BGP Sessions - Wrong IPv4 Local IP

Hi There!

Im new to LibreNMS, just working with for a few days in a test environment.

The Local Address on BGP IPv4 sessions shows 0.0.127.252 for the 4 BGP Sessions I have. Peers IPv4 addresses and all IPv6 addresses are show correctly. Only Local for IPv4 shows wrong.

The Updates are also wrong. In the graph of updates for a single session I see:
image

But for the same session, in the general tab of BGP I get:

Updates (in) 5.530M (out) 18.000

The router is an ASR1k2X.

Thanks in advance, if everything goes well, LibreNMS will replace shortly our OpenNMS installation.

Regards,

Hi again!

Should I send a bug with this in the bug tracker?

By the way, here is the validate.php

====================================

Component Version
LibreNMS 1.32-18-ge25f473
DB Schema 210
PHP 7.0.22-0ubuntu0.16.04.1
MySQL 10.0.31-MariaDB-0ubuntu0.16.04.2
RRDTool 1.5.5
SNMP NET-SNMP 5.7.3

====================================

[OK] Database connection successful
[OK] Database schema correct

./discovery.php -r -f -h HOSTNAME -d

and use pastebin.com to post your output please

Here is!

https://pastebin.com/9N2ap2Ab

Thanks in advance!

Hi Kevin,

Did you had time to take a look to this? I’ve searched for 0.0.127.252 in the output of the discovery and cant find it. I had been also looking for bgpupdates but cant find them…

Thanks!

what vendor is this router?

Lol, sorry. Cisco ASR1002-X. Also know as 1k2X :slight_smile:

@murrant or @FTBZ
yall have any idea on this?

I’m using two ASR920 with BGP without problems, but the value 0.0.127.252 is strange. It’s probably a wildcard mask and not the correct information. Is that value used somewhere in your router configuration?

Can you post the full polling of the device (removing confidential data first, but keeping the 0.0.127.252)?

1 Like

Hi @FTBZ,

I’ve sended you the pastebin in pm. There are a lot of sensitive information to be “removed” so I prefer this if you dont mind.

Thanks!

Perhaps 0.0.127.252 is an erroneous database entry? maybe try deleting it from the db directly.

Hi @murrant,

No, its not. @ftbz have the last pastebin from a poller.php -v and I found where the problem is, but I dont know how to fix it (as i said to FTBZ).

Here is a little extract:

cbgpPeer2LocalAddr.ipv4."10.50.2.230" = "00 00 7F FC "

Is taking the hexa response instead of the index for the localaddr.

Thanks!

Copy here of my private message:

[e[0;31mRRD Disablede[0m]SQL[e[0;33mUPDATE `bgpPeers` set `bgpPeerState` ='established',`bgpPeerAdminStatus` ='start',`bgpPeerFsmEstablishedTime` ='313705',`bgpPeerInUpdates` ='1486066',`bgpLocalAddr` ='0.0.127.252',`bgpPeerOutUpdates` ='3' WHERE `device_id` = '2' AND `bgpPeerIdentifier` = '10.x.x.x'e[0m]

I compared your poller with my ASR poller for BGP and the result are different. The local address seems returned by the OID bgpLocalAddr. Your device is returning 0.0.127.252 for this OID, so for LibreNMS this is the corrected information from the router.

On my ASR, the bgpLocalAddr is returning the correct local BGP address.

So it’s probably a specific configuration problem on your ASR or an IOS-XE bug. I don’t think that we can help more here :confused: . Perhaps you can parse your router configuration file and check if 0.0.127.252 exists.

After my post I found that: https://quickview.cloudapps.cisco.com/quickview/bug/CSCva76268

Hi @FTBZ,

Thank you very much. Im not using that versioin but its very probably that mine is affected too.

What about the BGP Updates? Do you want me to open another thread for this?

Regards,

IMHO this is the same problem, the update is returned in the same time/request of the bgpLocalAddr. So if the address is wrong the update will be wrong too.

I dont thing so, as with cacti BGP updates are showed correctly.

Also, there are discrepancies inside LNMS.

In the main eBGP tab, I have for one peer:
image

And, the same peer in the Updates tab:
image

Cacti, for same peer:
image