Re: Fwd: Fail to boot Fujitsu M10-4

[Kevin Stabel]

Hi,

When i get fast data mmu miss, it is usually related to device scanning and
such on scsi controllers.
Check that your nvramrc doesn't have things in it like scsi-probe-all and
such.

I've also had this problem once trying to load a kernel and initrd that was
too big.

HTH

-Kevin

On Tue, Jun 4, 2019 at 12:28 PM Anatoly Pugachev  wrote:

> On Tue, Jun 4, 2019 at 12:42 PM Sonnie Hook  wrote:
> >
> > I inserted a USB disk on M10-4 and partitioned it with grub bios in
> rescue mode. The installer automatically installed GRUB to /dev/sda. I
> chroot to /target and executed
> >
> > # grub-install --force-extra-removable "/dev/sdb"
> >
> > Installing for sparc64-ieee1275 platform.
> >
> > Installation finished. No error reported.
> >
> >
> > Then something different happened:
> >
> > {0} ok boot /pci@8000/pci@4/pci@0/pci@1/pci@0/usb@4,1/disk@0,0:a
> >
> > Boot device: /pci@8000/pci@4/pci@0/pci@1/pci@0/usb@4,1/disk@0,0:a  File
> and args:
> >
> > GRUB Loading kernel.
> >
> > vitual-device not found.
> >
> > ERROR: Last Trap: Fast Data Access MMU Miss
>
>
> JFYI
>
> Last time i've seen this error on 220R  (like 10 years ago) , the only
> way to boot linux on it was to hard power reset
>
>

Re: Updated installation images for Debian Ports 2019-05-09

[Kevin Stabel]

Hi,

I don't know for this machine.
On my V880, there was a second loop going to the disk backplane.  I
unplugged it, and everything started working.
I know this is pretty circumstantial, but don't know if i can do more for
you.

I did exchange some emails with the guy that wrote the driver to try and
fix it before, but he had no idea, and i'm guessing no time/interest in
fixing this either.

Kind Regards,

-Kevin

On Fri, May 10, 2019 at 7:00 AM Meelis Roos  wrote:

> > Hello Meelis,
> >
> > Perhaps of interest ... i was being hit with the qla2xxx driver panics
> as well on my v880.
> > I unplugged the second FCAL loop to the backplane, and this fixed it.
> >
> > Perhaps this is something you could try?
>
> I have not tried on my V880 but I got similar trouble on V480.
>
> But where is the secondary loop on SB1000/E280R (having the same
> mainboard)? Hmm,
> come to think of it, there is a connector marked with double arrow, the
> back of mainboard but I do
> not think I have seen any cable that would fit there.
>
> --
> Meelis Roos
>

Re: Updated installation images for Debian Ports 2019-05-09

[Kevin Stabel]

Hello Meelis,

Perhaps of interest ... i was being hit with the qla2xxx driver panics as
well on my v880.
I unplugged the second FCAL loop to the backplane, and this fixed it.

Perhaps this is something you could try?

On Thu, May 9, 2019 at 11:03 PM Meelis Roos  wrote:

> > Given that it works on a sun4u then I am guessing that a Sunblade 2000
> > should be perfect for this sort of thing. I know that I have one of
> > those in the warehouse. However I think they need FCAL disks. Anyways I
> > will give that a look see and also I hope serial console works out of
> > the box.
>
> Yes, I should try it on some sun4u too. And probably bare hardware sun4v -
> all my sparcs run Linux natively.
>
> About Blade 1x00/2x00 and E280R: they use qla2xxx for root disks and last
> I tried this was broken in the kernel. It may be better now but older
> Symbios
> and newer MPT etc controllers would probably be a safer try if you have
> them
> available.
>
> --
> Meelis Roos
>
>

Re: Latest Debian Sparc64 installer

[Kevin Stabel]

Hi,

I had a problem with the driver, but i disconnected the second loop to the
drive backplane, and it worked.
It's a long shot, but i guess you could try?

Kind Regards,

-Kevin

On Mon, Dec 17, 2018 at 2:10 PM Rick Leir  wrote:

> Thomas
> Does your machine have FiberChannel disks (FC-AL?)
>
> There was a problem with the driver the last time I tried to install. That
> was a year ago, and I was too busy to fix it. My guess is that there is an
> endian problem in whatever boot code that calls the driver. I say that
> because the driver is old and has never been changed.
>
> You might want to try using an ATA or SCSI disk temporarily. Please let me
> know how it goes.
> Cheers
> Rick
>
> On December 17, 2018 2:27:46 AM EST, Thomas D Dial  wrote:
>>
>> A few months back I made a half hearted try at installing Debian on a
>> Sun Blade 1000. The installer failed instantly and I deferred this in
>> favor of more immediately pressing matters, but now will have time to
>> spend on it.
>>
>> I have found debian-10.0-sparc64-NETINST-1.iso (apparent date 5/18/18
>> and debian-9.0-sparc64-NETINST-1.iso (apparent date 4/4/18) at
>> cdimage.debian.org. Is there a later version of either?
>>
>> Thanks in advance
>> Tom Dial
>>
>>
> --
> Sorry for being brief. Alternate email is rickleir at yahoo dot com
>

Re: qlogic fibrechannel

[Kevin Stabel]

Hi Rick,

Is it qla2200?

Kind Regards,

-Kevin

On Sep 7, 2017 2:18 PM, "Rick Leir"  wrote:

> Hi all
> I am still hoping to get my Sun 2000 (vintage 2002) running with Debian.
> The problem was the Qlogic fibrechannel disk driver, so perhaps I can add
> in a different disk controller temporarily. What economical solution would
> you recommend? Perhaps a USB disk?
>
> By the way, OpenBSD installs fine. I might be able to compare drivers and
> fix this bug, but the problem would be much easier to debug with Debian
> booting from a SCSI or SATA disk.
> Thanks
> Rick
>
> More Oracle analysis:
> https://meshedinsights.com/2017/09/03/oracle-finally-killed-sun/
>
> --
> Sorry for being brief. Alternate email is rickleir at yahoo dot com

Re: How can i boot *anything* on Sun Blade 100?

[Kevin Stabel]

Cool, just a bit more patience, TCH, we will get this machine running :)

Usually the part where the cable is the longest goes on the motherboard.
Then you need to set the jumpers correctly as well for slave/master.

These are the positions: http://support.seagate.com/kbimg/3050-2.jpg

Please try it and let us know.

