Sparc CPU chart

CPU Type

 SPARC64-VII+ (M3)  T3 T4 T5 M5  M10/SPARC64-X

Servers

M3000, M4000,
M5000, M8000,
M9000,
M9000-64

T3-1, T3-2
T3-4 

T4-1, T4-2
T4-4
T5-2, T5-4
T5-8
M5-32 M10-1, M10-4, 
M10-4S

CPU Core Codename

Jupiter++ S2 S3 S3 S3 Athena
CPU Clockfrequency 2.66-3 GHz 1.65 GHz 2.85-3GHz 3.6 GHZ 3.6 GHZ  2.8-3 GHz
Number of Cores per socket 4 16

 8 

16 6 16
Number of Threads per core 2 8

 8 

8 8 2
Minimum amount of RAM,
number of sockets,
and threads per Server

 8GB
1 socket
8 threads

8GB
1 socket
128 threads

 16GB
1 socket
64 threads

256GB
2 sockets
256 threads
1TB
8 sockets
384 threads
32GB
1 socket
4 threads (2 cores activated from 16)
Maximum amount of RAM, number

of sockets and threads per Server

4TB
64 sockets
512 threads 

128 GB
4 sockets
512 threads

2TB
4 sockets
256 threads 

4TB
8 sockets
1024 threads
32 TB
32 sockets
1536 threads
32TB
64 sockets
2048 threads
L3 Cache no L3$
(12MB L2$ though!)
no L3$

4MB

8MB 48MB no L3$
(24MB shared L2$ though!)

Virtualization supported

Dynamic Domains
(and Zones)

LDoms
(and Zones)

LDoms
(and Zones) 

LDoms
(and Zones) 
Hard Domains, 
LDoms
(and Zones) 
LDoms
(and Zones) 

CPU architecture (ISA)

sun4u sun4v sun4v sun4v sun4v sun4v

 OS supported

Solaris 10, 11 Solaris 10, 11 Solaris 10, 11 Solaris 10, 11 Control Domain: S11 only.
Guest Domains: Solaris 10, 11
Solaris 10, 11

noteworthy features

electronically separated domains 512 threads in 2010! First S3 based, single thread AND throughput 1024 threads in 8 RU
Mainframe class,
Mission critical
Software on Chip

Make Watchdog reset force a Coredump

When a system becomes unresponsive, without any messages provided for diagnosis of the problem, the deadman kernel code can be enabled to force a core dump during the systems unresponsive period. However if the system is equipt with an ALOM, there is a function known as the ALOM watchdog which will reset them system according to set perameters when the system goes unresponsive. By default the watchdog will just power cycle the system, however this gathers no addition as to the cause of the unresponsive period for the system.

You can set the watchdog to dump a corefile however with the following command:

set sys_autorestart=dumpcore

This effectively will turn the LOM watchdog into the deadman timer.

System won’t boot. But I have an OK prompt

System won’t boot. But I have an OK prompt

1. Can you run the banner?
ok banner

2. Can the Machine boot from CDROM into Single User Mode?
#boot cdrom -s
– YES – but fails to boot to the boot disk, then there is a software problem with the OS on the boot disk.
– NO – then it is most likely a Hardware Problem.

a. Verify that the boot cd is a supported OS for this system.
b. Try to boot the exact same OS CD as is loaded on the boot disk.

3. Have any changes been made to the system since it last booted correctly?
– Patches applied, Packages added, Hardware installed or removed

4. Are you getting past the banner and ok prompt?

5. Can you see your boot device from the ok prompt?
ok setenv auto-boot? false
ok reset
ok probe-scsi-all
-or-
ok probe-ide-all

6. Are you attempting to boot from your default boot device?
ok printenv boot-device
ok devalias {output from previous command}
ok boot -v
– Now just watch to make sure that the output from the devalias command matches the device that the system is attempting to boot from.

7. Is the system automatically attempting to boot from your diag device (net by default)?
ok printenv diag-device
ok setenv diag-switch? false

NOTES:

A. If the diag-switch OBP variable is set to true, then the machine will attempt to boot from the diag-device which is usually different from the default boot device. Set the diag-switch to false.

B. If the system hangs at the "twirly bird" ( | / – \ ) or the configuring devices statement, then there may be hardware on the system that is not recognized or is failed.

C. If the system comes up and stops at the maintenance prompt (press control-d for normal start-up or enter root password for system maintenance) the files systems need to be fsck’ed. Enter the root password and fsck the filesystems.

Enter Root Password: ******
#fsck -y /dev/rdsk/cXtXdXsX

D. If the root drive is encapsulated or under control of a volume management utility, then the issue needs to be handled as a storage issue.

SSH and ALOM V240

SSH and ALOM V240 / V210

Unfortunately there is no support for SSH in 1.x Versions, so there is
no way to configure or workaround it.

The network management port does not support ssh,The ALOM is very simple
and only understands basic protocols.

DCE to DTE serial connections

DCE equipment is typically a modem or other type of communication device. The DCE sits between the DTE (data terminal equipment) and a transmission circuit such as a phone line. Originally, the DTE was a dumb terminal or printer, but today it is a computer, or a bridge or router that interconnects local area networks. In an IBM mainframe environment, a communication controller and a link-attached cluster controller are examples of DTEs.

