Booting x86 with serial console by default

Let's say I have a x86 box with no monitor and no keyboard. I'd like to boot the x86 Solaris 10 installation dvd and have it booted with the 'serial console' option. However, it's not selected by default and I can't select it without using the keyboard.
Is it possible to have the x86 installation DVD boot with serial console support by default?

Well, https://wiki.archlinux.org/index.php/Re … nstall_ISO doesn't help, genisoimage created iso is still not bootable.
The iso created with Archiso boots but - it creates an x86_64 kernel only...
Last edited by piwwo (2013-06-14 13:55:31)

Similar Messages

Ultra sparc IIe insalling solaris with serial console

I bought a sparc and I'm trying to install solaris 10. It came with solaris installed on a 15gb hard drive, and I want to replace the hard drive.
As it came, when I boot, it says that no keyboard was detected and that ttya will be used for input/output. Then X starts and I can use the USB keyboard and mouse.
When attempting to install, with a blank hard drive installed, I get the same message, then the screen goes black. So, I've hooked up a terminal emulator running on linux (cutecom), and I can see messages on the console, but I can't seem to get a prompt. It says OpenBoot etc. then bad magic number in disk label, then "Timeout waiting for ARP/RARP packet" over and over.
Is there some key stroke I need to send to get a prompt? I've tried various keystrokes.

sid_sanders wrote:
yes. do you have the install media (cd/dvd)? if so, at the ok prompt type: boot cdrom -s
you will get # prompt and can label the disk using format. once that is done, you can do init 0. at ok prompt do
boot cdrom - nowinMy question is: how do I get to the prompt? Is there a keystroke I must send? I don't get a prompt. I had tried typing "boot cdrom - text" and I get no response.

Serial Console on x86 Platform

After several hours of searching online documentation I have been unable to find any information on using the serial port as the console on x86. I'm working on drivers for x86 and of my realmode drivers panics during the install. The system then reboots itself. The messages scroll to fast to read, but if I could get serial console working, I would have them saved off. If anyone know to get this working, I would appreciate a hint. If it's even supported.
-bmb-

Recently, I faced a similar situation that I have to get serial console from x86 machine on sparc machine.
By setting up a tipline, you can access a console from x86 remotely.
1. Setting on x86
First, you should set "serial console" on x86 Configuration Assistant Screen.
Please refer to docs.sun.com(http://docs.sun.com).
Solaris 7:
"x86: Menus Displayed During
the Device Identification Phase"
in System Administration Guide, Volume I
Solaris 8:
"Menus Displayed During the Device
Identification Phase"
in System Administration Guide, Volume I
As other way, you can use the following eeprom command also,
or you can edit /boot/solaris/bootenv.rc file directly to have an effect permanently.
(in case of using COM1)
# eeprom input-device=ttya
# eeprom output-device=ttya
2. Get console output
If you connect serial cable between x86 machine and sparc machine and executed the tip command,
# tip hardwire
you can get console output from x86.
As default, you can connect COM1(x86) to serial port A(sparc). If you use port B on sparc, you have to modify /etc/remote file.
If you can not succeed, there is a possibility that the serial cable is not suitable.
Although I've not tried between two x86 machines, the same way might be applicable.
Thanks,
- Hidenobu

Booting with mode CLI and not with the console.

hi,
On a SF300-08, is anybody knows how to work with CLI and not with the Console Menu Interface ?
Thank you for your help.
Pascal Linglin

Hi Pascal.
Firstly make sure your switch is using the most recent firmware, which is now as of yesterday 1.1.2.0 .
When that firmware is running on the switch, a textview CLI option should be seen on the main menu as per the following screen capture;
to get back to the menu, just type menu from the cli interface.
Hope i answered your question?
regards Dave

Install over serial console

Hi all,
I've recently purchased an Ultra5 that I'm using to learn a bit more about SPARC Solaris (previously used x86). However, my cheapo PC VGA monitor can't handle the Sun resolutions, particularly the one used by the console on boot-up (everything from the OpenBoot banner to dtlogin, at which point I can use my monitor). I'm intending to update the system from Solaris 8 to 9 once I get my media pack - but I'll have to do it via the serial console.
Has anyone here installed Solaris via the serial console - if so, what should I expect, and are there any particular "gotchas" ? I've used the OpenWindows-based setup on x86 and SPARC, but I've never encountered the command-line version. Does it follow the same steps, just as text - or are there major differences ?
Any information or pointers would be much appreciated!
Thanks in advance,
-Mark