On Mon, Apr 10, 2017 at 7:37 PM, transmail  wrote:

> I see, so it can go for a while, even if it does not connected properly.
> Good to know.
>
> I replaced the 80 pin cables to 40 pin ones. Does it matter for the 40 pin
> one, how it's connected?
>
> John Paul Adrian Glaubitz  írta:
> >I wouldn't be surprised if it crashed with the cabling being messed up.
> >
> >Frans is absolutely right. Unless you connect the cable properly, don't
> be surprised by obscure crashes or hangs.
> >
> >Adrian
> >
> >> On Apr 10, 2017, at 7:25 PM, transmail  wrote:
> >>
> >> But it works if it's not the blue is connected to the motherboard!
> "boot cdrom" works and it's not the blue which is connected to the
> motherboard. Or even if the machine begins to boot, the later crash is
> because of this? What if i connect the devices with a 40 pin cable?
> >>
> >> Frans van Berckel  írta:
>  On Mon, 2017-04-10 at 19:09 +0200, transmail wrote:
>  And which cable is correct for this machine? Because now "boot cdrom"
>  works with 80 pin too.
> 
>  It's impossible to plug the blue connector to the motherboard. The
>  DVD's and the HDD's connector is too far from each other then. I need
>  to put the middle (gray) one to the motherboard. When i got the
>  machine it was that way.
> >>>
> >>> Okay clear, stop searching any further. Because this the problem. The
> >>> blue cable connector got to be connected to the motherboard, indeed.
> >>>
>  It did not make any change if i mapped the DVD as slave or CS. What
>  mattered that i had to put the DVD on the same bus as the HDD. If the
>  DVD is alone then it only appears if it's the master.
> 
>  I'll get a serial cable.
> >>>
> >>> Thanks,
> >>>
> >>> Frans van Berckel
> >>>
> >>>
> >
> >
>
>

Re: How can i boot *anything* on Sun Blade 100?

[Kevin Stabel]

With IDE/ATA, it is vital that:
- You use the correct cable, 40 or 80 pins
- You place the devices in the right order on the cable. 80 pin cables will
usually have color coding.  Usually blue goes on the motherboard.
- Jumper the options correctly on the devices.  Master/slave on 40 pin
cables, cable select preferred on 80 pin cables.

For the console ...
You will need to set your output-device and input-device via setenv to
switch the console to serial.

On Mon, Apr 10, 2017 at 5:45 PM, transmail  wrote:

> Ok, i'll get a serial cable then. But now, the booting up works, it's now
> the OS-es what do not want to work. I'll get a DVD-R for Solaris and i'll
> wait for the next Debian image. Maybe this'll solve. As for FreeBSD i have
> no idea why it does freeze. Does the Blade still sends messages via serial
> cable after "boot cdrom" ?
>
> John Paul Adrian Glaubitz  írta:
> >On 04/10/2017 03:40 PM, transmail wrote:
> >> I think, it was the SSD which confused the system. Now as SSD is alone
> on the secondary
> >> channel and on the primary, the HDD is the master and the DVD is the
> slave, everything
> >> is visible in "probe-ide" and "boot cdrom" works. But boot FreeBSD and
> Debian 9 dies under
> >> the boot procedure.
> >
> >If you want people to help you, you have to provide the boot messages
> from the serial
> >console. Anything else ends up in a guessing game.
> >
> >Adrian
> >
> >--
> > .''`.  John Paul Adrian Glaubitz
> >: :' :  Debian Developer - glaub...@debian.org
> >`. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
> >  `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
> >
>

Re: How can i boot *anything* on Sun Blade 100?

[Kevin Stabel]

What i think is going on ...

According to this: http://oscomp.hu/depot/probe-ide.jpg
The cdrom drive is not where it should be (it should be slave on the first
bus).
Thus the devalias for "cdrom" is wrong.

OP, move the cdrom to the correct bus, as described in the manual.
https://docs.oracle.com/cd/E19127-01/blade100.ws/806-3416-10/806-3416-10.pdf
Please reference sections 6.2.5 and 6.2.6

You might not think this matters on these machines, but it does.

On Mon, Apr 10, 2017 at 3:47 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 04/10/2017 03:40 PM, transmail wrote:
> > I think, it was the SSD which confused the system. Now as SSD is alone
> on the secondary
> > channel and on the primary, the HDD is the master and the DVD is the
> slave, everything
> > is visible in "probe-ide" and "boot cdrom" works. But boot FreeBSD and
> Debian 9 dies under
> > the boot procedure.
>
> If you want people to help you, you have to provide the boot messages from
> the serial
> console. Anything else ends up in a guessing game.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Sun Blade 100, problems with DVD drive

[Kevin Stabel]

If you don't mind the cost of one disc, you could.
If you do mind the cost, you could try to look up what your drive supports.

On Sun, Apr 9, 2017 at 5:11 PM, transmail <transm...@freemail.hu> wrote:

> Yes, it is. It's a DVD+RW. And both the FreeBSD and Debian CD-s i tried to
> boot up were CD-RW. Should i try CD-R and DVD-R?
>
> Kevin Stabel <kevin.sta...@gmail.com> írta:
> >This is a burned DVD yeah?It could actually be a fairly obscure issue,
> namely DVD-R vs DVD+RPlease have a look here: https://pc.net/
> helpcenter/answers/different_dvd_r_formatsOn Sun, Apr 9, 2017 at 4:51 PM,
> transmail transm...@freemail.hu> wrote:Ive tried to set auto-boot?
> to false but its still ejecting my Solaris DVD. What can cause this?
> >
> >Mark Morgan Lloyd markmll.debian-sp...@telemetry.co.uk> írta:
> >>On 09/04/17 12:00, John Paul Adrian Glaubitz wrote:
> >>> Thats not a proof that this drive works. Try reading a known good
> medium, e.g. a commercially pressed CD and see if that works.
> >>>
> >>> And, even if that works, it could still be that the old CD-ROM drive
> in your SPARC doesnt handle CD-Rs.
> >>>
> >>> You really need to make sure the drive can deal with the CDs
> youve created. Testing them on a modern drive is moot. Modern drives
> are more tolerant with CD-Rs and CD-RWs.
> >>
> >>I think that system had a comparatively-recent IDE-connected DVD, but
> >>mines somewhere waiting to be scrapped and Im afraid Im
> not going
> >>digging for it.
> >>
> >>--
> >>Mark Morgan Lloyd
> >>markMLl .AT. telemetry.co .DOT. uk
> >>
> >>[Opinions above are the authors, not those of his employers or
> colleagues]
> >>
> >>
> >
> >
> >
>

Re: Sun Blade 100, problems with DVD drive

[Kevin Stabel]

This is a burned DVD yeah?

