Operational Defect Database
BugZero found this defect 46 days ago.
Data sources
All data on this page is proprietary to BugZero® or gathered from public sources
Hewlett Packard Enterprise | a00138603en_us
Advisory: (Revision) HPE M-Series Switches - When Two ONYX Switches are Configured in a Multi-Chassis Link Aggregation (MLAG), They are Unable to Form a Master / Standby Connection
Last update date:
4/4/2024
Affected products:
HPE M-series SN2010M Switch
HPE M-series SN2100M Switch
HPE M-series SN2410M Switch
HPE M-series SN2700M Switch
HPE M-series SN3700M Switch
HPE Storage Switch M-series SN4600M
Affected releases:
No affected releases provided.
Fixed releases:
No fixed releases provided.
Description:
Info
Document Version Release Date Details 2 April 4, 2024 Added "ONYX" to the Title and revised the Affected Products list to include HPE Storage Switch M-series SN4600M. 1 April 3, 2024 Original Document Release. A failure may occur when forming a Multi-Chassis Link Aggregation (MLAG) master/standby connection in configurations where an intermediate switch is used to connect the M-series switches mgmt0 ports. The M-series MLAG configuration requires MDNS packet exchanges over the mgmt0 ports for purposes of "heartbeat" and status communications in addition to Standard Boot Order (IPL) communications. When the mgmt0 MDNS information fails to be exchanged, the two M-series switches display as master/master or master/unknown; although the switch configuration for MLAG and MLAG-VIP are correct, and the mgmt0 ports are able to ping to each other. A constant heartbeat exchange is required through the mgmt0 ports of two switches to establish a MLAG-VIP master/standby connection. The switches in MLAG constantly exchange MDNS packets, i.e., multicast traffic sent to multicast address 224.0.0.251. However, if the intermediate switch connecting the M-series mgmt0 ports does not permit this multicast IP address to be forwarded, MLAG cannot resolve the master/standby roles. See the example below:
Scope
HPE M-Series ONYX switches in an MLAG configuration are affected by this advisory: HPE SN2100M 100GbE 16QSFP28 Half Width Switch (Q2F23A) HPE SN2100M 100GbE 8QSFP28 Half Width Switch (Q2F24A) HPE SN2100M 100GbE 16QSFP28 Power to Connector Airflow Half TAA-compliant (R0P75A) HPE SN2100M 100GbE 16QSFP28 Connector to Power Airflow Half TAA-compliant (R0P76A) HPE SN2010M 25GbE 18SFP28 4QSFP28 Switch (Q9E63A) HPE SN2010M 25GbE 18SFP28 4QSFP28 Power to Connector Airflow TAA-compliant (R0P77A) HPE SN2010M 25GbE 18SFP28 4QSFP28 Connector to Power Airflow TAA-compliant (R0P78A) HPE SN2700M 100GbE 32QSFP28 Switch (Q2F21A) HPE SN2700M 100GbE 32QSFP28 Power to Connector Airflow TAA-compliant (R0P71A) HPE SN2700M 100GbE 32QSFP28 Connector to Power Airflow TAA-compliant (R0P72A) HPE SN2700M 100GbE 16QSFP28 Switch (Q6M26A) HPE SN2410M 25GbE 48SFP28 8QSFP28 Switch (Q2F22A) HPE SN2410M 25GbE 48SFP28 8QSFP28 Power to Connector Airflow TAA-compliant (R0P73A) HPE SN2410M 25GbE 48SFP28 8QSFP28 Connector to Power Airflow TAA-compliant (R0P74A) HPE SN2410bM 10GbE 48SFP+ 8QSFP28 Switch (Q6M28A) HPE SN2410M 25GbE 24SFP28 4QSFP28 Switch (Q6M27A) HPE SN2410bM 10GbE 24SFP+ 4QSFP28 Switch (Q6M29A) HPE SN3700cM 100GbE 32QSFP28 Power to Connector (P2C) Airflow Switch (R3B14A) HPE SN4600cM 100GbE 64QSFP28 Power to Connector (P2C) Airflow Switch (R6R24A) HPE SN3700M 200GbE 32QSFP28 Power to Connector (P2C) Airflow Switch (R5Z74A)
Resolution
The intermediate switch or device to which the mgmt0 ports of M-series switches in an MLAG configuration are connected must allow 224.0.0.251 (MDNS protocol) through the intermediate switch regardless of any multicast protocols that are running (igmp, igmp-snooping, etc.) Contact the device vendor to address this issue. If assistance is needed to analyze the TCP dump outputs from the tcpdump commands, contact your local country HPE Customer Support or log a case via the HPE Support Center . To debug the issue, the tcpdump command can be used. One instance on each switch is as follows: On one switch, “switch a”: tcpdump vrf mgmt -i mgmt0 host 224.0.0.251 or host <IP address of mgmt0 on switch b> -c 50 On the other switch, “switch b”: tcpdump vrf mgmt -i mgmt0 host 224.0.0.251 or host <IP address of mgmt0 on switch a> -c 50 The output of those two tcpdump commands should show whether MDNS is being sent/received properly between the two mgmt0 interfaces. This output can also be provided to the HPE ERT team to analyze. When MDNS packets are being exchanged through an intermediate switch, the show mlag-vip command or the show mlag command displays the working mlag-vip master/standby connection: RECEIVE PROACTIVE UPDATES : Receive support alerts (such as Customer Advisories), as well as updates on drivers, software, firmware, and customer replaceable components, proactively in your e-mail through HPE Support Alerts. Sign up for Support Alerts at the following URL: HPE Email Preference Center
