NetBSD on Sun's 64-bit Ultrasparc based machines (NetBSD/sparc64)

Chris Ross | 27 Feb 2012 02:58

Netbooting a sparc64 (netbsd-6)


  I know I've netbooted sparc64's from my i386 build machine in the past, but it's been a couple years, and the
i386 build machine has been replaced at least once.  I've been following the documentation at
www.netbsd.org, and have it *mostly* working.  I'm also not sure if something has "changed" with netbsd-6.

  Right now, I've got a dhcpd running on my primary router that points the sparc64 box to the i386 NFS server,
and gives it an address.  I've got a rarpd and tftpd running on the i386 box serving up the new netbsd-6 ofwboot.net.

  I even have it loading a kernel at this point.  But, after the kernel loads via nfs root, I only see:

root on bge0
nfs_boot: trying DHCP/BOOTP
nfs_boot: timeout...
nfs_boot: timeout...
nfs_boot: timeout...
nfs_boot: trying RARP (and RPC/bootparam)
revarp failed, error=51
Supported file systems: union umap tmpfs smbfs puffs ptyfs procfs overlay null nfs msdos mfs lfs kernfs ffs
fdesc cd9660
no file system for bge0
cannot mount root, error = 79
root device (default bge0):

  I've tried a few different things.  I've mucked with the rc.conf and ifconfig.bge0 on the NFS root of the
host, but I don't think it's even getting far enough to be reading those, so I don't think it matters.

  Is this an NFS issue, where the kernel is being loaded via NFS somehow, but then NFS is not working to mount the filesystem?

  If anyone has any ideas, let me know.  If you want all of my various config files, let me know off-list and I can
send any of them to you for review.

(Continue reading)

headers

Erik Fair | 27 Feb 2012 04:33

Re: Netbooting a sparc64 (netbsd-6)

There are two ways to deal with this for Suns:

the old way: RARP + bootparamd

the new way: DHCP (looks like your sparc64 is trying new way first)

The variants of firmware I've netbooted over the years have taught me that firmware is kinda particular
about what it gets back from the network as responses to its queries.

I'll describe new way, with DHCP:

key dhcp parameters:

	option next-server 1.2.3.4 # or a hostname; this is where secondary boot will be TFTP'd from
	server-name "a-hostname-string-here-for-nfs-mounts-usually-short";

	host net-boot-client-host {
		hardware ethernet xx:xx:xx:xx:xx:xx;	# the client's MAC
		fixed-address 4.5.6.7;	# or a hostname
		option root-path "/an/nfs/exported/root/for/this/host"
	}

Your netboot server [1.2.3.4] will have to have the NetBSD/sparc64 netboot secondary booter available
for TFTP (configure TFTP in /etc/inetd.conf) and it will have to be named by the client's assigned IP
address ... in capital hexadecimal. Once your netboot client has its IP address, you'll see the TFTP
requests in your server's log.

I like to set up TFTPd so that it gets every unadorned filename from /tftpboot, e.g.

	tftp            dgram   udp     wait    root    /usr/libexec/tftpd      tftpd -l -s /tftpboot

(Continue reading)

headers

Martin Husemann | 27 Feb 2012 10:08

Re: Netbooting a sparc64 (netbsd-6)

On Sun, Feb 26, 2012 at 08:58:16PM -0500, Chris Ross wrote:
> root on bge0

I think we have seen firmware versions that break the bge interface they
use to download ofwboot/kernel so it would not work properly afterwards.
Do you have the latest firmware version for your machine installed?

Check on the dhcp server (or nfs server) with tcpdump if it gets any packets
after the kernel has taken over...

Martin

headers

Chris Ross | 27 Feb 2012 15:17

Re: Netbooting a sparc64 (netbsd-6)

On 02/27/2012 04:08 AM, Martin Husemann wrote:
> On Sun, Feb 26, 2012 at 08:58:16PM -0500, Chris Ross wrote:
>> root on bge0
> I think we have seen firmware versions that break the bge interface they
> use to download ofwboot/kernel so it would not work properly afterwards.
> Do you have the latest firmware version for your machine installed?
>
> Check on the dhcp server (or nfs server) with tcpdump if it gets any packets
> after the kernel has taken over...
>

   Thanks.  I'll check that first, as I believe it is not the most 
up-to-date OBP on the machine.  And, I'm pretty sure the DHCP server 
didn't see any packets, at least not enough of a packet for it to log 
seeing a request, so.

