Flashes (Alerts)
Abstract
IBM has discovered a potential migration issue that can affect some users of the IBM MQ Appliance M2002 models, who are migrating from a 9.2 firmware level to a 9.3.0 or 9.3.1 firmware level.
Content
This issue affects M2002 appliances configured in a high availability (HA) group, where only some of the HA queue managers are configured to use a floating IP address. IBM strongly recommends that IBM MQ Appliance M2002 users with this configuration defer planned 9.3 migration activities until a fix for this issue is available.
The issue does not affect:
- Appliances that are not configured in an HA group
- Appliances where no floating IP addresses are configured
- Appliances where all HA queue managers are configured to use a floating IP address
To check if HA queue managers are configured with a floating IP address, use the dspmq -o ha command from the mqcli. If a floating IP address is configured, a non-blank value is displayed for the HALSADDR and HALSINT attributes. In the following example, HAQM1 and HAQM3 have floating IP addresses defined, while HAQM2 does not:
QMNAME(HAQM1) HA(Replicated) HALSADDR(172.20.52.2) HALSINT(eth14)
QMNAME(HAQM2) HA(Replicated) HALSADDR() HALSINT()
QMNAME(HAQM3) HA(Replicated) HALSADDR(172.20.52.4) HALSINT(eth14)
Symptoms
On encountering the issue, HA queue managers for which the appliance is the preferred location do not automatically start on the appliance. A failure data capture (FDC) file is created in the mqerr: filesystem when the appliance reboots after installing the new firmware. The FDC file header reports probe ID XC022001 and Major Errorcode aplE_HA_INIT_FAILED, similar to the following:
+-----------------------------------------------------------------------------+
| |
| IBM MQ First Failure Symptom Report |
| ========================================= |
| |
| Product Long Name :- IBM MQ Appliance |
| Vendor :- IBM |
| Installation Name :- MQAppliance (1) |
| Probe Id :- XC022001 |
| Application Name :- MQM |
| Component :- xcsDisplayMessage |
| Source filename :- /build/jslot0/p930_P/src/lib/cs/amqxeida.c |
| Effective UserID :- 0 (root) |
| Real UserID :- 0 (root) |
| Program Name :- mqdpinit |
| Arguments :- -d /dev/sdb |
| Thread :- 1 |
| Thread Name :- ProcessInit |
| UserApp :- FALSE |
| Last HQC :- 0.0.0-0 |
| Last HSHMEMB :- 0.0.0-0 |
| Last ObjectName :- |
| Major Errorcode :- aplE_HA_INIT_FAILED |
| Minor Errorcode :- OK |
| Probe Type :- INCORROUT |
| Probe Severity :- 2 |
| Probe Description :- AMQ6125E: An internal IBM MQ Appliance error has |
| occurred. |
| FDCSequenceNumber :- 0 |
| |
+-----------------------------------------------------------------------------+
[some lines removed for brevity]
Attempting to move queue managers to the current appliance manually, using the sethagrp -u [qmgr] command, reports:
High availability configuration of [qmgr] already migrated
The sethagrp command succeeded
but the queue manager does not start on the local appliance.
Users who have already migrated and are experiencing these symptoms should contact IBM support for assistance in resolving the issue.
IBM is preparing a fix for this issue under APAR IT42853, which will be released on IBM Fix Central when complete. The affected 9.3 firmware images have been withdrawn from Fix Central while these updates are prepared.
[{"Type":"MASTER","Line of Business":{"code":"LOB36","label":"IBM Automation"},"Business Unit":{"code":"BU053","label":"Cloud \u0026 Data Platform"},"Product":{"code":"SS5K6E","label":"IBM MQ Appliance"},"ARM Category":[{"code":"a8m0z00000008L0AAI","label":"Migration"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"9.2.0;9.3.0"}]
Product Synonym
IBM MQ Appliance M2002; IBM MQ Appliance M2002A; IBM MQ Appliance M2002B
Was this topic helpful?
Document Information
Modified date:
10 January 2023
UID
ibm16853397