As long as your cursor keys (arrows up, down, left & right) and Tab key work you should be all set.
boot cdrom - w (space on both sides) will get you started with "no windows" option. It's a lot faster since you don't have to load up WebStart or OpenWin from the CD. Don't bother with Install CD. Laying out filesystems can be a little trickey but you'll get the hang of it. Ctrl d is down a page & Ctrl u is up a page if you get into large menus.

Please Help :( iMac 20" refuses to boot up with anymore than 1gb ram?

Please Help iMac 20" refuses to boot up with anymore than 1gb ram? It was fine 4 days ago and had been working with a 1gb stick in one slot and a 2gb stick, which had replaced the orignal 1gb stick that I got it with, in the other and it had shown 3gb of ram and was working just fine with that since November? It went to sleep briefly 4 days ago while I was waiting on a customer and I came back to try to wake it up and it never would forcing me to hard boot it. When I did that it would only go to a black screen with a solid white light on the power indicator. Over and over again and same thing. I read and tried every suggestion I could possibly find in this and a few other mac forums to no avail.
I finally after reading about RAM issues etc. I decided to power off and take out the ram sticks and try them one at a time in each slot. So I took out the 2gb stick and did the SMC reset and it would boot up and work just fine with the 1 gb of ram. So then I went through the same process with the 2gb stick and it would not boot up only go to a black screen with white solid light again (NOT flashing). So then I tried the other original 1gb I had replaced with the 2gb and it worked just find like the other 1gb. So then I tried just 1gb stick in the other ram slot (back facing one) and with just the 1gb strip it booted up and worked fine. Tried it with the other 1gb stick and once again that one too worked just fine. So I determined it wasn't the slot itself I thought? I then tried the 2gb in the back slot and it would not boot only go to the black screen white solid light deal like before. So I thought well ok it's obviously the 2gb ram must have somehow went bad (even though I have only had it since November) and I thought well I will just use the (2) 1gb sticks in each slot and then that will at least give me 2gb because only 1gb really ***** LOL. So I proceeded to go through the same shut down process installed both sticks (I always push them to I feel them click and my fingers have a nice indention on them) reset the SMC and turned the power one and.... It would only go to the black screen with white solid light on power indicator???? I went through the same above process at least 3 times and again today a few times and I can't figure it out and I tried to search and search and I couldn't find anything like what I am experiencing.....What happened? Please help this slow Imac is killing me. It will work with 1gb ram only nothing more? Here are the specifics of my Imac.
It is a 20" Late 2006 Intel Core 2 Duo 2.16 ATI Radeon X1600 256 MB graphics card and Ram with now only 1gb (did have 3gb) and I am running OSX Lion 10.7.5
When I go to system report and go under memory it shows this:
BANK 0/DIMM0:
Size:          1 GB
Type:          DDR2 SDRAM
Speed:          667 MHz
Status:          OK
Manufacturer:          0xCE00000000000000
Part Number:          0x4D342037305432393533435A332D43453620
BANK 1/DIMM1:
Size:          Empty
Type:          Empty
Speed:          Empty
Status:          Empty
Manufacturer:          Empty
Part Number:          Empty
Serial Number:          Empty
Any help I can get would really be appreciated. Thanks in advance
ALSO...Here is also a pic of the system log right before it happened. It even changed the date to 07/2014????
I have some more if needed???????????????
<Personal Information Edited by Host>

