Serial redirection halts on Dell PowerEdge 1955 Blade (mfsBSD)

Wed Aug 10 15:21:16 UTC 2016

Fixed, thanks to this thread: http://mailing.freebsd.questions.narkive.com/1jubDzGu/r220-s-with-serial

Adding 
    hw.uart.console="io:0x2f8,br:57600"
to /boot/loader.conf seems to have fixed the issue.

-----Original Message-----
From: Lavery, Benjamin (IT Enterprise Services) 
Sent: 08 August 2016 11:48
To: 'freebsd-questions at freebsd.org'
Subject: FW: Serial redirection halts on Dell PowerEdge 1955 Blade (mfsBSD)

Good morning all,

(Apologies, cross-posting from freebsd-hardware)

I have been given an old Dell PowerEdge chassis filled with PowerEdge 1955 blades.  I also have a FreeNAS which I'm using to serve DHCP addresses and TFTP for PXE booting.

I can get the blades to PXE boot (I'm using pxelinux to load mfsbsd-10.3-RELEASE-amd64), but the serial redirection is only working for the first part of the boot sequence.  After select mfsBSD at the menu, I get the following (verbose boot chosen at the boot menu):

    Loading memdisk... ok
                                                             ok
    e820: 0000000000000000 00000000000a0000 1
    e820: 0000000000100000 00000000cfea8000 1
    e820: 00000000cffa8000 000000000000fc00 3
    e820: 00000000cffb7c00 0000000000048400 2
    Total size needed = 2513 bytes, allocating 3K
    Old dos memory at 0x8a000 (map says 0xa0000), loading at 0x89400
    1588: 0xffff  15E801: 0x3c00 0xcbba
    INT 13 08: Success, count = 1, BPT = 0000:0000
    Drive probing gives drive shift limit: 0x82
    old: int13 = ca001da6  int15 = f000f859  int1e = f000efc7
        new: int13 = 8940000a  int15 = 894003ba  int1e = f000efc7
    Loading boot sector... booting...
    /boot/config: -D

    FreeBSD/x86 boot
    Default: 0:ad(0p2)/boot/loader
    boot:
    BTX loader 1.00  BTX version is 1.02
    console comconsole failed to initialize
    Consoles: internal video/keyboard  serial port
    BIOS drive A: is disk0
    BIOS drive C: is disk1
    BIOS drive D: is disk2
    BIOS 549kB/3353216kB available memory

    FreeBSD/x86 bootstrap loader, Revision 1.1
     (root at releng1.nyi.freebsd.org, Fri Mar 25 02:04:38 UTC 2016)
    \oading /boot/defaults/loader.conf
    lapic3: Routing NMI -> LINT1
    lapic3: LINT1 trigger: edge
    lapic3: LINT1 polarity: high
    lapic7: Routing NMI -> LINT1
    lapic7: LINT1 trigger: edge
    lapic7: LINT1 polarity: high
    MADT: Interrupt override: source 0, irq 2
    ioapic0: Routing IRQ 0 -> intpin 2
    MADT: Interrupt override: source 9, irq 9
    ioapic0: intpin 9 trigger: level
    ioapic0 <Version 2.0> irqs 0-23 on motherboard
    ioapic1 <Version 2.0> irqs 32-55 on motherboard
    ioapic2 <Version 2.0> irqs 192-215 on motherboard
        cpu0 BSP:
         ID: 0x00000000   VER: 0x00050014 LDR: 0x00000000 DFR: 0xffffffff
      lint0: 0x00010700 lint1: 0x00000400 TPR: 0x00000000 SVR: 0x000001ff
      timer: 0x000100ef therm: 0x00010000 err: 0x000000f0 pmc: 0x00010400
    wlan: <802.11 Link Layer>
    snd_unit_init() u=0x00ff8000 [512] d=0x00007c00 [32] c=0x000003ff [1024]
    feeder_register: snd_unit=-1 snd_maxautovchans=16 latency=5 feeder_rate_min=1 fe
    eder_rate_max=2016000 feeder_rate_round=25
    null: <null device, zero device>
    nfslock: pseudo-device
    ▒▒▒▒▒▒▒▒▒▒▒

The odd glitch (▒) is me trying to input after output stops.  I'm accessing the console by TELNETing to the blade chassis, and connecting to the blade console.  I believe this runs at a baud rate of 57600.  Because of this I've modified the following files within mfsBSD:

etc/ttys (from inside mfsroot.gz)
    # Serial terminals
    ttyu0   "/usr/libexec/getty std.57600"  vt100   on secure
    ttyu1   "/usr/libexec/getty std.57600"  dialup  on secure
    ttyu2   "/usr/libexec/getty std.57600"  vt100   off secure
    ttyu3   "/usr/libexec/getty std.57600"  vt100   off secure
    # Dumb console
    dcons   "/usr/libexec/getty std.9600"   vt100   off secure

boot/loader.conf
    boot_multicons="YES"
    boot_serial="YES"
    comconsole_speed="57600"
    console="comconsole"

The Java console from the chassis web interface doesn't work on my laptop, but via another server I've been able to grab the attached screenshot.  It states:
    getty[706]: open /dev/ttyu0: No such file or directory

Unfortunately the Java console doesn't work well enough for me to use to log in and get at any settings (getting it to load that far took 10 minutes...)

The blades BIOS is set up to redirect to COM2 (it's either that or be disabled) and to emulate vt100/vt200 (could also be ANSI, but this doesn't seem to allow input).

Would anyone have any ideas about how I can get console redirection working past this point?  Or where I can look to get more information?

Many thanks,
Ben

This email is security checked and subject to the disclaimer on web-page: http://www.capita.co.uk/email-disclaimer.aspx