It could actually be a fairly obscure issue, namely DVD-R vs DVD+R

Please have a look here:
https://pc.net/helpcenter/answers/different_dvd_r_formats

On Sun, Apr 9, 2017 at 4:51 PM, transmail  wrote:

> I've tried to set auto-boot? to false but it's still ejecting my Solaris
> DVD. What can cause this?
>
> Mark Morgan Lloyd  írta:
> >On 09/04/17 12:00, John Paul Adrian Glaubitz wrote:
> >> That's not a proof that this drive works. Try reading a known good
> medium, e.g. a commercially pressed CD and see if that works.
> >>
> >> And, even if that works, it could still be that the old CD-ROM drive in
> your SPARC doesn't handle CD-Rs.
> >>
> >> You really need to make sure the drive can deal with the CDs you've
> created. Testing them on a modern drive is moot. Modern drives are more
> tolerant with CD-Rs and CD-RWs.
> >
> >I think that system had a comparatively-recent IDE-connected DVD, but
> >mine's somewhere waiting to be scrapped and I'm afraid I'm not going
> >digging for it.
> >
> >--
> >Mark Morgan Lloyd
> >markMLl .AT. telemetry.co .DOT. uk
> >
> >[Opinions above are the author's, not those of his employers or
> colleagues]
> >
> >
>
>

Re: /boot partition failing to mount

[Kevin Stabel]

I C ... my bad

On Wed, Mar 29, 2017 at 9:29 AM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> Then you don't have a merged /usr yet. This is only present in the 2017
> images I created.
>
> On Mar 29, 2017, at 9:25 AM, Kevin Stabel <kevin.sta...@gmail.com> wrote:
>
> This install is about 6 months old i think.
>
> On Wed, Mar 29, 2017 at 9:21 AM, John Paul Adrian Glaubitz <
> glaub...@physik.fu-berlin.de> wrote:
>
>> Was that system recently installed?
>>
>> If the installation is older, you don't have a merged /usr yet as this is
>> an option to debootstrap which is run during installation.
>>
>> Adrian
>>
>> On Mar 29, 2017, at 9:14 AM, Kevin Stabel <kevin.sta...@gmail.com> wrote:
>>
>> From my system:
>> root@Noise ~# which mount
>> /bin/mount
>>
>>
>> On Wed, Mar 29, 2017 at 8:59 AM, John Paul Adrian Glaubitz <
>> glaub...@physik.fu-berlin.de> wrote:
>>
>>> His problem could be the separate /usr partition which is no longer
>>> supported on modern Linux distributions because of the usr-merge. See his
>>> attached fstab.
>>>
>>> I'm not sure whether the mount command has been moved to /usr/bin yet
>>> though. If yes, this could explain the problem.
>>>
>>> Adrian
>>>
>>> On Mar 29, 2017, at 8:52 AM, Kevin Stabel <kevin.sta...@gmail.com>
>>> wrote:
>>>
>>> Hi Jesse,
>>>
>>> Wrong fs type in fstab?  Is it ext3?
>>> Wrong label in fstab?  Try replacing the UUID=etc etc with /dev/sda1
>>>
>>> On Wed, Mar 29, 2017 at 2:35 AM, Jesse Talavera-Greenberg <
>>> jessetalav...@aol.com> wrote:
>>>
>>>>
>>>> On 03/28/2017 05:30 AM, Jesse Talavera-Greenberg wrote:
>>>>
>>>> However, the /boot partition (which uses ext3) is failing to mount
>>>>
>>>> How does that manifest? What error message do you get? What are the 
>>>> contents
>>>> of your /etc/fstab?
>>>>
>>>> Attached to this e-mail.  And the error's manifestation appeared in the
>>>> logs I posted in my previous e-mail.  Specifically this part:
>>>>
>>>> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /boot...
>>>> Mar 27 22:39:23 motherfscker systemd[1]: var.mount: Directory /var to 
>>>> mount over is not empty, mounting anyway.
>>>> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /var...
>>>> Mar 27 22:39:23 motherfscker kernel: des_sparc64: sparc64 des opcodes not 
>>>> available.
>>>> Mar 27 22:39:23 motherfscker kernel: md5_sparc64: sparc64 md5 opcode not 
>>>> available.
>>>> Mar 27 22:39:23 motherfscker kernel: aes_sparc64: sparc64 aes opcodes not 
>>>> available.
>>>> Mar 27 22:39:23 motherfscker systemd[1]: boot.mount: Mount process exited, 
>>>> code=exited status=32
>>>> Mar 27 22:39:23 motherfscker systemd[1]: Failed to mount /boot.
>>>> Mar 27 22:39:23 motherfscker systemd[1]: Dependency failed for Local File 
>>>> Systems.
>>>>
>>>> and I don't know why.  The weird thing is that I can mount it manually 
>>>> just fine,
>>>>
>>>> How do you mount it manually? Have you compared it to what's in /etc/fstab?
>>>>
>>>> I mount it through `mount /dev/sda1 /boot`.  That's about it.
>>>>
>>>> though if I run systemctl default the console stops responding.
>>>>
>>>> Did you actually read the manpage for systemctl to understand what 
>>>> "systemctl
>>>> default" does?
>>>>
>>>> Quoting:
>>>>
>>>>default
>>>>Enter default mode. This is mostly equivalent to isolate 
>>>> default.target.
>>>> and:
>>>>"isolate" is only valid for start operations and causes all other units 
>>>> to
>>>>be stopped when the specified unit is started. This mode is always used 
>>>> when
>>>>the isolate command is used.
>>>>
>>>> So, "systemctl default" on Debian effectively kills all units except for 
>>>> the ones
>>>> that are wanted by default.target. Don't run "systemctl default".
>>>>
>>>> Probably the default.target should be reconfigured in Debian's systemd 
>>>> package
>>>> to avoid this problem.
>>>>
>>>> I don't understand what this means, can you elaborate?  (I don't know
>>>> very much about configuring Debian.)
>>>>
>>>> That being said, after I manually mounted /boot I was able to SSH into
>>>> the machine like nothing ever happened; it seems like the default Linux
>>>> login prompt just wasn't showing up.  I think there's a boot parameter to
>>>> that effect?  Now I'm confused.
>>>>
>>>
>>>
>>
>

Re: /boot partition failing to mount

[Kevin Stabel]

This install is about 6 months old i think.

On Wed, Mar 29, 2017 at 9:21 AM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> Was that system recently installed?
>
> If the installation is older, you don't have a merged /usr yet as this is
> an option to debootstrap which is run during installation.
>
> Adrian
>
> On Mar 29, 2017, at 9:14 AM, Kevin Stabel <kevin.sta...@gmail.com> wrote:
>
> From my system:
> root@Noise ~# which mount
> /bin/mount
>
>
> On Wed, Mar 29, 2017 at 8:59 AM, John Paul Adrian Glaubitz <
> glaub...@physik.fu-berlin.de> wrote:
>
>> His problem could be the separate /usr partition which is no longer
>> supported on modern Linux distributions because of the usr-merge. See his
>> attached fstab.
>>
>> I'm not sure whether the mount command has been moved to /usr/bin yet
>> though. If yes, this could explain the problem.
>>
>> Adrian
>>
>> On Mar 29, 2017, at 8:52 AM, Kevin Stabel <kevin.sta...@gmail.com> wrote:
>>
>> Hi Jesse,
>>
>> Wrong fs type in fstab?  Is it ext3?
>> Wrong label in fstab?  Try replacing the UUID=etc etc with /dev/sda1
>>
>> On Wed, Mar 29, 2017 at 2:35 AM, Jesse Talavera-Greenberg <
>> jessetalav...@aol.com> wrote:
>>
>>>
>>> On 03/28/2017 05:30 AM, Jesse Talavera-Greenberg wrote:
>>>
>>> However, the /boot partition (which uses ext3) is failing to mount
>>>
>>> How does that manifest? What error message do you get? What are the contents
>>> of your /etc/fstab?
>>>
>>> Attached to this e-mail.  And the error's manifestation appeared in the
>>> logs I posted in my previous e-mail.  Specifically this part:
>>>
>>> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /boot...
>>> Mar 27 22:39:23 motherfscker systemd[1]: var.mount: Directory /var to mount 
>>> over is not empty, mounting anyway.
>>> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /var...
>>> Mar 27 22:39:23 motherfscker kernel: des_sparc64: sparc64 des opcodes not 
>>> available.
>>> Mar 27 22:39:23 motherfscker kernel: md5_sparc64: sparc64 md5 opcode not 
>>> available.
>>> Mar 27 22:39:23 motherfscker kernel: aes_sparc64: sparc64 aes opcodes not 
>>> available.
>>> Mar 27 22:39:23 motherfscker systemd[1]: boot.mount: Mount process exited, 
>>> code=exited status=32
>>> Mar 27 22:39:23 motherfscker systemd[1]: Failed to mount /boot.
>>> Mar 27 22:39:23 motherfscker systemd[1]: Dependency failed for Local File 
>>> Systems.
>>>
>>> and I don't know why.  The weird thing is that I can mount it manually just 
>>> fine,
>>>
>>> How do you mount it manually? Have you compared it to what's in /etc/fstab?
>>>
>>> I mount it through `mount /dev/sda1 /boot`.  That's about it.
>>>
>>> though if I run systemctl default the console stops responding.
>>>
>>> Did you actually read the manpage for systemctl to understand what 
>>> "systemctl
>>> default" does?
>>>
>>> Quoting:
>>>
>>>default
>>>Enter default mode. This is mostly equivalent to isolate 
>>> default.target.
>>> and:
>>> "isolate" is only valid for start operations and causes all other units 
>>> to
>>> be stopped when the specified unit is started. This mode is always used 
>>> when
>>> the isolate command is used.
>>>
>>> So, "systemctl default" on Debian effectively kills all units except for 
>>> the ones
>>> that are wanted by default.target. Don't run "systemctl default".
>>>
>>> Probably the default.target should be reconfigured in Debian's systemd 
>>> package
>>> to avoid this problem.
>>>
>>> I don't understand what this means, can you elaborate?  (I don't know
>>> very much about configuring Debian.)
>>>
>>> That being said, after I manually mounted /boot I was able to SSH into
>>> the machine like nothing ever happened; it seems like the default Linux
>>> login prompt just wasn't showing up.  I think there's a boot parameter to
>>> that effect?  Now I'm confused.
>>>
>>
>>
>

Re: /boot partition failing to mount

[Kevin Stabel]

>From my system:
root@Noise ~# which mount
/bin/mount


On Wed, Mar 29, 2017 at 8:59 AM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> His problem could be the separate /usr partition which is no longer
> supported on modern Linux distributions because of the usr-merge. See his
> attached fstab.
>
> I'm not sure whether the mount command has been moved to /usr/bin yet
> though. If yes, this could explain the problem.
>
> Adrian
>
> On Mar 29, 2017, at 8:52 AM, Kevin Stabel <kevin.sta...@gmail.com> wrote:
>
> Hi Jesse,
>
> Wrong fs type in fstab?  Is it ext3?
> Wrong label in fstab?  Try replacing the UUID=etc etc with /dev/sda1
>
> On Wed, Mar 29, 2017 at 2:35 AM, Jesse Talavera-Greenberg <
> jessetalav...@aol.com> wrote:
>
>>
>> On 03/28/2017 05:30 AM, Jesse Talavera-Greenberg wrote:
>>
>> However, the /boot partition (which uses ext3) is failing to mount
>>
>> How does that manifest? What error message do you get? What are the contents
>> of your /etc/fstab?
>>
>> Attached to this e-mail.  And the error's manifestation appeared in the
>> logs I posted in my previous e-mail.  Specifically this part:
>>
>> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /boot...
>> Mar 27 22:39:23 motherfscker systemd[1]: var.mount: Directory /var to mount 
>> over is not empty, mounting anyway.
>> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /var...
>> Mar 27 22:39:23 motherfscker kernel: des_sparc64: sparc64 des opcodes not 
>> available.
>> Mar 27 22:39:23 motherfscker kernel: md5_sparc64: sparc64 md5 opcode not 
>> available.
>> Mar 27 22:39:23 motherfscker kernel: aes_sparc64: sparc64 aes opcodes not 
>> available.
>> Mar 27 22:39:23 motherfscker systemd[1]: boot.mount: Mount process exited, 
>> code=exited status=32
>> Mar 27 22:39:23 motherfscker systemd[1]: Failed to mount /boot.
>> Mar 27 22:39:23 motherfscker systemd[1]: Dependency failed for Local File 
>> Systems.
>>
>> and I don't know why.  The weird thing is that I can mount it manually just 
>> fine,
>>
>> How do you mount it manually? Have you compared it to what's in /etc/fstab?
>>
>> I mount it through `mount /dev/sda1 /boot`.  That's about it.
>>
>> though if I run systemctl default the console stops responding.
>>
>> Did you actually read the manpage for systemctl to understand what "systemctl
>> default" does?
>>
>> Quoting:
>>
>>default
>>Enter default mode. This is mostly equivalent to isolate 
>> default.target.
>> and:
>>  "isolate" is only valid for start operations and causes all other units 
>> to
>>  be stopped when the specified unit is started. This mode is always used 
>> when
>>  the isolate command is used.
>>
>> So, "systemctl default" on Debian effectively kills all units except for the 
>> ones
>> that are wanted by default.target. Don't run "systemctl default".
>>
>> Probably the default.target should be reconfigured in Debian's systemd 
>> package
>> to avoid this problem.
>>
>> I don't understand what this means, can you elaborate?  (I don't know
>> very much about configuring Debian.)
>>
>> That being said, after I manually mounted /boot I was able to SSH into
>> the machine like nothing ever happened; it seems like the default Linux
>> login prompt just wasn't showing up.  I think there's a boot parameter to
>> that effect?  Now I'm confused.
>>
>
>

Re: /boot partition failing to mount

[Kevin Stabel]

Hi Jesse,

Wrong fs type in fstab?  Is it ext3?
Wrong label in fstab?  Try replacing the UUID=etc etc with /dev/sda1

On Wed, Mar 29, 2017 at 2:35 AM, Jesse Talavera-Greenberg <
jessetalav...@aol.com> wrote:

>
> On 03/28/2017 05:30 AM, Jesse Talavera-Greenberg wrote:
>
> However, the /boot partition (which uses ext3) is failing to mount
>
> How does that manifest? What error message do you get? What are the contents
> of your /etc/fstab?
>
> Attached to this e-mail.  And the error's manifestation appeared in the
> logs I posted in my previous e-mail.  Specifically this part:
>
> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /boot...
> Mar 27 22:39:23 motherfscker systemd[1]: var.mount: Directory /var to mount 
> over is not empty, mounting anyway.
> Mar 27 22:39:23 motherfscker systemd[1]: Mounting /var...
> Mar 27 22:39:23 motherfscker kernel: des_sparc64: sparc64 des opcodes not 
> available.
> Mar 27 22:39:23 motherfscker kernel: md5_sparc64: sparc64 md5 opcode not 
> available.
> Mar 27 22:39:23 motherfscker kernel: aes_sparc64: sparc64 aes opcodes not 
> available.
> Mar 27 22:39:23 motherfscker systemd[1]: boot.mount: Mount process exited, 
> code=exited status=32
> Mar 27 22:39:23 motherfscker systemd[1]: Failed to mount /boot.
> Mar 27 22:39:23 motherfscker systemd[1]: Dependency failed for Local File 
> Systems.
>
> and I don't know why.  The weird thing is that I can mount it manually just 
> fine,
>
> How do you mount it manually? Have you compared it to what's in /etc/fstab?
>
> I mount it through `mount /dev/sda1 /boot`.  That's about it.
>
> though if I run systemctl default the console stops responding.
>
> Did you actually read the manpage for systemctl to understand what "systemctl
> default" does?
>
> Quoting:
>
>default
>Enter default mode. This is mostly equivalent to isolate 
> default.target.
> and:
>   "isolate" is only valid for start operations and causes all other units 
> to
>   be stopped when the specified unit is started. This mode is always used 
> when
>   the isolate command is used.
>
> So, "systemctl default" on Debian effectively kills all units except for the 
> ones
> that are wanted by default.target. Don't run "systemctl default".
>
> Probably the default.target should be reconfigured in Debian's systemd package
> to avoid this problem.
>
> I don't understand what this means, can you elaborate?  (I don't know very
> much about configuring Debian.)
>
> That being said, after I manually mounted /boot I was able to SSH into the
> machine like nothing ever happened; it seems like the default Linux login
> prompt just wasn't showing up.  I think there's a boot parameter to that
> effect?  Now I'm confused.
>

Re: Subject=Re: New sparc64 installation images 2017-03-14

[Kevin Stabel]

Hi,

This is indeed openboot throwing this error.

Make sure your nvramrc doesn't contain any strange or unnecessary commands.
If you for example try to do a probe-scsi or probe-scsi-all before booting,
you have a high chance of getting this error.
Random commands in OBP can leave the memory in a state that prevents
booting.

I've seen it happen many times when i was playing with the fcode functions
of my onboard FC adaptor.

I just do a reset-all when it happens, and that seems to clear it up.

Kind Regards,

-Kevin

On Sat, Mar 25, 2017 at 6:43 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 03/25/2017 06:04 PM, Anatoly Pugachev wrote:
> > I had this error on my e250 about 10 years ago when I had debian
> > installed on it. The only way I could boot succesfully next time,
> > is to power off machine completely via alom (or whatever e250 had
> > for remote access, can't remember), and after power on it could
> > load/boot to silo.
>
> Either way, I think this error message comes from the firmware and
> not SILO, so this is something someone needs to answer who is more
> knowledgeable with the inner workings of the firmware.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>
>

Re: non-free firmware

[Kevin Stabel]

Thank you, Adrian.

These are not stupid questions at all :)
I have booted the right kernel, updated the initrd, the firmware was there,
etc.
Patch was applied with the usual "patch -p1 < patch" method.
Two chunks patched.
Looked ok.

I will contact Meelis Roos and will keep you all posted here.
It would be great if we could get this working.

Kind Regards,

-Kevin

On Tue, Mar 21, 2017 at 10:34 AM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On Tue, Mar 21, 2017 at 10:02:56AM +0100, Frans van Berckel wrote:
> > And yes, because Debian sparc64 runs very well with a 3.16 kernel-
> > image. The sparc backported one to be clear. I extracted the package
> > and copied it manually into the new rootfs. And updated the silo.conf.
>
> Are you talking about the driver or the firmware?
>
> > But maybe we just have to build a sparc64 package for it :-)
>
> No, we have to fix the actual bug. Jamming an old, unsupported kernel
> into a sparc64 package is a workaround, not a fix.
>
> > > I also saw your report in the mailing list archives.
> >
> > And a real bug post, so being able tracking it ...
> >
> > https://bugzilla.kernel.org/show_bug.cgi?id=110081
>
> You should add some SPARC kernel people to the CC of this bug report
> and also post this to the sparclinux kernel mailing list so you get
> some attention.
>
> > > I applied the following patch to a vanilla 4.9.0 kernel, and it
> > > didn't work either.
> > > https://patchwork.kernel.org/patch/6533041/
>
> This very much looks like the root of the cause. Are you sure you
> properly applied and booted the right kernel? Sorry for asking stupid
> questions, but I have made similar mistakes in the past myself, so
> sometimes this cannot be ruled out :).
>
> > > If anyone has any ideas, i would greatly appreciate their input, as
> > > it would be nice to use the controller given that i just got a batch
> > > of disks to use with the built in backplane.
> >
> > I am open for new ideas as well.
>
> I would talk Meelis Roos who wrote the above patch. He probably has
> some insight.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: non-free firmware

[Kevin Stabel]

Hello Frans,

Did you get the qlogic FC controller working in the end?
I am facing the same issue now actually.  It's still broken in the latest
kernel in the repo.

I also saw your report in the mailing list archives.

I've been able to find some information on it.
It seems to have to do with endianness, and the way the configuration of
the nvram on the card is read.
Do you also get a message about inconsistent nvram configuration?

I applied the following patch to a vanilla 4.9.0 kernel, and it didn't work
either.
https://patchwork.kernel.org/patch/6533041/

If anyone has any ideas, i would greatly appreciate their input, as it
would be nice to use the controller given that i just got a batch of disks
to use with the built in backplane.