The Memtest just completed and came back like this? It only test the single 1gb chip because anything more in this computer it won't go past the black screen with solid white power light.
Memtest version 4.22 (64-bit)
Copyright (C) 2004 Charles Cazabon
Copyright (C) 2004-2008 Tony Scaminaci (Macintosh port)
Licensed under the GNU General Public License version 2 only
NOTE: No command-line arguments have been specified
Using defaults: Test all available memory, one test pass, no logfile
Mac OS X 10.7.5 (11G63) running in multiuser mode
Memory Page Size: 4096
System has 2 Intel core(s) with SSE
Requested memory: 250MB (263086080 bytes)
Available memory: 250MB (263086080 bytes)
Allocated memory: 250MB (263086080 bytes) at local address 0x0000000101000000
Attempting memory lock... locked successfully
Partitioning memory into 2 comparison buffers...
Buffer A: 125MB (131543040 bytes) starts at local address 0x0000000101000000
Buffer B: 125MB (131543040 bytes) starts at local address 0x0000000108d73000
Running 1 test sequence... (CTRL-C to quit)
Test sequence 1 of 1:
Running tests on full 250MB region...
Stuck Address       : ok
Linear PRN          : ok
Running comparison tests using 125MB buffers...
Random Value        : ok
Compare XOR         : ok
Compare SUB         : ok
Compare MUL         : ok
Compare DIV         : ok
Compare OR          : ok
Compare AND         : ok
Sequential Increment: ok
Solid Bits          : ok
Block Sequential    : ok
Checkerboard        : ok
Bit Spread          : ok
Bit Flip            : ok
Walking Ones        : ok
Walking Zeroes      : ok
All tests passed! Execution time: 362 seconds.
logout
[Process completed]
Seems Okay I guess?

No access to serial console in ACS appliance 111

We have 2 Cisco ACS appliances running version ...
Cisco Secure ACS 3.2.2.5
Appliance Management Software 3.2.2.5
Appliance Base Image 3.2.2.1
The fact is that after initial setup, we have never used the console mainly because in a production environment we manage them through the Web Admin application. Now we have decided to upgrade both appliances to the latest version (3.3.3) and when we tried to connect to the serial console (115200,N,8,1, no flow control) we don't get any response from none of both ACS. It's quiet strange but we have found no way to make them work. We have tried several things I expose to you in case you can give us any hint:
1. We have rebooted the appliance and we can see through the console all the start-up process but when it finally finishes the start-up, we see no login prompt.
2. We have also shutdown the appliance properly and power it off and on again. Same results. The appliances boot normal but still we don't have console access.
3. We have tried boot the appliance with the recovery CD-ROM and the console works fine. I can reset the Admin password, but when it restart from its own system ( I mean without the recovery CD_ROM), I can see all the starting messages but when it finish the start-up process ... no console access.
4. Finally I have connected a monitor and a keyboard to the appliance ( I know Cisco dosn not recommned it but when in trouble....) and I see the full start-up process and it includes the base Windows 2000 server operating system startup. When Windows finishes loading, we get a lock screen in which the appliance informs you that it have started correctly and that we could access it for management through the serial console port or through the web console. 10 seconds later I see a pop up window stating that on or more services have not started correctly and that we shoulkd check the Event viewer, something we wished we could do but as you you, this is a secured system and I don't know if there is a back door method to verify windows services in this appliance.
Any help would be appreciated, as the problem is identical in both the appliances and upgrading them without access to the admin console is difficult and risky.
Kind regards.

Hi
I had similair problem being locked out of console after initial configuration wizard.
I think there is a bug within the console session in that if you input a hostname of more than 15 characters, it locks up the ACS service when the server reboots. If you keep your hostname to less than 15 characters, the server reboots and you get console access. If you then access the GUI, you will see that 15 characters is the maximum, and you cannot enetr any more than this. This is not the case with the console, where you can enter more than 15 without getting an error message.
I rescued the server by doing F8 and rebooting server with last known good configuration. from there, you can reset the hostname to something valid. You can check to see which CS services are running through console session, and start any services that may not be running..
deliverance1> start CSAgent
Starting service: CSAgent..
CSAgent is starting
CSAgent is running
Regards
Ian

Help - Custom Jumpstart 10/09 x86 with ZFS

Hi,
I'm trying to Jumpstart Solaris 10/09 x86 with ZFS root paritition. Everything seems to be going okay. My problems are
1. The Server does not automatically boot after completing jumpstart
2. After installation, everytime I reboot the newly jumpstarted machine, I see the following messages on the console, that I don't see with regular installation
# init 6
propagating updated GRUB menu
File </boot/grub/menu.lst> propagation successful
File </etc/lu/GRUB_backup_menu> propagation successful
File </etc/lu/menu.cksum> propagation successful
File </sbin/bootadm> propagation successful
Do you know why I see this progagating message?
My rules.ok file is
root@dev # cat rules.ok
any - - my_profile -
# version=2 checksum=3418
root@dev #
root@dev # cat my_profile
install_type initial_install
system_type standalone
cluster SUNWCreq
pool rpool auto auto auto mirror c1t0d0s0 c1t1d0s0
bootenv installbe bename s10x_u8wos_08a dataset /var
root@dev #
Thank you,
Jacob

