v1.4.8rcX Bug reports and comments

DOWNLOAD THE LATEST FIRMWARE HERE
User avatar
Eric Stern
Employee
Employee
 
Posts: 532
Joined: Wed Apr 09, 2014 9:41 pm
Location: Toronto, Ontario
Has thanked: 0 time
Been thanked: 130 times

Re: v1.4.8rcX Bug reports and comments

Thu Jun 08, 2017 3:01 pm

The management VLAN should not be in the monitored VLANs. It could potentially be used for the Control VLAN, but my understanding is that most configurations used a separate dedicated VLAN for control.

User avatar
mike99
Associate
Associate
 
Posts: 837
Joined: Tue Nov 25, 2014 10:53 am
Location: Quebec, Canada
Has thanked: 95 times
Been thanked: 245 times

Re: v1.4.8rcX Bug reports and comments

Thu Jun 08, 2017 3:40 pm

CCM won't connect if we add MEP without specifing the port.



Code: Select all
20# show erps
MEP 1
ID: 20
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 0
VLAN: 4080
Peer ID: 00-00-00-00-00-00
CCM Status: Down
20#
21# show erps
MEP 1
ID: 21
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 0
VLAN: 4080
Peer ID: 00-00-00-00-00-00
CCM Status: Down
21#
22# show erps
MEP 1
ID: 22
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 0
VLAN: 4080
Peer ID: 00-00-00-00-00-00
CCM Status: Down


Just by adding the ports, CCM session goes up:
20# show erps
MEP 1
ID: 2001
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 13
VLAN: 4080
Peer ID: ec-13-b2-61-43-36
CCM Status: Up
MEP 2
ID: 2002
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 14
VLAN: 4080
Peer ID: ec-13-b2-61-b4-5c
CCM Status: Up
20#
20#
21# show erps
MEP 1
ID: 21
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 13
VLAN: 4080
Peer ID: ec-13-b2-61-13-f6
CCM Status: Up
Code: Select all
21#
22# show erps
MEP 1
ID: 22
MD: distribution
MA: ISP1
Level: 5
CCM Interval: 10 ms
Port: 14
VLAN: 4080
Peer ID: ec-13-b2-61-13-f6
CCM Status: Up

User avatar
mike99
Associate
Associate
 
Posts: 837
Joined: Tue Nov 25, 2014 10:53 am
Location: Quebec, Canada
Has thanked: 95 times
Been thanked: 245 times

Re: v1.4.8rcX Bug reports and comments

Thu Jun 08, 2017 3:46 pm

Eric Stern wrote:The management VLAN should not be in the monitored VLANs. It could potentially be used for the Control VLAN, but my understanding is that most configurations used a separate dedicated VLAN for control.


On my config, management VLAN is 5 and control VLAN 4080. Those are separeted but control VLAN is not related to my problem.

All switch share a management VLAN to access those switch. How can I avoiding loop on management VLAN if I can't add the management VLAN to the ring ? Even with a management VLAN by switch (what would make no sense), I still have the same problem.

User avatar
mike99
Associate
Associate
 
Posts: 837
Joined: Tue Nov 25, 2014 10:53 am
Location: Quebec, Canada
Has thanked: 95 times
Been thanked: 245 times

Re: v1.4.8rcX Bug reports and comments

Thu Jun 08, 2017 3:57 pm

Here's a drawing so it will be easier to understand. As soon as VLAN 5 is part of the monitored VLAN list, I lost access via port 11 but switch can still be ping from one to a other so it's seem to affect every port outside of the ring.
Attachments
ERPS management lost.PNG

User avatar
Eric Stern
Employee
Employee
 
Posts: 532
Joined: Wed Apr 09, 2014 9:41 pm
Location: Toronto, Ontario
Has thanked: 0 time
Been thanked: 130 times

Re: v1.4.8rcX Bug reports and comments

Thu Jun 08, 2017 4:04 pm

You should not be able to add a MEP without specifying a port. I'll fix that.

Its my understanding that ERPS is used on customer VLANs, you would use STP on the management VLAN/port.

User avatar
mike99
Associate
Associate
 
Posts: 837
Joined: Tue Nov 25, 2014 10:53 am
Location: Quebec, Canada
Has thanked: 95 times
Been thanked: 245 times

Re: v1.4.8rcX Bug reports and comments

Thu Jun 08, 2017 4:49 pm

Can't find anything about it anywhere. Is It this from vitesse documentation ?

The only information about running both in the same port I can find is the following from Juniper:
Note: Spanning-tree protocols and ERPS cannot both be configured on a ring port. Because RSTP is the spanning-tree protocol enabled in the default switch configuration, this example shows disabling RSTP on each ring port before configuring ERPS. If another spanning-tree protocol is enabled, you must disable that first instead.
Source: https://www.juniper.net/documentation/e ... eries.html

Anyway, STP/RSTP is not a option since those could block the ring by disabling the whole port. Is MSTP priority fix ? I tested a little and is seemed like not.

