly from the Cisco.
>
> When i discover the Cisco ME3600, i get these errors
> Port for IID:2 not resolved.
> Port for IID:26 not resolved.
> Port for IID:25 not resolved.
> And these are the ports where i have the H3C/Comware switches on.
>
> Is there a issue with Cisco ME3600 mapping to H3C/Comware switches via
> LLDP? Does something need to be tweaked to get it working and map it
> correctly.
>
> When i look at the 5130, it is able to accurately get the interfaces from
> the Cisco via LLDP , however it marks the ports in blue color, which is
> blocking. But really it shouldnt be in blue as they are not blocking ports.
>
> Thanks again for your assistance.
>
--- End Message ---
--- Begin Message ---
Hi Muris,
On 2016-06-02 04:07, Muris wrote:
Is it possible to use the vendor model name instead in the config file
to use for GETNEXT only? For example the 5130 are listed under vendor
name as hp513024GSFP4SFPPlusEI and hp513048GPoEPlus4SFPPlus370WEI . If
its possible to use vendor model name, can you let me know how i would
include the parameter in the config file? This way when it sees this
model it only does a GETNEXT, but the rest use bulkwalk. This is for
netdisco v1.
In Netdisco 1 you should be able to configure bulkwalk_no with
"model:hp5130.*", according to the manual:
bulkwalk_no = model:hp5130.*
see https://sourceforge.net/p/netdisco/code/ci/master/tree/README#
In Netdisco 2 the feature works similarly:
bulkwalk_no:
- 'model:hp5130.*'
You could use vendor: or model: depending which field you want to match.
In Netdisco 2 any field of the Device table in the database can be used.
see
https://metacpan.org/pod/distribution/App-Netdisco/lib/App/Netdisco/Manual/Configuration.pod#bulkwalk_no
Hope this helps,
regards,
oliver.
And with the ME3600, yes when i do a discover and do a debug on it, it
lists "port for IID" not resolved. Not a macsuck or arpnip, but a
discover command on it. I think its getting some kind of invalid IID
recognising the 5130 on those ports.
On the 5130 yes the uplinks get marked as blocking marked in blue.
Like I have 2 uplinks, in a port channel going to the ME3600.
PortChannel on the HP they call it BridgeAggregation. The individual
uplinks are marked as blue, but the Bridge Aggregation is normal
colour. BridgeAggregation doesnt show the remote connected port
channel, its just blank. However the individual interfaces the 2
uplinks show the remote Cisco correctly and ports, highlighted in blue
however.
Thanks again for your work contributing to this magnificent tool.
On Mon, May 23, 2016 at 8:11 PM, Muris <alcat...@gmail.com> wrote:
Thanks Jeroen, I will try that.
What i actually managed to do is set bulk_walk off to true, to use
the GETNEXT process to pull the data out instead of BULKWALK. This
worked.. and I was able to pull data out, i think because went
slower to pull out the data..
Do you recommend using BULKWALK instead and increasing the snmp
timeout Value?
Also on another note, these 5130s in stack mode, are connected to
Cisco ME3600 distribution switches, everything between them has LLDP
enabled, but on netdisco it cant get a mapping correctly from the
Cisco.
When i discover the Cisco ME3600, i get these errors
Port for IID:2 not resolved.
Port for IID:26 not resolved.
Port for IID:25 not resolved.
And these are the ports where i have the H3C/Comware switches on.
Is there a issue with Cisco ME3600 mapping to H3C/Comware switches
via LLDP? Does something need to be tweaked to get it working and
map it correctly.
When i look at the 5130, it is able to accurately get the interfaces
from the Cisco via LLDP , however it marks the ports in blue color,
which is blocking. But really it shouldnt be in blue as they are not
blocking ports.
Thanks again for your assistance.
------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and
traffic
patterns at an interface-level. Reveals which users, apps, and
protocols are
consuming the most bandwidth. Provides multi-vendor support for
NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports.
https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Netdisco mailing list
netdisco-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/netdisco-users
--- End Message ---
--- Begin Message ---
Hello All,
I am running with Netdisco2 and have some 'ghost devices' in my job
queue. Three device which has been already delete, appears in in the job
queue. How can I remove this queue entries permanent ?
Regards,
Oliver Buehrig
--- End Message ---
------------------------------------------------------------------------------
What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
patterns at an interface-level. Reveals which users, apps, and protocols are
consuming the most bandwidth. Provides multi-vendor support for NetFlow,
J-Flow, sFlow and other flows. Make informed decisions using capacity
planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
_______________________________________________
Netdisco mailing list - Digest Mode
netdisco-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/netdisco-users
