Ps5 OID's

[Frotihngdog.ca]

OID's for PS5 Station-WDS...if anyone is interested

Signal: 1.3.6.1.4.1.14988.1.1.1.1.1.4.4
TxRate: 1.3.6.1.4.1.14988.1.1.1.1.1.2.4
RxRate: 1.3.6.1.4.1.14988.1.1.1.1.1.3.4
Hostname: 1.3.6.1.4.1.14988.1.1.1.1.1.5.4
Channel: 1.3.6.1.4.1.14988.1.1.1.1.1.7.4
MAC of AP: 1.3.6.1.4.1.14988.1.1.1.1.1.6.4
Station MAC: 1.2.840.10036.1.1.1.1.4
Firmware Version: 1.2.840.10036.3.1.2.1.4.4

[JustJoe]

Thanks! ... That's very useful.

I have used SNMP quite a bit .... But by no means am I any sort of expert! ;-)

1) In addition to Signal, is there an OID for Noise ?

2) Now that the firmwares for LSx, PSx and NSx have converged at V3...

...could UBNT post all the supported OIDs in one place on the web site ???

[Frotihngdog.ca]

There's no OID for noise YET! :) But I believe they are working on that.

Personally I hope they put their efforts towards the PS9's and the SR71's first. I can do without graphing noise for now. Really REALLY can't wait for the PS9's and the SR71's.

Keep up the GREAT work guys!!

[Frotihngdog.ca]

Ignore this sticky. The OID's listed are not static and change when the system is rebooted or updated. UBNT is working is looking into it.

[UBNT-keba]

OIDs changes just last number in OID. It is interface index. And it changes on each soft restart (apply changes). And I'd like to say that it will not we fixed in v3.1 for sure. It is not trivial task. Sorry guys, wait for v3.2.

[netmaster]

OIDs changes just last number in OID. It is interface index. And it changes on each soft restart (apply changes). And I'd like to say that it will not we fixed in v3.1 for sure. It is not trivial task. Sorry guys, wait for v3.2.

but why they changes? I have never seen behavior like that, or if some OID changes, then there must be some special OID what say a current lucky number. Is there? I have tried to write for our support staff some readonly way to look status over SNMP, but soon realize, that with dynamic OID's, SNMP is too hard to use. Much better and more predictable way, was to parse html output from web pages (Show Stations etc...) .

[UBNT-kesha]

of course there is -- if you would look closely onto MIB descriptions you use, you would see that they reference interface index. You can fetch all available interfaces and their IDs using IF-MIB (http://www.mibsearch.com/vendors/RFC/mibview/IF-MIB), then choose interface you need and use it's index in other MIBs - see supported MIBs in http://ubnt.com/forum/viewtopic.php?t=582 .

On "why they change" - the most of the interfaces are recreated during soft-restart, hence getting new index numbers, some of them (e.g. ppp) even can change at runtime - and all of this is considered "normal". yes, not so comfortable to SNMP users, but that's what SNMP is all about.

In the future we might try to do some semi-static index mapping, however, this needs careful consideration, as we shouldn't break SNMP systems, which already do interface lookups by their names from IF-MIB (which is pretty common for solid SNMP monitoring systems)

[netmaster]

is there any way to find from AP, correspondending IP address or unit name from client MAC address? All CPE's broadcasts CDP periodically, including name's and ip's, but is that information stored somewhere in AP?

Would be great, if "Show Station" or SNMP will show that information too.

[tonymobile]

OID's for PS5 Station-WDS...if anyone is interested

Signal: 1.3.6.1.4.1.14988.1.1.1.1.1.4.4
TxRate: 1.3.6.1.4.1.14988.1.1.1.1.1.2.4
RxRate: 1.3.6.1.4.1.14988.1.1.1.1.1.3.4
Hostname: 1.3.6.1.4.1.14988.1.1.1.1.1.5.4
Channel: 1.3.6.1.4.1.14988.1.1.1.1.1.7.4
MAC of AP: 1.3.6.1.4.1.14988.1.1.1.1.1.6.4
Station MAC: 1.2.840.10036.1.1.1.1.4
Firmware Version: 1.2.840.10036.3.1.2.1.4.4

Is possible OID for Noise ????

Is possible to use this OIDs in NS5???


Thanks.