Okay, I figured out why I'm getting the File </boot/grub/menu.lst> propagation successful,etc. messages. Its because I had
bootenv installbe bename s10x_u8wos_08a dataset /var
in my jumpstart profile. When that line exists, another boot env is installed, also Live Upgrade is installed when that line exists. Removing that line Live Upgrade is not installed and an alternate boot environment is not installed. But now the problem is, I would like to have a separate dataset for /var and unfortunately it is not possible to create it without the bootenv line.
Does anyone know how I could have the "best of both worlds", ie have separate dataset for /var and not having to install a new boot env.
Thank you,
Jacob.

Client Push on Windows 7 x86 with IE9 open.

I everyone, I need to validate something.
I am currently deploying SCCM 2012 R2 for a client of mine.
They are using mainly Windows 7 with SP1 x86 with IE9 installed.
When doing client push from the console or installing any SCCM agent update, we are getting an error message stating the IE just crashed and the error is talking about IEFrame.dl:
Doing the same thing on a Windows 7 with SP1 x64 is not getting an error message.
Anyone can reproduce the error on is side? Or anyone know a workaround for this issue?
We are talking about SCCM 2012 R2 CU3 doing a client push on a Windows 7 with SP1.
There is no update done on Windows (I was using the default install.wim at first then applied all updates) and IE9 is installed (installation file downloaded at
http://techdows.com/2011/03/download-ie9-final-offlinestandalone-installer.html).
I will try the same thing with a all the updates installed on Windows 7 x86.
Thank you for your help.
Mathieu

After some troubleshooting and a bit of luck, the problem is the following:
The December update KB3008923 (Cumulative Update for IE December 2014) was causing the problem on IE9.
The update released on January 2015 KB3025945 is fixing the issue.
This is not an update available through WSUS so you have to look for it at
http://catalog.update.microsoft.com.
There is a bit more information, the problem is recurring but I did not found which task was causing the error so it was not only the installation that was causing the problem. After installing the KB3008923, the problem was present on any x86 Windows 7
computer running any version of configMgr agent (tested on 2012 SP1 and 2012 R2).
After installing the January update, the problem is gone.
Mathieu

Sun E450 serial console problem

Hello! I have an old Sun E450 server and can't make console work. I tried both ttya and ttyb (I connect appropriate pins of Serial-to-USB cable and choosing output in Stop + A OpenBoot shell), tried different port speeds, tried resetting NVRAM by Stop+N etc. Nothing helps. Unfortunately, I can't connect VGA monitor because don't have an adapter and can't find it there in Russia, Moscow (an original Sun monitor don't know where to get too). Terminal emulation programs (minicom, screen) outputs garbage instead of text. I.e., boot process — garbage, I press Enter — garbage, I do command to net boot — rarpd and tftpd works OK, I see logs on Linux server, but I see garbage in serial console. How to fix, please advise.

Hi.
1. Disconnect Keyboard.
    By default, this server use keyboard and screen for console.
    In case keyboard not connected, ttya used as console.
2. Use Null-Modem Cable. ( Cross RX-TX singnal)
3. Use next settings for connect to console: 9600,1 - stop bit, no parity, no flow control.
Regards.
PS.
   It's not problem find   VGA adapter or another SPARC server at Moscow. Just try call to some Oracle's partners.

USB serial console

I need to run a serial console on my laptop in order to get information on a kernel panic. Since the laptop does not have a serial connector, I would like to use a serial console over USB. I have enabled the required options in the kernel (CONFIG_USB_SERIAL=y and CONFIG_USB_SERIAL_CONSOLE=y), but I have a problem due to Arch's device naming. I believe that the kernel expects a device name of the form ttyUSB0, which should be located under /dev. However, with Arch (because of udev?) this device is called USB0 and is located under /dev/tts.
How can I pass this information to the kernel? I tried console=tts/USB0, but then it just hangs during boot.
Elad

After your system has booted, the device name you see under /dev is indeed due to udev - here's the relevant rule:
KERNEL=="ttyUSB[0-9]*", NAME="tts/USB%n"
However, at boot time, udev hasn't run yet, so the kernel name ttyUSBx should work.

