NMS-18059: Fix LLDP handling for SNMP agents from Planet and Microsense

[rssntn67]

This PR regards an issue related to couple of devices supporting lldp MIB.

The devices are Microsense and Planet.

In particular Planet reports mac addresses using the - So mac address 00:00:00:00:00:00 is reported ad 00-00-00-00-00-00.
So first of all I moved all the methods to decode from snmp agent into SnmpLldpUtils class. I guess that is a better place, before where in the tracker classes. Because of weird snmp agent implementation I have moved all the stuff converting snmp data into proper string in one place.

The Microsense also has wrong decoding of the LLDP mib and data needs to be properly decoded. Inparticular Hex Strings are coded as Strings.

finally I have reworked the LldpServiceImpl this class is fundamental to produce topology used in topology maps. I have reworked the class to let find links in more ways. Also to support Microsense to Microsense Lldp links beceuase I found that there is a bug in the lldp oimplementation reporting the wrong port on sides.

I ave run all the IT test of enlinkd and everything is fine.

Also I changed some Thread.sleep statements from 1000ms to 200ms to speed up test running

https://opennms.atlassian.net/browse/NMS-18059

[cgorantla]

@rssntn67 Can you update description on what you are trying to fix in this PR.

I see some common utils getting refactoring to LldpSnmpUtils. Other than that what else is important to review here ?

Also we require a JIRA issue for this contribution. Let me know if you need any help here.

[marshallmassengill]

I appreciate this contribution @rssntn67! I really appreciate the work on the "formatMacAddress" function. Will be good to see that working. Let us know if you need any help with the items @cgorantla mentioned.

[rssntn67]

I appreciate this contribution @rssntn67! I really appreciate the work on the "formatMacAddress" function. Will be good to see that working. Let us know if you need any help with the items @cgorantla mentioned.

I have updated the informations. It is a work I've done a coule of months ago. I guess it is all.
The main problem with planet and micro sense is decondig properly the mac addresses they provide.

[rssntn67]

@rssntn67 Can you update description on what you are trying to fix in this PR.

I see some common utils getting refactoring to LldpSnmpUtils. Other than that what else is important to review here ?

Also we require a JIRA issue for this contribution. Let me know if you need any help here.

Done

[marshallmassengill]

@rssntn67, how are the MAC addresses stored in the database? Is it as collected from the devices or are they decoded and sanitized before storing?

[rssntn67]

@rssntn67, how are the MAC addresses stored in the database? Is it as collected from the devices or are they decoded and sanitized before storing?

Well, they are decoded ad sanitized and the final format is "001122334455", but on the other end if the sanitation does not work I store the string. In the case of planet the data was formatted as a string:
00-11-22-33-44-55, and I add to the method now also the replace for - with blanck, as is was for ":" and so on.

[christianpape]

Only a few minor things to change...

[christianpape]

LGTM.

[cgorantla]

why is this custom null ?

[rssntn67]

It is simply a way of returning the value in the database.
It looks better in UI.

[cgorantla]

Is it better to throw exception instead of custom return values or use Optional ?

[rssntn67]

It is simply a way of returning the value in the database.
It looks better in UI.
If I throw the exception then I have to forgot the data.
Too many weird Snmp implementation of Lldp.
The most important is having data saved.

[cgorantla]

LGTM!

Thanks for your contribution.