   Thanks for the pointer.

                                    - Chris

headers

Chris Ross | 27 Feb 2012 15:43

Re: Netbooting a sparc64 (netbsd-6)

On 02/27/2012 04:08 AM, Martin Husemann wrote:
> I think we have seen firmware versions that break the bge interface 
> they use to download ofwboot/kernel so it would not work properly 
> afterwards. Do you have the latest firmware version for your machine 
> installed? Check on the dhcp server (or nfs server) with tcpdump if it 
> gets any packets after the kernel has taken over... Martin 

   Do you, or anyone else, remember which firmwares have seen this, or 
other, problems?  I have checked, and I'm running 4.22.33, which looks 
to be the most recent in the 4.22.xx line (patch 121683-06).  There are 
also two patches in the 142700 line, which provide OBP 4.30.xx.

   I'm trying to get a line on 142700-02 at the moment, but just wanted 
to see if anyone knew of versions that did or didn't work.  Thanks.

                                - Chris

headers

Martin Husemann | 27 Feb 2012 15:46

Re: Netbooting a sparc64 (netbsd-6)

On Mon, Feb 27, 2012 at 09:43:25AM -0500, Chris Ross wrote:
>   Do you, or anyone else, remember which firmwares have seen this, or 
> other, problems?  I have checked, and I'm running 4.22.33, which looks 
> to be the most recent in the 4.22.xx line (patch 121683-06).  There are 
> also two patches in the 142700 line, which provide OBP 4.30.xx.

121683-06 is the one I have (which works for me).

Martin

headers

Chris Ross | 27 Feb 2012 15:48

Re: Netbooting a sparc64 (netbsd-6)

On 02/27/2012 09:46 AM, Martin Husemann wrote:
> On Mon, Feb 27, 2012 at 09:43:25AM -0500, Chris Ross wrote:
>>    Do you, or anyone else, remember which firmwares have seen this, or
>> other, problems?  I have checked, and I'm running 4.22.33, which looks
>> to be the most recent in the 4.22.xx line (patch 121683-06).  There are
>> also two patches in the 142700 line, which provide OBP 4.30.xx.
> 121683-06 is the one I have (which works for me).

   Hmm.  Okay.  Then I guess while I'll still try to update to 4.30.xx, 
that may not be the issue.  I'll start watching the network traffic to 
the DHCP server and the tftp/nfs server this evening or next time I look 
at it.

   Thanks again.

                                - Chris

headers

Martin Husemann | 27 Feb 2012 16:04

Re: Netbooting a sparc64 (netbsd-6)

On Mon, Feb 27, 2012 at 09:48:54AM -0500, Chris Ross wrote:
>   Hmm.  Okay.  Then I guess while I'll still try to update to 4.30.xx, 
> that may not be the issue.  I'll start watching the network traffic to 
> the DHCP server and the tftp/nfs server this evening or next time I look 
> at it.

There is an easy way to verify the problem I had in mind: connect another
port as well and use that to netboot - e.g. have the prom use bge1 to load
the kernel, then the kernel will be able to use bge0.

Martin

headers

Chris Ross | 28 Feb 2012 03:15

Re: Netbooting a sparc64 (netbsd-6)

On Feb 27, 2012, at 10:04 , Martin Husemann wrote:
> On Mon, Feb 27, 2012 at 09:48:54AM -0500, Chris Ross wrote:
>>  Hmm.  Okay.  Then I guess while I'll still try to update to 4.30.xx, 
>> that may not be the issue.  I'll start watching the network traffic to 
>> the DHCP server and the tftp/nfs server this evening or next time I look 
>> at it.
> 
> There is an easy way to verify the problem I had in mind: connect another
> port as well and use that to netboot - e.g. have the prom use bge1 to load
> the kernel, then the kernel will be able to use bge0.

  The first attempt to do this, with a manual "move the cable between ports, and use the same IP address for
both" didn't work.  If I get too much further, I'll make a new cable (it's a long run, this server isn't in the
service area/room/closet yet) and try it the hard way.  Would you suggest using two different IP
addresses, given that there are two MACs ?