Kind Regards,

-Kevin


On Tue, Mar 21, 2017 at 8:38 AM, Frans van Berckel 
wrote:

> Hi Rick,
>
> On Tue, 2017-03-21 at 03:06 -0400, Rick Leir wrote:
> > Hi all,
> >
> > I am just trying out Adrian's latest iso (thanks Adrian!!).
> >
> > (On a SunBlade 2000 (circa 2002) with two Cu processors and 1800M
> > RAM, booting from the CDROM): I tapped on the 'choose language,
> > choose keyboard, choose..' screen a bit and it jumped ahead without
> > me being finished there. OK, now I understand: the fourth item is
> > 'Detect and mount CD-ROM', and I must have chosen that.
> >
> > Now it is telling me that I need some non-free firmware, filename
> > ql2200_fw.bin, and that I should load it from removable media. Does
> > this firmware matter, and where can I get it? There is something by
> > that name on the qlogic.com site. I guess it could be for the
> > FibreChannel interface. OK, ignoring it for now.
> >
> > After doing 'set up users and passwords' correctly it goes on to
> > 'Configure the clock'. It returns from that immediately without
> > doing anything. I was hoping to set the timezone.
> >
> > Now it is doing 'Detecting disks'. Gee, it took a long time on that,
> > maybe a few minutes. Then it says 'No disk drive was detected' and
> > gives me a list of drivers to select from. I am guessing now: qla2xxx
> > ... long pause, then I am back to the list of drivers screen. Try
> > 'qla1280' ... 'qla4xxx' .. help! I think the disk hardware is OK
> > because OpenBSD is installed and can boot. What can I use for a disk
> > driver? How could I load the ql2200_fw.bin?
>
> I have done some SunBlade 2000 installs. If i am well the Debian
> installer ask for the firmware while installing.
>
> What I did just copy the firmware to a USB stick. It automatily detects
> the bin and I was able to detect the disks.
>
> But just after that, on early version my kernel crashes several times,
> because it's not handling the FibreChannel controller well. Did report
> this issue upstream a year ago, so maybe its solved now.
>
> I downloaded this firmware-qlogic package. Extracted it manually and
> copied my bin to the root of the USB stick. If in your case the
> installer does not ask for the firmware, you can manually copy it to
> /lib/firmware of the installer. With ctrl + f2 you will get a prompt.
>
> https://packages.debian.org/sid/firmware-qlogic
>
> Thanks,
>
>
> Frans van Berckel
>
>

Re: Firefox build environment

[Kevin Stabel]

OK, i tried several times from their source tree with a custom mozconfig,
but could never get it to work.

I'll check it out.  Thanks!

On Sun, Mar 12, 2017 at 12:12 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 03/12/2017 12:04 PM, Kevin Stabel wrote:
> > Can you share your build environment settings for firefox please?
>
> I just use the Debian packages from unstable and experimental when testing
> my
> patches, I haven't built Firefox from git yet as it requires additional
> patches
> and configure options to build on Debian.
>
> I simply build with sbuild --purge=never --arch=sparc64 -d sid
> --no-arch-all.
>
> After the build fails, I just chroot into the still existing build chroot.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Firefox build environment

[Kevin Stabel]

Hello Adrian,

Can you share your build environment settings for firefox please?

I would like to have a go at building it, and try to resolve an issue which
seems to be related to endianness in the image rendering.

Thank you for your time and kind regards.

-Kevin

Re: X in debian-9.0-sparc on qemu

[Kevin Stabel]

Can you run just Xtigervnc?
Just type Xtigervnc and see what happens.

There might be a problem with the package ...

On Tue, Mar 7, 2017 at 10:19 AM, sacarde <saca...@tiscali.it> wrote:

> >Alle martedì 7 marzo 2017, Kevin Stabel ha scritto:
> >> Tigervnc works well for me.
>
> I try installing:
> tigervnc-common
> tigervnc-standalone-server
>
> but when I start:
>
> tigervncserver -interface 192.168.1.7 -geometry 800x600 -cleanstale
>
> I have:
> tigervncserver: /usr/bin/Xtigervnc did not start up, please look
> into '/home/sacarde/.vnc/debian:1.log' to determine the reason! -1
> Killing Xtigervnc process ID 719... success!
>
> and /home/sacarde/.vnc/debian:1.log is empy
>
>
> thanks
>
> saca...@tiscali.it
>
>

Re: X in debian-9.0-sparc on qemu

[Kevin Stabel]

Tigervnc works well for me.

On Tue, Mar 7, 2017 at 12:23 AM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 03/07/2017 12:05 AM, Kevin Stabel wrote:
> > I've had some mixed success in building x2go.
> > I will give it another go soon.
>
> Alternatively, the OP can just use TightVNCServer like we did
> successfully in December [1].
>
> Adrian
>
> > [1] https://lists.debian.org/debian-sparc/2016/12/msg00054.html
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: X in debian-9.0-sparc on qemu

[Kevin Stabel]

I've had some mixed success in building x2go.
I will give it another go soon.

On Mon, Mar 6, 2017 at 11:54 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 03/06/2017 11:12 PM, Rick Leir wrote:
> > Have you tried X2go? If it works for you it will be faster than vnc
> > http://x2go.org
> > https://launchpad.net/~x2go/+archive/ubuntu/stable
>
> There are currently no sparc64 binaries available for x2go-server.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>
>

X2go

[Kevin Stabel]

Hi,

As suggested a few days ago, perhaps a working x2go build would be useful.
So far, i have nxlibs, x2goserver, x2goagent, etc all built and running.

However ... the client connects, something dies, and then the client
immediately disconnects.

Adrian, you said you had previous experience with x2go ... perhaps you know
where to look?
I have set the log level for x2goserver to debug, but nothing useful about
how and when it dies exactly.
Just the X clients complaining about a disappearing display.

Kind Regards,

-Kevin

V880: bbc.ko panics the machine

[Kevin Stabel]

Hi,

Is anyone else encountering this problem?
I get a hard crash, machine proceeds to dump cpu registers, etc etc.

It'd be nice to get envctrl to do it's thing and slow down the fans a bit
...

FYI:

Machine is a V880, 8 cpus

# cat /proc/cpuinfo
cpu : TI UltraSparc III+ (Cheetah+)
fpu : UltraSparc III+ integrated FPU
pmu : ultra3+
prom: OBP 4.18.2 2005/06/30 07:53
type: sun4u
ncpus probed: 8
ncpus active: 8
D$ parity tl1   : 0
I$ parity tl1   : 0
cpucaps :
flush,stbar,swap,muldiv,v9,ultra3,mul32,div32,v8plus,vis,vis2
Cpu0ClkTck  : 47868c00
Cpu1ClkTck  : 47868c00
Cpu2ClkTck  : 47868c00
Cpu3ClkTck  : 47868c00
Cpu4ClkTck  : 47868c00
Cpu5ClkTck  : 47868c00
Cpu6ClkTck  : 47868c00
Cpu7ClkTck  : 47868c00
MMU Type: Cheetah+
MMU PGSZs   : 8K,64K,512K,4MB
State:
CPU0:   online
CPU1:   online
CPU2:   online
CPU3:   online
CPU4:   online
CPU5:   online
CPU6:   online
CPU7:   online