[drwho17]

I'm looking for some additional snmp information from the Ubiquities. I'm using the latest 3.3.2 firmware. In AP mode I'm looking for the noise floor, and I'm also looking for the client association time, both of which are available in the Ubiquity web interface.

[WHT]

Are you looking for the ambient noise floor of the RF landscape or the RX sensitivity?

[drwho17]

Well, I'm looking for the value that is displayed under "show stations", which appears to match the "Noise Floor" value shown in the main page. I want to flag bad margin's. I've rebuilt the "show station" tool on my stats page, added the username/ip address as well (from my redback), just need the last column, and association time, via snmp preferably, I'm sure I could get the values via ssh.

[dGlueguy]

I see there "should" be an OID for noise floor now, but I've never seen it posted.

Care to post it here?

[dGlueguy]

Have they fixed the stupid OID indexing so that the OIDs stay put? I don't see anything about this in the release notes.

[drwho17]

Have they fixed the stupid OID indexing so that the OIDs stay put? I don't see anything about this in the release notes.
It's not a problem, it's the way SNMP works. Some OID's change dynamically.

[dGlueguy]

Some OID's change dynamicallyNot on any of the other systems we use.

Not clear what the point is. It just seems idiotic to me.

[drwho17]

Because you can't have OID"s for dynamic interfaces, such as wifi interfaces for example. How are you supposed to know what OID to check for a wireless client that may not be on? Do you have any Cisco's/Redbacks/Lucent Stingers? They have some static OID's like for eth0 or eth1, but they even will change if you add new gear while they are live, and then later reboot. What are you trying to monitor that is changing anyway? I can probably tell you how to monitor it.

[dGlueguy]

Well, any of the NS or PS radios are only certified as PTP devices, so there can only legally be one wireless interface. At least in the US.

And on our PtMP legal devices, we just check the CPE directly.

[WHT]

Well, any of the NS or PS radios are only certified as PTP devices,Can you expand on that and how they are not legal PtMP devices.

The FCC is the exclusive regulatory authority in the U.S. and does not make any certification distinction between PtP and PtMP use of the NS, PS, Bullet, Pico, or Loco product line.

[dGlueguy]

The FCC is the exclusive regulatory authority in the U.S. and does not make any certification distinction between PtP and PtMP use of the NS, PS, Bullet, Pico, or Loco product line.Actually, they do.

They make a distinction with regard to EIRP limts, which are different for PTP and PtMP devices.

In addition, you just have to refer to the documentation supplied by Ubiquity to the FCC in this document: (https://fjallfoss.fcc.gov/prod/oet/forms/blobs/retrieve.cgi?attachment_id=909562&native_or_pdf=pdf)

You will see at the very bottom, it explicitly states "This is a Point-to-Point CPE Wireless Device."

So the PTP limitation is explicitly stated in the document submitted to the FCC.

[WHT]

It does not say it is exclusively restricted for use as a PtP device, it merely says it is one.

They make a distinction with regard to EIRP limts, which are different for PTP and PtMP devices.Thats not relevant here, its only relevant to FCC 15.247 that defines the usage between PtP and PtM.

[dGlueguy]

Whatever.

You read it your way, I'll take the conservative route and stay legal.

Until it says multipoint, or facsimile thereof in the FCC filing, it will only be used PTP in our network.

[Frotihngdog.ca]

If you use Cacti and the UBNT templates that are I available it works just fine. You need to monitor the interface directly not the OID.

[dGlueguy]

Well, we do use Cacti along with the published templates. But the RRD files are filled with NANs.

Which prompted me to start looking at the OIDs. Which is where I found each one on a different index number.

Got a link to the most current UBNT templates?

[WHT]

Mike...

Can you jump in here and offer an explaination for the disclaimer on the FCC's version of the user guide that says "This is a Point-to-Point CPE Wireless Device."

Its fairly apparent (at least to me) this was added as an after thought to the original user guide.

While I realize the FCC places the responsibility for proper operation with the installing party; it is very, very unlike that the installing party would even check for FCC certification documents.

On a side note, there is no way the NS2 could exceed the 36 dBm EIRP with max TPO of 26 dBm and 10 dBi antenna.

Furthermore the NS5 with the internal antenna can not be used at all on the UNII-2 band which has a 30 dBm EIRP limit *and* 6 dBi antenna restriction.

[dGlueguy]

...and it gets even shakier with something like the PS2 that has 26 db TPO plus a 17 db antenna putting it in the 43db EIRP area.