        - Chris

headers

Chris Ross | 28 Feb 2012 03:39

Re: Netbooting a sparc64 (netbsd-6)

On Feb 27, 2012, at 21:15 , Chris Ross wrote:
>> There is an easy way to verify the problem I had in mind: connect another
>> port as well and use that to netboot - e.g. have the prom use bge1 to load
>> the kernel, then the kernel will be able to use bge0.
> 
>  The first attempt to do this, with a manual "move the cable between ports, and use the same IP address for
both" didn't work.  If I get too much further, I'll make a new cable (it's a long run, this server isn't in the
service area/room/closet yet) and try it the hard way.  Would you suggest using two different IP
addresses, given that there are two MACs ?

  In case this helps, continuing the "switch the cable and wait, after the kernel fails to mount the nfs filesystem".

  I loaded the bootloader via tftp, and the kernel via NFS, over bge1.  Then, after the kernel loads and fails to
configure bge1, I switched the cable to bge0.  I waited a while, checked that the arp cache on the router
didn't have anything, and cleared it on the NFS (/rarpd/tftpd) server.  I told the sparc to use bge0, and it
claims to be trying to get an address via DHCP/BOOTP, but there is never any traffic from the mac(s) on that
machine [seen by the router/DHCP server].

  So, it looks more like this version of the kernel (netbsd-6, circa 25-Feb) just isn't passing packets at
all.  So, with the kernel in question, I loaded the 5.99.59 userland I had on a disk.  I then tried to dhcpcd
from it, and also saw nothing on the dhcp server.

  So.  I then cold-booted into the 5.99.59 image on the disk, which was never really fully configured.  It's in
full Amnesiac mode.  If I run dhcpcd on bge0, it gives up assigning a 169.254/16 address, my dhcpd never
having seen a packet.  But, if I kill dhcpcd and manually configure an address, it seems to pass packets, on
net and through the router.

  This is smelling more like a sparc64 bge bug as time goes on...

(Continue reading)

headers

Martin Husemann | 28 Feb 2012 09:52

Re: Netbooting a sparc64 (netbsd-6)

On Mon, Feb 27, 2012 at 09:39:58PM -0500, Chris Ross wrote:
>   So, it looks more like this version of the kernel (netbsd-6, circa 25-Feb) just isn't passing packets at
all.  So, with the kernel in question, I loaded the 5.99.59 userland I had on a disk.  I then tried to dhcpcd
from it, and also saw nothing on the dhcp server.

Well, the regular test runs of netbsd-6 on sparc64 are on a v210 and run
via ssh - so I am pretty sure a feb 24 kernel worked fine. I'll do a new
test run today.

>   This is smelling more like a sparc64 bge bug as time goes on...

Or a bge driver bug (more likely).

Martin

headers

Martin Husemann | 28 Feb 2012 11:32

Re: Netbooting a sparc64 (netbsd-6)

On Tue, Feb 28, 2012 at 09:52:15AM +0100, Martin Husemann wrote:
> On Mon, Feb 27, 2012 at 09:39:58PM -0500, Chris Ross wrote:
> >   So, it looks more like this version of the kernel (netbsd-6, circa 25-Feb) just isn't passing packets at
all.  So, with the kernel in question, I loaded the 5.99.59 userland I had on a disk.  I then tried to dhcpcd
from it, and also saw nothing on the dhcp server.
> 
> Well, the regular test runs of netbsd-6 on sparc64 are on a v210 and run
> via ssh - so I am pretty sure a feb 24 kernel worked fine. I'll do a new
> test run today.

Just finished with no unexpected failures:

	http://www.netbsd.org:~martin/sparc64-atf-netbsd6

Could you try burning the iso and boot from that, then when in sysinst
go to the utility menu, do the network setup via dhcp and see if that
works?

Martin

headers

Eric Schnoebelen | 28 Feb 2012 09:04

Re: Netbooting a sparc64 (netbsd-6)


Which switches are you using? And which model sparc64?

I have had no end of problems netbooting when connected to some
models of Cisco Catalyst switch.  The dominate model in use here
is a 2924, and it seems to take *forever* to do it's
autonegociation.  So long that first ofwboot, and then the
kernel frequently timeout.

I seem to remember adding a loop to an older bit of the netboot
portion of ofwboot to try to broadcast the dhcp packet multiple
times before giving up.. (and trying to avoid resetting the port
in between.)

FYI: you don't need to hardcode the loader name as a hex
representation of the client IP address when using DHCP.  Here
is the configuration block I've used successfully:

    # get any vendor-class-identifier information
    set vendor-class-id = option vendor-class-identifier;

    # standard netboot for the SunFire V120's 
    # (so we don't need to create and entry for each one of them!)
    class "SunFire-V120" {
	match if substring(vendor-class-id,1,13) = "SUNW.UltraAX-";
	spawn with hardware;
	# next-server 192.67.63.1;
	server-name "192.67.63.1";
	filename "/netboot.netbsd.sparc64";
	option root-path "192.67.63.1:/boot.d/netboot/install/sparc64";

(Continue reading)

headers

Dave McGuire | 28 Feb 2012 09:13

Re: Netbooting a sparc64 (netbsd-6)

On 02/28/2012 03:04 AM, Eric Schnoebelen wrote:
> Which switches are you using? And which model sparc64?
>
> I have had no end of problems netbooting when connected to some
> models of Cisco Catalyst switch.  The dominate model in use here
> is a 2924, and it seems to take *forever* to do it's
> autonegociation.  So long that first ofwboot, and then the
> kernel frequently timeout.

   Pardon me for butting in, but this rang a bell so I went looking 
through my old notes.  This was in there from way, way back:

"Cisco's policy to enable arbitration for all protocols.  The first
protocols that switches arbitrate for are trunk and channeling, and
since the Sun NICs don't, it cause the Sun NICs to mistake the
trunking and channeling arbitration for 100Mbps, half duplex."

                -Dave

--

-- 
Dave McGuire, AK4HZ
New Kensington, PA

headers

Mouse | 28 Feb 2012 15:38

Re: Netbooting a sparc64 (netbsd-6)

> I have had no end of problems netbooting when connected to some
> models of Cisco Catalyst switch.  The [dominant] model in use here is
> a 2924, and it seems to take *forever* to do [its] autonegociation.
> So long that first ofwboot, and then the kernel frequently timeout.

I've had related issues with a SS20 connected to a 2924 (specifically,
a WS-C2924-XL-EN) - unless I specifically tell it to use twisted pair,
it will time out and switch to AUI when the ROMs are trying to netboot.

In my case, since the SS20's interface is 10-only, I put a small 10Mb
hub between the SS20 and the Cisco, which "fixed it" - ie, made the
symptom go away, though of course it didn't actually repair the
underlying problem.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse <at> rodents-montreal.org
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B

headers

Martin Mersberger | 28 Feb 2012 15:51

Re: Netbooting a sparc64 (netbsd-6)

Hi...

>> I have had no end of problems netbooting when connected to some
>> models of Cisco Catalyst switch.  The [dominant] model in use here is
>> a 2924, and it seems to take *forever* to do [its] autonegociation.
>> So long that first ofwboot, and then the kernel frequently timeout.
> 
> I've had related issues with a SS20 connected to a 2924 (specifically,
> a WS-C2924-XL-EN) - unless I specifically tell it to use twisted pair,
> it will time out and switch to AUI when the ROMs are trying to netboot.
> 
> In my case, since the SS20's interface is 10-only, I put a small 10Mb
> hub between the SS20 and the Cisco, which "fixed it" - ie, made the
> symptom go away, though of course it didn't actually repair the
> underlying problem.

I had similar problems on catalyst as long as 'spanning-tree portfast'
was not configured on the port.
(but it should be configured by default on access ports to avoid blocked
switchport issues on the client side)

regards
 Martin

headers

Chris Ross | 28 Feb 2012 15:50

Re: Netbooting a sparc64 (netbsd-6)

On 02/28/2012 03:04 AM, Eric Schnoebelen wrote:
> Which switches are you using? And which model sparc64?

   I'm using a Dell 2824.  And the sparc64 is a SunFire v240.  
Negotiation takes a bit, but isn't the core of the problem.