# cat /sys/firmware/devicetree/base/banner-name
Sun Fire 880⏎

# cat /sys/firmware/devicetree/base/openprom/version
OBP 4.18.2 2005/06/30 07:53⏎



# uname -a
Linux Noise 4.8.0-2-sparc64-smp #1 SMP Debian 4.8.11-1 (2016-12-02) sparc64
GNU/Linux

#lspci
:00:01.0 SCSI storage controller: LSI Logic / Symbios Logic 53c875 (rev
04)
:00:03.0 Fibre Channel: Emulex Corporation LP8000 Fibre Channel Host
Adapter (rev 02)
:00:04.0 SCSI storage controller: Marvell Technology Group Ltd.
MV88SX6081 8-port SATA II PCI-X Controller (rev 09)
:00:05.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
:01:04.0 SCSI storage controller: QLogic Corp. QLA2200 64-bit Fibre
Channel Adapter (rev 05)
:01:05.0 SCSI storage controller: QLogic Corp. QLA2200 64-bit Fibre
Channel Adapter (rev 05)
0001:00:01.0 Ethernet controller: Oracle/SUN GEM 10/100/1000 Ethernet [ge]
(rev 01)
0001:00:02.0 SCSI storage controller: QLogic Corp. QLA2200 64-bit Fibre
Channel Adapter (rev 05)
0002:00:01.0 Bridge: Oracle/SUN RIO EBUS (rev 01)
0002:00:01.1 Ethernet controller: Oracle/SUN RIO 10/100 Ethernet [eri] (rev
01)
0002:00:01.3 USB controller: Oracle/SUN RIO USB (rev 01)
0002:00:02.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
0002:00:03.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
0002:00:04.0 Fibre Channel: Emulex Corporation LP8000 Fibre Channel Host
Adapter (rev 02)
0002:01:00.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
0002:01:04.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
0002:02:00.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:02:01.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:03:02.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:03:03.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:04:00.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
0002:04:04.0 PCI bridge: Intel Corporation 21154 PCI-to-PCI Bridge
0002:05:00.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:05:01.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:06:02.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0002:06:03.0 Ethernet controller: National Semiconductor Corporation
DP83065 [Saturn] 10/100/1000 Ethernet Controller (rev 30)
0003:00:01.0 Fibre Channel: QLogic Corp. ISP2422-based 4Gb Fibre Channel to
PCI-X HBA (rev 02)
0003:00:02.0 Fibre Channel: QLogic Corp. ISP2422-based 4Gb Fibre Channel to
PCI-X HBA (rev 02)

Any help would be appreciated.

Kind Regards,

-Kevin

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

Hi Joost,

Does you machine have a gfx card and monitor?
If the machine is reasonably specced, perhaps you could try playing some
video on it to see if you get the strange blue tint as well?

Kind Regards,

-Kevin

On Wed, Dec 21, 2016 at 5:05 PM, Joost van Baal-Ilić <
joostvb-debian-sp...@mdcc.cx> wrote:

> On Mon, Dec 19, 2016 at 04:29:26PM +0100, John Paul Adrian Glaubitz wrote:
> > On 12/19/2016 04:27 PM, Kevin Stabel wrote:
> > > I (cautiously) report success!
> > > In fact, i am replying to you right now from within your build :)
> > >
> > > Here is a screenshot of a site that crashed the browser before:
> >
> > Man, this is so awesome. Makes me proud we already got this far!
>
> Congrats indeed!  Thanks a lot for your work,
>
> Joost - who showed a sparc running a previous Debian @ the Debian booth at
> the
>  T-DOSE conference in Eindhoven, The Netherlands, some months ago
>
>
>

Re: Suggested VNC server setup

[Kevin Stabel]

I've also used x2go on sparc when this machine was running gentoo.
It performs better, uses nox.
I don't see it in the debian repos, though.  Maybe i'll try and build the
server.

On Tue, Dec 20, 2016 at 8:47 AM, Mark Morgan Lloyd <
markmll.debian-sp...@telemetry.co.uk> wrote:

> On 19/12/16 18:30, John Paul Adrian Glaubitz wrote:
>
>> Hi Kevin!
>>
>> Can you tell me which VNC server you are using on your sparc64 box?
>> I have tried various servers and none really worked for me.
>>
>
> As a general point- not specifically SPARC- I think the work is now going
> into TigerVNC. It certainly has improvements compared with the standard
> server available in Jessie and older, but IME some ways of using it require
> XDMCP in the display manager and that's now getting comparatively rare-
> we've had to fall back to xdm here.
>
> --
> Mark Morgan Lloyd
> markMLl .AT. telemetry.co .DOT. uk
>
> [Opinions above are the author's, not those of his employers or colleagues]
>
>

Re: Suggested VNC server setup

[Kevin Stabel]

OK, it's working now?

On Mon, Dec 19, 2016 at 7:51 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/19/2016 07:41 PM, Kevin Stabel wrote:
> > sudo apt-get install tigervnc-standalone-server
> > echo "exec mate-session" > ~/.Xsession# Replace mate
> with whatever
> > tigervncserver -localhost no -cleanstale
>
> Ok, that's actually what I did. But I think the problem is IPv6
> which must be turned off.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Suggested VNC server setup

[Kevin Stabel]

sudo apt-get install tigervnc-standalone-server
echo "exec mate-session" > ~/.Xsession# Replace mate with
whatever
tigervncserver -localhost no -cleanstale

This is assuming you already have a working Xorg setup with all necessary
libs of course.

On Mon, Dec 19, 2016 at 7:36 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/19/2016 07:34 PM, Kevin Stabel wrote:
> > Are you running into trouble?
> > Starting the tigervncserver in the way i indicated will start it's own X
> session, aka it won't "share" your desktop.
>
> Yes. And can you quickly share what you did?
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Suggested VNC server setup

[Kevin Stabel]

I just installed it from the repos.

Are you running into trouble?
Starting the tigervncserver in the way i indicated will start it's own X
session, aka it won't "share" your desktop.
To do that' you'll need x0vncserver, or x11vnc.
x11vnc is in the repos.

On Mon, Dec 19, 2016 at 7:28 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/19/2016 07:16 PM, Kevin Stabel wrote:
> > Tigervnc works well.
> > Use RAW encoding and disable JPEG compression if used on a local network
> to limit the overhead.
> > Firefox is very usable like this actually!  I am still enjoying it and
> testing it as we speak :)
>
> Is there some sort of idiot's guide for setting this up? I don't want to
> spend the rest of the
> evening trying to figure this out. I just want to connect to a running X
> server to play around
> with Firefox and Co, but getting VNC set up is frustratingly overly
> complicated.
>
> Thanks,
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Suggested VNC server setup

[Kevin Stabel]

Hey :)

Tigervnc works well.
Use RAW encoding and disable JPEG compression if used on a local network to
limit the overhead.
Firefox is very usable like this actually!  I am still enjoying it and
testing it as we speak :)

once installed, define your Xsession, and launch the server as unprivileged
user with:
tigervncserver -localhost no -cleanstale

On Mon, Dec 19, 2016 at 7:14 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> Hi Kevin!
>
> Can you tell me which VNC server you are using on your sparc64 box?
> I have tried various servers and none really worked for me.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

This machine is headless.  It's a V880.
I could add a PCI video card to it if i could find one that would work
among all my old stuff, but it would definitely not be something that i
could do quickly.  Most of it sits in storage.

