Re: [gentoo-user] Why no love for amd64 hardened?

[Pandu Poluan]

On Thu, Nov 3, 2011 at 12:56, Francisco Blas Izquierdo Riera
(klondike)  wrote:
>
> Issues with the autobuild system which we still haven't managed to figure
> out and which seem related to lack of memory. Jmbsviceto is publishing some
> more recent ones on http://www.jmbsvicetto.name and this is the place where
> you should look for the stage3s until we fix the issues.

I see...

Thanks for the information!

I've been building my own hardened stage3.1 for amd64. Just in case
this might be helpful to you:

* To compile gcc-4.5.3, I need 1 GiB of RAM (800 MiB also seems to
work -- tested using VMware so I can experiment using strange numbers)
* MAKEOPTS="-j5" *always* resulted in errors; MAKEOPTS="-j3" seems to
work okay. MAKEOPTS="-j1" is of course the safest.

Rgds,
-- 
FdS Pandu E Poluan
~ IT Optimizer ~

 • LOPSA Member #15248
 • Blog : http://pepoluan.tumblr.com
 • Linked-In : http://id.linkedin.com/in/pepoluan

Re: [gentoo-user] Why no love for amd64 hardened?

[Francisco Blas Izquierdo Riera (klondike)]

El 03/11/11 06:41, Pandu Poluan escribió:
> Why amd64 hardened stage3 never got any updates?
>
> I've checked here:
>
> http://distfiles.gentoo.org/releases/amd64/autobuilds/current-stage3-amd64-hardened/
>
> And here:
>
> http://ftp.jaist.ac.jp/pub/Linux/Gentoo/releases/amd64/autobuilds/current-stage3-amd64-hardened/
>
> the latest amd64 hardened stage3 are both dated 20110625 there.
>
> Contrast this with x86 hardened:
>
> http://distfiles.gentoo.org/releases/x86/autobuilds/current-stage3/hardened/
>
> or
>
> http://distfiles.gentoo.org/releases/x86/autobuilds/current-stage3/hardened/
>
> What's up with the 4 months non-update?
>
> Rgds,
Issues with the autobuild system which we still haven't managed to
figure out and which seem related to lack of memory. Jmbsviceto is
publishing some more recent ones on http://www.jmbsvicetto.name and this
is the place where you should look for the stage3s until we fix the issues.


signature.asc
Description: OpenPGP digital signature

[gentoo-user] Why no love for amd64 hardened?

[Pandu Poluan]

Why amd64 hardened stage3 never got any updates?

I've checked here:

http://distfiles.gentoo.org/releases/amd64/autobuilds/current-stage3-amd64-hardened/

And here:

http://ftp.jaist.ac.jp/pub/Linux/Gentoo/releases/amd64/autobuilds/current-stage3-amd64-hardened/

the latest amd64 hardened stage3 are both dated 20110625 there.

Contrast this with x86 hardened:

http://distfiles.gentoo.org/releases/x86/autobuilds/current-stage3/hardened/

or

http://distfiles.gentoo.org/releases/x86/autobuilds/current-stage3/hardened/

What's up with the 4 months non-update?

Rgds,
-- 
FdS Pandu E Poluan
~ IT Optimizer ~

 • LOPSA Member #15248
 • Blog : http://pepoluan.tumblr.com
 • Linked-In : http://id.linkedin.com/in/pepoluan

Re: [gentoo-user] Does this drive need a funeral?

[Mick]

On Tuesday 01 Nov 2011 18:58:14 Dale wrote:
> Hi,
> 
> For the first time in my life, I think I have a drive failing on me.
> Here is the info:
[snip ...]

> === START OF READ SMART DATA SECTION ===
> SMART overall-health self-assessment test result: FAILED!
> Drive failure expected in less than 24 hours. SAVE ALL DATA.

That's not encouraging ...  :-/

Then this:

> SMART Attributes Data Structure revision number: 16
> Vendor Specific SMART Attributes with Thresholds:
> ID# ATTRIBUTE_NAME  FLAG VALUE WORST THRESH TYPE
> UPDATED  WHEN_FAILED RAW_VALUE
>1 Raw_Read_Error_Rate 0x000b   018   001   051Pre-fail
> Always   FAILING_NOW 1904

The Value (018) being lower than the Threshold (051) indicates failure is 
imminent.  Of course it also screams at you that it's "FAILING NOW" in case 
you missed it!  :-))


>3 Spin_Up_Time0x0007   087   084   021Pre-fail
> Always   -   2166
>4 Start_Stop_Count0x0032   099   099   040Old_age
> Always   -   1288
>5 Reallocated_Sector_Ct   0x0033   199   199   140Pre-fail
> Always   -   1
>7 Seek_Error_Rate 0x000b   200   200   051Pre-fail
> Always   -   0
>9 Power_On_Hours  0x0032   023   023   000Old_age
> Always   -   56466
>   10 Spin_Retry_Count0x0013   100   100   051Pre-fail
> Always   -   0
>   11 Calibration_Retry_Count 0x0013   100   100   051Pre-fail
> Always   -   0
>   12 Power_Cycle_Count   0x0032   099   099   000Old_age
> Always   -   1039
> 194 Temperature_Celsius 0x0022   110   253   000Old_age
> Always   -   33
> 196 Reallocated_Event_Count 0x0032   199   199   000Old_age
> Always   -   1
> 197 Current_Pending_Sector  0x0012   199   199   000Old_age
> Always   -   17
> 198 Offline_Uncorrectable   0x0012   200   200   000Old_age
> Always   -   10
> 199 UDMA_CRC_Error_Count0x000a   200   253   000Old_age
> Always   -   1155
> 200 Multi_Zone_Error_Rate   0x0009   195   085   051Pre-fail
> Offline  -   191

The remaining values look OK.
-- 
Regards,
Mick


signature.asc
Description: This is a digitally signed message part.

Re: [gentoo-user] OT: Arm Cluster board

[Nick Khamis]

Someone watched some news regarding HP moving towards ARM ;)

http://www.mercurynews.com/business/ci_19240331

Nick.

On Wed, Nov 2, 2011 at 4:07 PM, James  wrote:
> So Sorry,
>
> I just think that in a few years, Gentoo will be more
> about ARM( 64 or 128 bit) than Intel...
>
> