Symptom | Details | Resolution/Workaround |
Solaris 10 10/08 does not support FMA on Intel Nehalem |
Solaris cannot diagnose CPU, memory or I/O faults on Sun systems based on Intel Nehalem architecture. LOM FMA can diagnose some but not all faults
|
Upgrade to Oracle Solaris 10 05/09 (S10 U7) or later |
Solaris 10 10/09 hangs with SATA packet timeout messages on console. |
System hangs during normal operation after a week or more. pon rebooting, the console messages file has a number of SATA packet timeouts logged for any (or all) of the onboard SATA disks. e.g. Example error:
Nov 18 20:57:57 nsgvlan339-39-119 ahci: [ID 517647 kern.warning] WARNING: ahci0: watchdog port 1 satapkt 0xffffffff88a841c8 timed out
|
Workaround: Disable CPU deep C-state support by adding following line to /etc/system and reboot: set idle_cpu_no_deep_c = 1
Solution: Update to Oracle Solaris 10 8/10 (S10 U10) or later
|
System may not to auto reboot once AC is restored. |
In the event of a planned or unplanned system AC cycle, due to a marginal timing issue between the BIOS & iLOM, there is potential for the system not to auto reboot once AC is restored. A manual intervention such as a power button assertion or remote, SP based power on assertion will be necessary in this case to reboot the host. |
Fixed in Sun Fire X2270 ILOM FW 3.0.3.35 r49758 / BIOS vers. 1.14.1.1 or later |
ILOM KVMS fails to launch |
When redirection of the ILOM BUI is launched via the Remote Control -> Redirection tab and is successful, the Sun ILOM Remote Control window displays "Connected to XX.XX.XX.XX" (where XX.XX.XX.XX is the IP address of the service processor). If this issue occurs, the window will display "Logging into XX.XX.XX.XX" indefinitely, and the redirection will fail to complete. Also, if this issue occurs, previously connected redirection windows will appear to hang for between 20 and 40 minutes. This issue is more likely to occur under the following circumstances: 1) There is significant TCP network congestion on a connection used by at least one of the redirection clients. 2) network disruption occurs, including one of the following: a) Intermittent cable disconnect of an intermediate cable, b) rKVM client crash or hang, or c) IP change on the SP. |
Workaround: If redirection is not successful within 30 seconds, either: 1) wait up to 40 minutes and try again or reset the SP prior to retrying. 2) Reset the SP via the ILOM BUI: Maintenance->Reset SP tab. 3) Reset the SP via the CLI: (via SSH or serial console), cd to /SP and issue a reset command. Solution: Fixed in Sun Fire X2270 ILOM FW 2.0.2.18 r45343 or later |
SP might go into a non-responsive state |
The problem is that an internal sequencer/clock in the ADM1026 has stalled. In the 3.3VSTBY voltage comparator section, there is a calibration circuit which keeps the internal reference voltage stable. When this clock stops, the analog to digital converters stop working which freezes sensor updates which is not fatal. |
Fixed in Sun Fire X2270 ILOM FW 3.0.3.35 r47987 or later |
Upgrading firmware to SW 2.2.0 resets all ILOM settings to defaults |
Upgrading the firmware on an X2270 system to firmware bundle SW 2.2.0 (ILOM 3.0.9.18 build r55822 / BIOS 2.03) via the web GUI results in all ILOM settings resetting to defaults even if the "preserve settings" option is ticked. Root password resets back to a default of "changeme" and network settings are reset back to dhcp etc. It is possible that the password may also reset back to an old password instead of "changeme" depending on actions taken. |
Workaround: Upgrading the firmware via the CLI does not seem to have this issue and all settings are preserved when the option is selected. |