[WHT]

Just back off the TPO on the PS2 by 2 or 3 dBm and you'll satisfy the "add three, subtract one" rule.

[Frotihngdog.ca]

dGlueguy

templates dated: 20090210
CactiEZ: 0.6 : http://cactiez.cactiusers.org/download/

I had no problems. Everything graphed properly

[oojoshua]

Hey, this post is kind of confusing. Can you please keep the cross talk to a different thread?

OK, I figured out how to get the dynamic OID and thought I would share:

snmpwalk -v 1 -c ur_community_string ur_hostname .1.3.6.1.2.1.2.2.1.2

This returns a list of interfaces:

IF-MIB::ifDescr.1 = STRING: lo
IF-MIB::ifDescr.2 = STRING: tunl0
IF-MIB::ifDescr.3 = STRING: gre0
IF-MIB::ifDescr.4 = STRING: eth0
IF-MIB::ifDescr.5 = STRING: wifi0
IF-MIB::ifDescr.8 = STRING: ath0
IF-MIB::ifDescr.9 = STRING: br0

In my case I am wanting signal, Rx Rate, Tx Rate, Channel and noise floor. (UBiQUiTi Staff - We'd all like the OID for the noise floor!!!!) so I am interested in ath0 (8).

To get this I'll just parse the results:

snmpwalk -v 1 -c ur_community_string ur_hostname .1.3.6.1.2.1.2.2.1.2 | grep ath0 | head -c 17 | tail -c 1

This command returns "8" - The interface I actually want.

I can now use this ID to get info for my interface:

signal:

snmpget -v 1 -c ur_community_string ur_hostname 1.3.6.1.4.1.14988.1.1.1.1.1.4.8

Here is a perl script that can be used to get this info. You can use it in Cacti using the Script/Command option, or just run it from the command line.

#!/usr/bin/perl

# Stats Grabber for UBNT
# - Joshua Montgomery
# - http://www.lawrencefreenet.org
# - 6/28/09
#
# Description:
# Gathers data in interger format for the following variables:
# * signal
# * txrate
# * rxrate
# * channel

$argv0 = $ARGV[0]; # Host
$argv1 = $ARGV[1]; # Community String
$argv2 = $ARGV[2]; # Query
$string;

# First get the ID of the interface we care about.
$id = `snmpwalk -v 1 -c $argv1 $argv0 .1.3.6.1.2.1.2.2.1.2 | grep ath0 | head -c 17 | tail -c 1`;

if ($argv2 eq "signal") {
$string = `snmpget -v 1 -c $argv1 $argv0 1.3.6.1.4.1.14988.1.1.1.1.1.4.$id`;
$string =~ s/.*: //;
$string =~ s/\n$//;
print $string;
}
elsif ($argv2 eq "txrate") {
$string = `snmpget -v 1 -c $argv1 $argv0 1.3.6.1.4.1.14988.1.1.1.1.1.2.$id`;
$string =~ s/.*: //;
$string =~ s/\n$//;
print $string;
}
elsif ($argv2 eq "rxrate") {
$string = `snmpget -v 1 -c $argv1 $argv0 1.3.6.1.4.1.14988.1.1.1.1.1.3.$id`;
$string =~ s/.*: //;
$string =~ s/\n$//;
print $string;
}
elsif ($argv2 eq "channel") {
# Channels
# 2412 = 1
# 2417 = 2
# 2422 = 3
# 2427 = 4
# 2432 = 5
# 2437 = 6
# 2442 = 7
# 2447 = 8
# 2452 = 9
# 2457 = 10
# 2462 = 11
$string = `snmpget -v 1 -c $argv1 $argv0 1.3.6.1.4.1.14988.1.1.1.1.1.7.$id`;
$string =~ s/.*: //;
$string =~ s/\n$//;
if ($string eq "2412") {
print 1; }
elsif ($string eq "2417") {
print 2; }
elsif ($string eq "2422") {
print 3; }
elsif ($string eq "2427") {
print 4; }
elsif ($string eq "2432") {
print 5; }
elsif ($string eq "2437") {
print 6; }
elsif ($string eq "2442") {
print 7; }
elsif ($string eq "2447") {
print 8; }
elsif ($string eq "2452") {
print 9; }
elsif ($string eq "2457") {
print 10; }
elsif ($string eq "2462") {
print 11; }