A DCE provides a connection for the DTE into a communication network and back again. In addition, it terminates and provides clocking for a circuit. When analog telephone lines are the communication media, the DCE is a modem. When the lines are digital, the DCE is a CSU/DSU (channel service unit/data service unit).

DTE and DCE interfaces are defined by the physical layer in the OSI (Open Systems Interconnection) model. The most common standards for DTE/DCE devices are EIA (Electronic Industries Association) RS-232-C and RS-232-D. Outside the United States, these standards are the same as the V.24 standard of the CCITT (Consultative Committee for International Telegraphy and Telephony). Other DTE/DCE standards include the EIA RS-366-A, as well as the CCITT X.20, X.21, and V.35 standards. The later standards are used for high-speed communication over telephone lines.

DTE and DCE devices send and receive data on separate wires that terminate at a 25-pin connector. It is useful to know that DTE devices transmit on pin connector 2 and receive on pin 3. DCE devices are just the opposite-pin 3 transmits and pin 2 receives.

For pinouts see:
http://www.dracko.com/index.php?s=page&n=show_post&id=417

The serial pin designations were originally intended for connecting a dumb terminal to a modem. The terminal was DTE (Data Terminal Equipment) and the modem was DCE (Data Communication Equipment). Today the PC is usually used as DTE instead of a terminal (but real terminals may still be used this way). The names of the pins are the same on both DTE and DCE. The words: "receive" and "transmit" are from the "point of view" of the PC (DTE). The transmit pin from the PC transmits to the "transmit" pin of the modem (but actually the modem is receiving the data from this pin so from the point of view of the modem it would be a receive pin).

The serial port was originally intended to be used for connecting DTE to DCE which makes cabling simple: just use a straight-thru cable. Thus when one connects a modem one seldom needs to worry about which pin is which. But people wanted to connect DTE to DTE (for example a computer to a terminal) and various ways were found to do this by fabricating various types of special null-modem cables. In this case what pin connects to what pin becomes significant.

Set up V490 service processor

The rscadm utility and its subcommands allow administration of Sun Remote System Control (RSC) from the host. You must log in to the server as root to use rscadm.

Installation places the rscadm utility in the following directory:

/usr/platform/platform-name/rsc/

You can obtain the string to use for platform-name using the Solaris command
uname -i.

resetrsc [-s]

This subcommand resets RSC immediately. To terminate all connections cleanly before the reset, use the -s option. If no argument is supplied, this subcommand performs a hard reset and drops all connections.

Note – When you reset RSC on a Sun Enterprise 250 server without also resetting the server, RSC time defaults to 1/1/70. To synchronize RSC time with server time, reset the server, use the rscadmcommand rscadm date -s, or run the script /usr/platform/platform-name/rsc/rsc-initscript. You can obtain the string to use for platform-nameusing the Solaris command uname -i.

Redirect console to RSC

To use the RSC card the system administrator needs to redirect the console to the RSC. Certain faults are undiagnosable if the console log at the time of failure is not captured. Once the RSC is setup as the console device this vital data is captured automatically. In addition the system can be powered off and on remotely making remote management much easier.

After downloading and installing the RSC packages setup the RSC card from Solaris using the rsc-config command.

/usr/platform/`uname -i`/rsc/rsc-config

Confirm you are able to login to the RSC card via the RSC serial or the network connection. Once remote login to the RSC card is working you can setup the console redirection. From the ok prompt run the following commands, have a session open to the console on the RSC card as once you reset-all the POST output will be output to the console.

ok setenv diag-out-console true

ok setenv input-device rsc-console

ok setenv output-device rsc-console

ok reset-all

The same changes could be made from Solaris using the eeprom command, the changes would take affect at the next reboot.

# eeprom diag-out-console=true

# eeprom input-device=rsc-console

# eeprom output-device rsc-console

Switching the console back to TTYA

If you need to redirect the console back to TTYA there are a number of options. OBP 4.15.x introduced “Service mode”, after shutting down the system turn the keyswitch to the diagnostics position and power on. The console will be redirected to the ttya serial port until the keyswitch is returned to normal and the system is reset. This was introduced to allow engineers sent to site to more easily work on the systems without requiring knowledge of the customer RSC passwords.

Alternatively if a more persistent solution is required you can set the individual OBP variables back to the defaults:

ok setenv diag-out-console false

ok setenv input-device keyboard

ok setenv output-device screen

If access to Solaris and the RSC card is not available as a last resort the RSC card can be removed. This will temporarily reset the console output to the TTYA port. Once the OBP has been started, the parameters will need to be reset appropriately until the RSC card has been configured.

Hot Swap/Pluggable Hard Drives

Hot Swap/Pluggable Hard Drives

System Hot Swap/Pluggable
====== ==================
Sun Fire V210 Hot Swap
Sun Fire T2000 Hot Pluggable
Sun Fire 20z Hot Swap
Sun Fire V240 Hot Swap
Sun Fire V40z Hot Swap
Sun Fire V440 Hot Pluggable
Sun Fire X2100 Hot Pluggable
Sun Fire X4100 Hot Swap
Sun Fire X4200 Hot Swap
Sun Fire V480/V490 Hot Pluggable
Sun Fire V440 Hot Pluggable
Enterprise 220R Hot Pluggable
Enterprise 250 Hot Pluggable
Enterprise 3500 Hot Swap
Netra 440 Hot Pluggable