   And yes, I too remember fighting with Cisco 2900's (and 3500's) about 
that issue.  We used to hard-code the speed and duplex on them to avoid, 
or at least assist, that issue.

                        - Chris

headers

Chris Ross | 14 Mar 2012 04:21

Re: Netbooting a sparc64 (netbsd-6)

On Feb 27, 2012, at 10:04 , Martin Husemann wrote:
> On Mon, Feb 27, 2012 at 09:48:54AM -0500, Chris Ross wrote:
>>  Hmm.  Okay.  Then I guess while I'll still try to update to 4.30.xx, 
>> that may not be the issue.  I'll start watching the network traffic to 
>> the DHCP server and the tftp/nfs server this evening or next time I look 
>> at it.
> 
> There is an easy way to verify the problem I had in mind: connect another
> port as well and use that to netboot - e.g. have the prom use bge1 to load
> the kernel, then the kernel will be able to use bge0.

  Alright.  I've finally gotten around to stringing another cable, and it doesn't
work as hoped.  When I boot from "net1", aka bge1, the kernel loads, but then
when I try to tell the kernel that the root device should be bge0 it tries to DHCP,
but my dhcp server never sees it, and it fails to continue.

  Does someone else have netbsd-6 running on bge devices successfully?  I
could certainly try a CD install if I thought it would yield different results, although
that itself would concern me.

         - Chris

headers

Martin Husemann | 14 Mar 2012 09:58

Re: Netbooting a sparc64 (netbsd-6)

On Tue, Mar 13, 2012 at 11:21:32PM -0400, Chris Ross wrote:
>   Does someone else have netbsd-6 running on bge devices successfully?

Yes, the bge0..3 in my v210 work just fine with NetBSD 6_BETA.

Please file a PR, this very likely is a bge driver bug.

Martin

headers

Michael | 14 Mar 2012 17:34

Re: Netbooting a sparc64 (netbsd-6)

Hello,

On Wed, 14 Mar 2012 09:58:54 +0100
Martin Husemann <martin <at> duskware.de> wrote:

> On Tue, Mar 13, 2012 at 11:21:32PM -0400, Chris Ross wrote:
> >   Does someone else have netbsd-6 running on bge devices successfully?
> 
> Yes, the bge0..3 in my v210 work just fine with NetBSD 6_BETA.
> 
> Please file a PR, this very likely is a bge driver bug.

There is definitely something fishy.
When I tried to netboot my Blade 2500 things would work fine until the kernel took over - then the bge stopped
sending or transmitting packets. A workaround was to add another ethernet card ( some tlp clone in my case
), have OF boot from the bge and force the kernel to use the tlp. When not netbooting the bge works just fine in
NetBSD, so OF does something to it that confuses the hell out of the bge driver.
This is what's in the Blade:
bge0 at pci0 dev 3 function 0: Broadcom BCM5703 Gigabit Ethernet
bge0: interrupting at ivec 371c
bge0: ASIC BCM5702/5703 A2 (0x1002), Ethernet address xx:xx:xx:xx:xx:xx
brgphy0 at bge0 phy 1: BCM5703 1000BASE-T media interface, rev. 2

That said, I have two other bges where the driver has trouble without any firmware meddling:
bge0 at pci2 dev 2 function 0: SysKonnect SK-9Dx1 Gigabit Ethernet
bge0: interrupting at irq 52
bge0: ASIC BCM5700 Altima (0x7104), Ethernet address xx:xx:xx:xx:xx:xx
brgphy0 at bge0 phy 1: BCM5411 1000BASE-T media interface, rev. 1

have fun

(Continue reading)

headers

Chris Ross | 15 Mar 2012 03:21

Re: Netbooting a sparc64 (netbsd-6)


  pr filed.  port-sparc64/46197

  I tried to fill in other information from this thread, but anyone who can contribute info, please do.