User avatar
mike99
Associate
Associate
 
Posts: 837
Joined: Tue Nov 25, 2014 10:53 am
Location: Quebec, Canada
Has thanked: 95 times
Been thanked: 245 times

Re: v1.4.8rcX Bug reports and comments

Fri Jun 09, 2017 9:50 am

Ingnitenet Fusionswtich have a great documentation for ERPS.
https://ignitenet.uservoice.com/knowled ... er-manuals

At page 476, you can see limitation of ERPS. One of those limitation is:
Exclusive use of STP or ERPS on any port.

A other one is that control VLAN should not be management VLAN and should not have IP address. The CVLAN should also contain only ring port (so exclude of trunk). Ring can't be member of a trunk or LACP enabled port.

A other limitation is max 255 data VLAN but with QinQ support , that should not be a problem.

User avatar
tonym
Member
 
Posts: 31
Joined: Thu Nov 19, 2015 1:42 pm
Location: Havelock, IA
Has thanked: 0 time
Been thanked: 7 times

Re: v1.4.8rcX Bug reports and comments

Fri Jun 16, 2017 2:25 pm

I've been working with LLDP-MIB and CISCO-CDP-MIB and these two are not properly implemented.

The indexes are incorrect. All entries should have this index structure, please see LLDP-MIB
Code: Select all
# LLDP-MIB::lldpRemPortId.234523.20.1
# .1.0.8802.1.1.2.1.4.1.1.7[.234523.20.11]
[.lldpRemTimeMark.lldpRemLocalPortNum.lldpRemIndex]


From the MIB:
lldpRemEntry OBJECT-TYPE
INDEX { lldpRemTimeMark, lldpRemLocalPortNum, lldpRemIndex }

http://www.ieee802.org/1/files/public/M ... 60000Z.txt

CISCO-CDP-MIB has the same issue, but i different indexes:
cdpCacheEntry OBJECT-TYPE
INDEX { cdpCacheIfIndex, cdpCacheDeviceIndex }

https://raw.githubusercontent.com/libre ... CO-CDP-MIB

User avatar
sirhc
Employee
Employee
 
Posts: 7416
Joined: Tue Apr 08, 2014 3:48 pm
Location: Lancaster, PA
Has thanked: 1608 times
Been thanked: 1325 times

v1.4.8rc7 Released

Sat Jun 17, 2017 1:22 pm

- Fixed SNMP support for Q-SWITCH-MIB, LLDP and CDP - RC7

v1.4.8rcX wrote:NOTE: MSTP is still being tested so use MSTP at your own risk.

FIXED/CHANGED
- Fixed SNMP support for Q-SWITCH-MIB, LLDP and CDP - RC7
- Changed temp sensor readings on DC models to effect fan speeds - RC2
- Grey out unavailable bulk poe options - RC3
- Upper out of voltage range for IDC is now 54V - RC4

ENHANCEMENTS
- Added ERPS protocol implementation is g.8032v2 - Ring Protection - RC1
- Added BETA note to ERPS - RC3
- Added port name to "show interface status" CLI command - RC3
- ADD an input voltage calibration for DC models - RC4
- Improved Q-BRIDGE-MIB support - RC5
- improved fan control on DC models - RC6
- improved stability of ERPS- RC6

KNOWN ISSUES
- Some language templates need help - please private message Eric Stern to help

v1.4.8rc1 Released 5/9/2017
v1.4.8rc2 Released 5/11/2017
v1.4.8rc3 Released 5/15/2017
v1.4.8rc4 Released 5/23/2017
v1.4.8rc5 Released 5/26/2017
v1.4.8rc6 Released 6/7/2017
v1.4.8rc7 Released 6/17/2017
Support is handled on the Forums not in Emails and PMs.
Before you ask a question use the Search function to see it has been answered before.
To do an Advanced Search click the magnifying glass in the Search Box.
To upload pictures click the Upload attachment link below the BLUE SUBMIT BUTTON.

User avatar
tonym
Member
 
Posts: 31
Joined: Thu Nov 19, 2015 1:42 pm
Location: Havelock, IA
Has thanked: 0 time
Been thanked: 7 times

Re: v1.4.8rcX Bug reports and comments

Mon Jun 19, 2017 11:25 am

Sorry I didn't catch this in the first pass. I found lldpRemManAddrTable is incorrect.

lldpRemManAddrSubtype and lldpRemManAddr should be inaccessible. (only used in the index)
lldpRemManAddrIfSubtype, lldpRemManAddrIfId, and lldpRemManAddrOID should be available (read-only) (If there is no data for these entries, they should return default data)

There seems to be a .4 missing from the index (I'm not an expert at reading mibs, so I'm not sure what this is) but all vendor implementations I tested are consistent.
Code: Select all
lldpRemManAddrIfSubtype
1.0.8802.1.1.2.1.4.2.1.3.0.24.1.1.4.10.36.68.10 = 1


The rest of lldp is looking good. I'll be reviewing cdp soon.

PreviousNext
Return to Hardware and software issues

Who is online

Users browsing this forum: No registered users and 25 guests