Cisco WS-C2960S-24PS-L Nothing ever appears on the serial console

We have Two Cisco 2960s switches. All two of them are experiencing the same problem
Here's the problem:
When I plug in the power, all the ports lights flash once and the SYST, RPS, MSTR,STAT, DUPLX , SPEED, STCK and POE lights come on for a second. Then all the lights turn off, except for the SYST light which stays green. Nothing ever appears on the serial console. Holding the Mode button while applying power does nothing,
kindly help me with this...?

configure default settings on com port and you will see comand prompt on the serial consol

How to configure your PC as a SUN Blade 1000 serial Console

Does anybody have the experience to configure your PC as a Sun Serial Console? What pinout will need for a DB25M/DB9F serial cable? What terminal software you recommend?
Thanks!

It appears that the EDID information is not passing back from that Samsung LCD monitor.
That could be due to a missing wire in the dongle you purchased.
It may mean you need to battle through this in a more manual fashion.
That would explain the Creator card listing Sun's historic default 1152-mode
as the only viable screen resolution for the card (no asterisk on that line).
The man pages for the FFBCONFIG utility,
http://bama.ua.edu/cgi-bin/man-cgi?ffbconfig+1M
suggest you have the -prconf as well as the -propt command switches.
What results do you see with -prconf ?
Compare your results with the example in the man pages.
If no information for the monitor, then that might confirm the lack of EDID signalling.
You could also try using the svga setting instead of the numerical 1024x768x60 resolution.
Also consider using a CRT monitor to do all this investigation,
and only fighting with the LCD afterwards.
... and the output from PRINTENV is ??
We need to confirm that the system is indeed set to display to the screen.

Mac mini not booting up with 8GB memory after 10.7.2 update

My mac mini is not booting up with 8GB of memory after the 10.7.2 update. It just keeps on beeping and flashing the led.
If I take out the upper DIMM (Bank 1) It boots just fine, and it does so with either one of the 4GB DIMMs in Bank 0, so the DIMMs are fine. It also boots up fine with two 1GB DIMMs installed simultaneously, so the Bank 1 seems to be somehow working as well.
This trouble started right after the Lion recovery update, meaning the mini has never booted with the full 8GB installed after the update.
It worked just fine before that, with the full 8GB of memory installed.
Anyone else having similar problems?
Cheers, Otto.

I have upgraded my 2011 Macbook Pro and my 2011 Mini Server
to 8 gig with Corsair RAM and have had no issues
(CMSA8GX3M2A1333C9). What brand RAM was used?
There have been issues in the past with some RAM
brands.
Something to try is a PRAM reset with the 8 gig installed.
http://docs.info.apple.com/article.html?path=Mac/10.7/en/mh26871.html
Also, make sure the Minis are booting into 64 bit mode.
Open Applications->Utilities->System Information
On the left side click the software heading. There should be a line
64-bit Kernel and Extensions: Yes
This should be the default for 2010 and 2011, but something could
have got hosed.

Serial Console for VM Manager not working

I can't seem to connect to an SPARC VM created via the Oracle VM Manager. The serial console opens and appears to connect to a port on my local desktop machine from which I'm running my browser but fails to connect. I'm assuming VM Manager serial console uses a tunneling technology? Is there anything anyone can think of that would help me debug this session? I've turned off the firewall on my desktop and turned off iptables
My VM's don't have IP's yet, so I can't directly ssh to them. There's a chance my VM's are bad, but I really don't know how to troubleshoot since I can't get a console. (I created these VM's from the Oracle provided templates.)
Does anyone have any troubleshooting tips?

It sounds like you've installed the JTA package on the system running the manager - that's a necessary first step. The other things that can affect this include enabling popups (sounds like you have that correct too) and the combination of browser, JVM on client, and OS. I've had the best results with JVM 6 or with the very latest JVM 7 (build 21). Hope this helps, Jeff

Booting x86 with serial console by default

Similar Messages

Maybe you are looking for