            - Chris

On Mar 14, 2012, at 04:58 , Martin Husemann wrote:

> On Tue, Mar 13, 2012 at 11:21:32PM -0400, Chris Ross wrote:
>>  Does someone else have netbsd-6 running on bge devices successfully?
> 
> Yes, the bge0..3 in my v210 work just fine with NetBSD 6_BETA.
> 
> Please file a PR, this very likely is a bge driver bug.
> 
> Martin

headers

Brian Buhrow | 28 Feb 2012 17:42

Re: Netbooting a sparc64 (netbsd-6)

	Hello.  On cisco 2924 switches, you can give the command:
spanning-tree portfast default
in the main configuration mode to turn off the 15 second auto-negotiation
process.  This should cause things to come up much faster.  The trade off
is that if you inadvertently introduce a loop into your switch topology,
you may experience a brief outage while the spanning-tree protocol
calculates where the loop is and disables the port.  The 15 second delay at
link up time prevents this outage possibility at the expense of slow
initial connect times.
-Brian
On Feb 28,  2:04am, Eric Schnoebelen wrote:
} Subject: Re: Netbooting a sparc64 (netbsd-6)
} 
} Which switches are you using? And which model sparc64?
} 
} I have had no end of problems netbooting when connected to some
} models of Cisco Catalyst switch.  The dominate model in use here
} is a 2924, and it seems to take *forever* to do it's
} autonegociation.  So long that first ofwboot, and then the
} kernel frequently timeout.
} 
} I seem to remember adding a loop to an older bit of the netboot
} portion of ofwboot to try to broadcast the dhcp packet multiple
} times before giving up.. (and trying to avoid resetting the port
} in between.)
} 
} FYI: you don't need to hardcode the loader name as a hex
} representation of the client IP address when using DHCP.  Here
} is the configuration block I've used successfully:
}

(Continue reading)

headers

Mouse | 29 Feb 2012 16:23

Re: Netbooting a sparc64 (netbsd-6)

> On cisco 2924 switches, you can give the command:
> spanning-tree portfast default
> in the main configuration mode to turn off the 15 second
> auto-negotiation process.

In my case, I have "no spanning-tree vlan 1", "no spanning-tree vlan
2", "no spanning-tree vlan 3", etc.  I also have "spanning-tree
portfast" set on each interface.

It still doesn't work - I just tried it.  Copied from the console:

ok boot net -s
Boot device: /iommu/sbus/ledma <at> f,400010/le <at> f,c00000  File and args: -s
Lost Carrier  (transceiver cable problem?)  
Cable problem or twisted pair hub link-test disabled.
Use the PROM command "help ethernet" for more information.

Timeout waiting for ARP/RARP packet

The switch reports

Feb 29 15:05:43: %LINK-3-UPDOWN: Interface FastEthernet0/23, changed state to up
Feb 29 15:05:44: %LINEPROTO-5-UPDOWN: Line protocol on Interface FastEthernet0/23, changed state to up

but apparently it didn't change state fast enough.

"help ethernet" output says, in part,

 For a TPE network, the 10BaseT Link-Integrity-Test function may be con-
 figured to be enabled or disabled.  For the TPE network port to work

(Continue reading)

headers

matthew green | 1 Mar 2012 01:03

re: Netbooting a sparc64 (netbsd-6)


> ok boot net -s
> Boot device: /iommu/sbus/ledma <at> f,400010/le <at> f,c00000  File and args: -s

so.. this is a sparc?  have you tried "boot net-tpe" for this one?

.mrg.

headers

Mouse | 1 Mar 2012 01:57

Re: Netbooting a sparc64 (netbsd-6)

>> Boot device: /iommu/sbus/ledma <at> f,400010/le <at> f,c00000  File and args: -s
> so.. this is a sparc?

Yes, a SS20.

> have you tried "boot net-tpe" for this one?

Yes.  It works (except when I have tpe-link-test? set false; I've yet
to make it work at all with tpe-link-test? set false).

The point is mostly that the 2924 is known to fail to get along with
one Sun device, so it would hardly be a shocker if it failed to get
along with another - and does not necessarily point to driver bugs,
which was a possibility I saw mentioned upthread.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse <at> rodents-montreal.org
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B

headers

Chris Ross | 1 Mar 2012 02:38

Re: Netbooting a sparc64 (netbsd-6)

On Feb 29, 2012, at 10:23 , Mouse wrote:
> I don't know how much of this is relevant to the original sparc64
> problem.  But it does seem to me to indicate that issues talking with a
> 2924 may well not indicate driver bugs; my experience is with the
> SS20's on-board Ethernet, which is not bge; it's Lance:

  In my case, I'm not talking to a cisco switch, but to a Dell 2824, and I've successfully netbooted other
sparcs (including the e420R that is now my primary sparc64 server) a year or more ago with this same switch. 
It still smells to me like a bge driver problem.  (My issue, that is)

          - Chris

headers

Mouse | 1 Mar 2012 02:49

Re: Netbooting a sparc64 (netbsd-6)

>   In my case, I'm not talking to a cisco switch, but to a Dell 2824, and I'v$

Please don't use paragraph-length lines.

