Code: Select all
Jun 15 12:44:18 orazone01 emlxs: [ID 349649 kern.info] [ 1.037A]emlxs0:WARNING:1540: Firmware update required. (To trigger an update, a manual HBA or link reset using luxadm, fcadm, or emlxadm is required.)
Jun 15 12:44:23 orazone01 emlxs: [ID 349649 kern.info] [ 1.037A]emlxs1:WARNING:1540: Firmware update required. (To trigger an update, a manual HBA or link reset using luxadm, fcadm, or emlxadm is required.)
Jun 15 12:46:00 orazone01 usba: [ID 691482 kern.warning] WARNING: /pci@400/pci@1/pci@0/pci@b/pci@0/usb@0,2/hub@2/hub@3 (hubd2): Connecting device on port 3 failed
Jun 15 12:47:25 orazone01 snmpXdmid: Registration with DMI failed. err = 831.
For the following info messages, I need to follow the steps:
Code: Select all
Jun 15 12:44:18 orazone01 emlxs: [ID 349649 kern.info] [ 1.037A]emlxs0:WARNING:1540: Firmware update required. (To trigger an update, a manual HBA or link reset using luxadm, fcadm, or emlxadm is required.)
Jun 15 12:44:23 orazone01 emlxs: [ID 349649 kern.info] [ 1.037A]emlxs1:WARNING:1540: Firmware update required. (To trigger an update, a manual HBA or link reset using luxadm, fcadm, or emlxadm is required.)For the following warning, I need to make some changes from ILOM:Warning:1540: Firmware Update Required. (A Manual Hba Reset Or Link Reset (Using Luxadm Or Fcadm) Is Required [ID 1356876.1] To BottomTo Bottom
Modified:Mar 5, 2013Type:PROBLEMStatus:PUBLISHEDPriority:3
There are no commentsComments (0) Rate this document Email link to this document Open document in new window Printable Page
In this Document
Symptoms
Cause
Solution
Applies to:
Sun Storage FC HBA - Version Not Applicable and later
Information in this document applies to any platform.
Symptoms
Errors seen in /var/adm/messages
Sep 3 23:12:52 <hostname>emlxs: [ID 349649 kern.info] [ 1.0340]emlxs0:WARNING:1540: Firmware update required. (A manual HBA reset or link reset (using luxadm or fcadm) is required.)
Sep 3 23:12:54 <hostname> emlxs: [ID 349649 kern.info] [ 1.0340]emlxs1:WARNING:1540: Firmware update required. (A manual HBA reset or link reset (using luxadm or fcadm) is required.)
Cause
Beginning with emlxs driver version 2.50p, the driver no longer automatically updates the firmware upon reboot. The user must issue a 'luxadm -e forcelip' to trigger the firmware update. This behavior maybe changed in the future eliminating the need for this extra step.
To confirm driver version:
From Explorer:
Check out the modinfo.out file in the Oracle explorer output under the "sysconfig" directory
grep emlxs modinfo*
modinfo-c.out: 83 1 emlxs LOADED/INSTALLED
modinfo.out: 83 7b600000 3e07f8 275 1 emlxs (SunFC emlxs FCA v20110324-2.60k)Check out the fcinfo.out file in the Oracle explorer output under the "sysconfig" directoryCode: Select all
orazone01:/opt/SUNWexplo/output/explorer.85b0bd58.orazone01-2013.06.12.23.24/sysconfig%grep emlxs modinfo* modinfo-c.out: 97 1 emlxs LOADED/INSTALLED modinfo.out: 97 7b600000 3b5de8 317 1 emlxs (SunFC emlxs FCA v20100108-2.50o)
egrep -i "OS Device Name:|Model:|Firmware Version:|FCode/BIOS Version:|Driver" fci* | sort | uniq -c
2 fcinfo.out: Driver Name: emlxs
2 fcinfo.out: Driver Version: 2.60k (2011.03.24.16.45)
2 fcinfo.out: FCode/BIOS Version: fcode: 1.16;
2 fcinfo.out: FCode/BIOS Version: Boot:5.03a8 Fcode:3.10a3
2 fcinfo.out: Firmware Version: 03.03.28
2 fcinfo.out: Firmware Version: 2.82a4 (B3F2.82A4)
2 fcinfo.out: Model: LP11002-S
1 fcinfo.out: OS Device Name: /dev/cfg/c1
1 fcinfo.out: OS Device Name: /dev/cfg/c2
ORCode: Select all
orazone01:/opt/SUNWexplo/output/explorer.85b0bd58.orazone01-2013.06.12.23.24/sysconfig%egrep -i "OS Device Name:|Model:|Firmware Version:|FCode/BIOS Version:|Driver" fci* | sort | uniq -c 2 fcinfo.out: Driver Name: emlxs 2 fcinfo.out: Driver Version: 2.50o (2010.01.08.09.45) 2 fcinfo.out: FCode/BIOS Version: Boot:5.03a0 Fcode:3.01a1 2 fcinfo.out: Firmware Version: 1.11a5 (U3D1.11A5) 2 fcinfo.out: Model: LPe12000-S 1 fcinfo.out: OS Device Name: /dev/cfg/c1 1 fcinfo.out: OS Device Name: /dev/cfg/c4
On Live system:
#modinfo | grep emlxs
83 7b600000 3e07f8 275 1 emlxs (SunFC emlxs FCA v20110324-2.60k)
# modinfo -c | grep emlxsCode: Select all
orazone01:%modinfo | grep emlxs 100 7b69a000 3f4a20 317 1 emlxs (emlxs FCA v20120917-2.80.8.0)
83 1 emlxs LOADED/INSTALLED
#fcinfo hba-port | egrep "OS Device Name:|Model:|Firmware Version:|Fcode/BIOS Version:|Driver" | sort | uniq -cCode: Select all
orazone01:%modinfo -c | grep emlxs 100 1 emlxs LOADED/INSTALLED
2 Driver Name: emlxs
2 Driver Version: 2.60k (2011.03.24.16.45)
2 Firmware Version: 03.03.28
2 Firmware Version: 2.82a4 (B3F2.82A4)
2 Model: LP11002-S
1 OS Device Name: /dev/cfg/c1
1 OS Device Name: /dev/cfg/c2
SolutionCode: Select all
orazone01:%fcinfo hba-port | egrep "OS Device Name:|Model:|Firmware Version:|Fcode/BIOS Version:|Driver" | sort | uniq -c 2 Driver Name: emlxs 2 Driver Version: 2.80.8.0 (2012.09.17.15.45) 2 Firmware Version: 1.11a5 (U3D1.11A5) 2 Model: LPe12000-S 1 OS Device Name: /dev/cfg/c1 1 OS Device Name: /dev/cfg/c4
# luxadm -e forcelip /dev/cfg/c1
# luxadm -e forcelip /dev/cfg/c2
NOTE: luxadm -e forcelip is intrusive. Do not issue this command while there is activity down the paths. Halt I/O to the targets on the paths that require the firmware upgrade prior to issuing the luxadm -e forcelip command.
NOTE: luxadm -e forcelip may also be required for a "firmware downgrade" in the event that the firmware detected on the HBA is different than the current patch level firmware in the OS.
Beginning with Solaris 11, fcadm may also be used to perform the firmware upgrade.Code: Select all
# luxadm -e forcelip /dev/cfg/c1 # luxadm -e forcelip /dev/cfg/c4
#fcadm hba-port
Make note of the WWN of the HBA port(s) that require a firmware upgrade.
#fcadm force-lip <Port_WWN>
Code: Select all
Jun 15 12:46:00 orazone01 usba: [ID 691482 kern.warning] WARNING: /pci@400/pci@1/pci@0/pci@b/pci@0/usb@0,2/hub@2/hub@3 (hubd2): Connecting device on port 3 failedCompleted on 06/19/2013USB warning "Connecting device on port 3 failed" on SPARC T-series servers during Solaris boot after installing April 2013 Solaris SPARC Critical Patch Update (CPU) [ID 1552329.1]
Symptoms
T3-2 reports the following warning during Solaris 10 boot:
usba: [ID 691482 kern.warning] WARNING: /pci@400/pci@1/pci@0/pci@b/pci@0/usb@0,2/hub@2/hub@3 (hubd2): Connecting device on port 3 failed
This error does not cause any loss of functionality to the front or rear USB ports on the T3-2.
Changes
Applying the April 2013 Solaris SPARC CPU (Critical Patch Update)
Cause
This error is related to an internal usb ILOM interconnect interface used by Oracle Hardware Management Pack (OHMP) to communicate between Solaris and the ILOM System Controller (SC). If you do not have Oracle Hardware Management Pack installed on your system, you will need to change the hostmanaged property in ILOM to false as a workaround to receiving the USB warning shown above.
Solution
1. Log in to the Oracle ILOM SP CLI.
Establish a local serial console connection or SSH connection to the server SP or CMM.
2. Navigate to the /SP/network/interconnect working directory on the server using the cd command.
-> cd /SP/network/interconnect
3. View the network interconnect targets and properties using the show command.
-> show
/SP/network/interconnect
Targets:
Properties:
hostmanaged = true
type = USB Ethernet
ipaddress = 169.254.182.76
ipnetmask = 255.255.255.0
spmacaddress = 02:21:28:57:47:16
hostmacaddress = 02:21:28:57:47:17
Commands:
cd
set
show
Note to TSC: If you gather a snapshot output from the customer, you can view the current setting of the hostmanaged property under the /ilom directory in the @usr@local@bin@spshexec_show_-d_properties_-level_all_@.out file.[This section is not visible to customers.]
4. Change the hostmanaged property to false.
-> set hostmanaged=false
Upon the next Solaris reboot, you should no longer see the USB warning.
For more information, please see the below ILOM 3.0 document:
Configure Local Interconnect Interface Between Server SP and Host OS (CLI)
As shown in the document linked above, "When you set the hostmanaged property to true, you must also install the Oracle Hardware Management Pack 2.1.0 (or later) software on your server and accept the installation default for enabling Local ILOM Interconnect. "
For the following error, a patch is needed:
Code: Select all
Jun 15 12:47:25 orazone01 snmpXdmid: Registration with DMI failed. err = 831.Completed on 06/19/2013Error on agent startup snmpXdmid: [ID xxxxx daemon.error] Registration with DMI failed. err = 831. [ID 1329827.1]
In this Document
Symptoms
Changes
Cause
Solution
Applies to:
Solaris SPARC Operating System - Version 10 5/09 U7 and later
Solaris x64/x86 Operating System - Version 10 5/09 U7 and later
Information in this document applies to any platform.
***Checked for relevance on 16-Nov-2012***
Symptoms
A error messages is received from snmpXdmid when it is started on a Solaris 10 system.
The message is "snmpXdmid: [ID xxxxx daemon.error] Registration with DMI failed. err = 831."
Changes
1. The patch to address CR 7010213 has not been applied and the snmpXdmid file in questions has a expired date.
2. The patch to address CR 7010213 has been applied but the customer was not provided the information to clear some snmpXdmid database files that store expired date information.
Cause
The message "snmpXdmid: [ID xxxxx daemon.error] Registration with DMI failed. err = 831." can occur for two reasons.
1. The patch to address CR 7010213 has not been applied and the snmpXdmid file in questions has a expired date.
2. The patch to address CR 7010213 has been applied but the customer was not provided the information to clear some snmpXdmid database files that store expired date information.
CR 7010213 snmpXdmid.conf WARNING_TIMESTAMP and EXPIRATION_TIMESTAMP are now in the past
dmispd creates a database from the config files on its first run. After installing the patch, the database will contain the old data. The entry needs to be cleared from the database.
Solution
The following steps need to be executed as root on the system.
The first step can be omitted if the patch 138361-02 or later has already been installed.
# patchadd 138361-02
# svcadm disable dmi
# rm /var/dmi/db/*
# svcadm enable dmi