On Mon, Dec 19, 2016 at 4:58 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/19/2016 04:56 PM, Kevin Stabel wrote:
> > I can even say HTML5 video seems to somewhat work.
>
> Cool.
>
> > There seems to be a colorspace issue at least.  I get a blue tint, but
> only on the video area (much like the X forwarding issue), and the frames
> seem to
> > overlay.  That is to say, the frames don't clear, they just add up
> infinitely (quite trippy), but the decoding works.
> > Video is not usable right now, but promising for sure.
>
> Might be related to VNC in this case. You don't really get hardware
> acceleration in this case.
>
> Any chance you can try this with a real monitor hooked up? I should try on
> my old desktop
> sparc64 machine as well. Although it will be a bit sluggish.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

Indeed!

I can even say HTML5 video seems to somewhat work.
There seems to be a colorspace issue at least.  I get a blue tint, but only
on the video area (much like the X forwarding issue), and the frames seem
to overlay.  That is to say, the frames don't clear, they just add up
infinitely (quite trippy), but the decoding works.
Video is not usable right now, but promising for sure.


On Mon, Dec 19, 2016 at 4:29 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/19/2016 04:27 PM, Kevin Stabel wrote:
> > I (cautiously) report success!
> > In fact, i am replying to you right now from within your build :)
> >
> > Here is a screenshot of a site that crashed the browser before:
>
> Man, this is so awesome. Makes me proud we already got this far!
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

>
> I see. Can you tell me what you did to trigger the crash?


Surf to gmail, engadget, gizmodo, etc.
Just some JS intensive sites.

Did browse osnews with it just fine.

On Sun, Dec 18, 2016 at 10:32 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/18/2016 10:30 PM, Kevin Stabel wrote:
> > It's more stable, until you try something with JS.
>
> I see. Can you tell me what you did to trigger the crash?
>
> > Do you want an account on this machine?
> > It has a DE setup and vnc working.
> > You can have a look maybe?
>
> Thanks, but that's not really necessary. I just need to know what you
> did to trigger the crash. It's very likely that the JS allocator does
> not ensure aligned access and therefore provokes crashes.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

It's more stable, until you try something with JS.

Do you want an account on this machine?
It has a DE setup and vnc working.
You can have a look maybe?

On Sun, Dec 18, 2016 at 10:29 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/18/2016 10:23 PM, Kevin Stabel wrote:
> > Tried this build as well.  Deleted all the cache, the profile, etc.
> Seems a bit more stable, but still crashes with bus error.
> > Colors are correct on vnc, however, i can confirm the blue tint over ssh
> forward.
>
> Huh? Did I forget to apply the patch? The patch definitely fixed it for me
> and I could browse normally without crashes. Odd.
>
> I will have a look to verify.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

Tried this build as well.  Deleted all the cache, the profile, etc.  Seems
a bit more stable, but still crashes with bus error.
Colors are correct on vnc, however, i can confirm the blue tint over ssh
forward.

On Sun, Dec 18, 2016 at 10:18 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/18/2016 10:16 PM, John Paul Adrian Glaubitz wrote:
> > On 12/18/2016 09:29 PM, Kevin Stabel wrote:
> >> I just tried it.  It launches and runs (!!!)
> >> Bus error as you said, but looks very promising :)
> >> I was able to get to google for a short time.
> >> I get a lot of "[GFX1} Unknown image format" errors.
> >
> > The bus error will go away with the next update. I have uploaded the
> > fixed version to the FTP servers. The fixed version is 50.0.2-1.1.
>
> And if you don't want to wait that long, just download the package here:
>
> > http://incoming.ports.debian.org/firefox_50.0.2-1.1_sparc64.deb
>
> As root:
>
> $ wget http://incoming.ports.debian.org/firefox_50.0.2-1.1_sparc64.deb
> $ dpkg -i firefox_50.0.2-1.1_sparc64.deb
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

Looks correct on vnc.
Got the same blue tint on ssh forwarding, so i can confirm.
I can try the new build if you want.

On Sun, Dec 18, 2016 at 10:16 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/18/2016 09:29 PM, Kevin Stabel wrote:
> > I just tried it.  It launches and runs (!!!)
> > Bus error as you said, but looks very promising :)
> > I was able to get to google for a short time.
> > I get a lot of "[GFX1} Unknown image format" errors.
>
> The bus error will go away with the next update. I have uploaded the
> fixed version to the FTP servers. The fixed version is 50.0.2-1.1.
>
> Btw, can you tell me whether the colors look correct for you? I tested
> it with X-forwarding and my Firefox had a blue tint.
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>

Re: Firefox now available on Debian/sparc64

[Kevin Stabel]

Hi!

I just tried it.  It launches and runs (!!!)
Bus error as you said, but looks very promising :)
I was able to get to google for a short time.
I get a lot of "[GFX1} Unknown image format" errors.

It doesn't run for very for me yet, but it shows a lot of promise!

Thank you so much!

Kind Regards,

-Kevin

On Sun, Dec 18, 2016 at 8:43 PM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> On 12/12/2016 03:38 PM, John Paul Adrian Glaubitz wrote:
> > So, I just tried it myself. It actually starts and shows the main window,
> > but it crashes with SIGBUS at some point. However, this is something that
> > should be easy to debug with gdb.
>
> Michael Karcher helped me locate the issue. And, of course, it's a matter
> of
> using memcpy() to fix the bug. Will send a patch upstream.
>
> I'm now browsing the web with Firefox on Linux/sparc64 :).
>
> Adrian
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>
>

Re: New Debian sparc installation images

[Kevin Stabel]

Hi Adrian,

Thank you for doing this!
I have tested the installation image, and encountered a problem with
installation on iscsi devices.

Just tells me it can't find any targets.  Nothing in the installer logs.

Are the necessary packages for iscsi included in the installation image?
If so, i guess i should investigate further.  It was rather late when i
tried this, and i didn't put much effort into it yet.

Kind Regards,

-Kevin

On Tue, Nov 22, 2016 at 9:43 AM, John Paul Adrian Glaubitz <
glaub...@physik.fu-berlin.de> wrote:

> Hi!
>
> After a half year break, I have finally created new installer images for
> Debian sparc64. These are currently being uploaded to my Debian webspace
> and should be completely uploaded within the next minutes, see [1].
>
> A quick and rough installation guide can be found in [2], please let
> us know if you encounter any issues or want to give feedback. These
> images have not been tested so far as I just generated them this
> night.
>
> Thanks,
> Adrian
>
> > [1] https://people.debian.org/~glaubitz/debian-cd/2016-11-22/
> > [2] https://lists.debian.org/debian-sparc/2016/06/msg00126.html
>
> --
>  .''`.  John Paul Adrian Glaubitz
> : :' :  Debian Developer - glaub...@debian.org
> `. `'   Freie Universitaet Berlin - glaub...@physik.fu-berlin.de
>   `-GPG: 62FF 8A75 84E0 2956 9546  0006 7426 3B37 F5B5 F913
>
>