> In my case, I'm not talking to a cisco switch, but to a Dell 2824,
> and I've successfully netbooted other sparcs [...] a year or more ago
> with this same switch.  It still smells to me like a bge driver
> problem.  (My issue, that is)

Ah, quite possibly, then.  I'd lost track of the differences between
what the thread I was writing in was about and your original issue.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse <at> rodents-montreal.org
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B

headers

Greg Earle | 1 Mar 2012 03:02

Re: Netbooting a sparc64 (netbsd-6)

On Feb 29, 2012, at 5:49 PM, Mouse wrote:

>>  In my case, I'm not talking to a cisco switch, but to a Dell 2824, and I'v$
> 
> Please don't use paragraph-length lines.

Hey Michael Parker, please wake up, the '90s are over.  We don't use ADM-3As anymore.

headers

Dave McGuire | 1 Mar 2012 05:01

Re: Netbooting a sparc64 (netbsd-6)

On 02/29/2012 09:02 PM, Greg Earle wrote:
>>>   In my case, I'm not talking to a cisco switch, but to a Dell 2824, and I'v$
>>
>> Please don't use paragraph-length lines.
>
> Hey Michael Parker, please wake up, the '90s are over.  We don't use ADM-3As anymore.

   Speak for yourself! ;)

     [Dave pats his VT-100 connected to PDP-11 next to desk, glances at 
DECsystem-20 in next room]

           -Dave

--

-- 
Dave McGuire, AK4HZ
New Kensington, PA

headers

David Laight | 1 Mar 2012 22:57

Re: Netbooting a sparc64 (netbsd-6)

On Wed, Feb 29, 2012 at 11:01:42PM -0500, Dave McGuire wrote:
> 
>     [Dave pats his VT-100 connected to PDP-11 next to desk...

You should have a VT50 or VT52, not a VT100 
(Or even the VT55 - I think that is the one with the wet-paper
screen print.)

	David

--

-- 
David Laight: david <at> l8s.co.uk

headers

Dave McGuire | 2 Mar 2012 00:20

Re: Netbooting a sparc64 (netbsd-6)

On 03/01/2012 04:57 PM, David Laight wrote:
> On Wed, Feb 29, 2012 at 11:01:42PM -0500, Dave McGuire wrote:
>>
>>      [Dave pats his VT-100 connected to PDP-11 next to desk...
>
> You should have a VT50 or VT52, not a VT100 
> (Or even the VT55 - I think that is the one with the wet-paper
> screen print.)

   I have both VT50 and VT52s here as well, but those are connected to 
my PDP-8s.

            -Dave

--

-- 
Dave McGuire, AK4HZ
New Kensington, PA

headers

Chris Ross | 1 Mar 2012 06:04

Re: Netbooting a sparc64 (netbsd-6)


On Feb 29, 2012, at 21:02 , Greg Earle wrote:

> On Feb 29, 2012, at 5:49 PM, Mouse wrote:
> 
>>> In my case, I'm not talking to a cisco switch, but to a Dell 2824, and I'v$
>> 
>> Please don't use paragraph-length lines.
> 
> Hey Michael Parker, please wake up, the '90s are over.  We don't use ADM-3As anymore.

  It was a polite request that I'll try to adhere to.  

  So, how to you tell Apple Mail to insert line-breaks?  

       - Chris

headers

Greg Earle | 1 Mar 2012 06:15

Re: Netbooting a sparc64 (netbsd-6)

On Feb 29, 2012, at 9:04 PM, Chris Ross wrote:

> On Feb 29, 2012, at 21:02 , Greg Earle wrote:
> 
>> On Feb 29, 2012, at 5:49 PM, Mouse wrote:
>> 
>>>> In my case, I'm not talking to a Cisco switch, but to a Dell 2824, and I'v$
>>> 
>>> Please don't use paragraph-length lines.
>> 
>> Hey Michael Parker, please wake up, the '90s are over.  We don't use ADM-3As anymore.
> 
>  It was a polite request that I'll try to adhere to.  

Why?

You should do whatever the f you want, not what one solitary clown who refuses to use a modern e-mail reader wants.

>  So, how to you tell Apple Mail to insert line-breaks?  

You can't, as I'm sure you know from the wink.

BTW 'Mouse' how's that Stamp Out HTML e-mails campaign coming along?  I bet the other two people in the world
who still use Pine are right in there with you ...

(OK, I've said my off-topic piece.  Good luck with the netbooting.  Back to lurker-land ... )

	- Greg

(Continue reading)

headers

Dave McGuire | 1 Mar 2012 06:21

Re: Netbooting a sparc64 (netbsd-6)

On 03/01/2012 12:15 AM, Greg Earle wrote:
> BTW 'Mouse' how's that Stamp Out HTML e-mails campaign coming along?
> I bet the other two people in the world who still use Pine are right
> in there with you ...

   Oh good heavens.  If you know THAT many people who insist that the 
Web and email are the same thing, you're hanging out in bad 
neighborhoods.  If you need different fonts and colors and graphical 
smileys in order to communicate, I respectfully submit that you have 
other, likely larger, problems that I'm in the wrong line of work to 
help you with.

   Out of an average of 691 email messages received to my account over 
the past week, precisely six of them were in HTML, and every one of them 
was spam that got through the filters.

   I run email services for an ISP; the spam filters there add a weight 
of 2 to HTML email.  The spam threshold is 5.

   Back to lurker-land, indeed.

           -Dave

--

-- 
Dave McGuire, AK4HZ
New Kensington, PA

headers

Mouse | 1 Mar 2012 13:59

Re: Netbooting a sparc64 (netbsd-6)

> You should do whatever the f you want, not what one solitary clown who refus$

You appear to have misspelled "mangles what the sender generated" as
"modern".  There is a perfectly good standard way for the sender to
indicate that it expects the recipient to reflow text; see RFC 3676.
People like you refuse, by your choice of user agent, to use it; I
don't see how it is a recipient's fault if the resulting email is
displayed as its markings direct.

Further than that I shall not feed this particular troll.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse <at> rodents-montreal.org
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B

headers

Greg Troxel | 1 Mar 2012 14:33

Re: Netbooting a sparc64 (netbsd-6)

Greg Earle <earle <at> isolar.DynDNS.ORG> writes:

> On Feb 29, 2012, at 9:04 PM, Chris Ross wrote:
>
>> On Feb 29, 2012, at 21:02 , Greg Earle wrote:
>> 
>>> On Feb 29, 2012, at 5:49 PM, Mouse wrote:
>>> 
>>>>> In my case, I'm not talking to a Cisco switch, but to a Dell 2824, and I'v$
>>>> 
>>>> Please don't use paragraph-length lines.
>>> 
>>> Hey Michael Parker, please wake up, the '90s are over.  We don't
>>> use ADM-3As anymore.
>> 
>>  It was a polite request that I'll try to adhere to.  
>
> Why?
>
> You should do whatever the f you want, not what one solitary clown who
> refuses to use a modern e-mail reader wants.

Ad hominem comments are inappropriate on this list.  The norms of mail
are pretty clear, and it's an unfortunate situation that Microsoft and
Apple mail clients are broken.  IIRC Apple's Mail.App used to correctly
implement format=flowed (so that the lines were a reasonable length, but
the receiving MUA had permission to rewrap them), but it no longer seems
to do that.

(Continue reading)

headers

Mouse | 1 Mar 2012 13:47

Re: Netbooting a sparc64 (netbsd-6)

> So, how to you tell Apple Mail to insert line-breaks?  

I don't know, since I don't use it.  Perhaps the RETURN key, or
whatever it's labeled on Mac keyboards?

If Apple Mail makes it difficult, I would say that's a good reason to
use something else instead.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse <at> rodents-montreal.org
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B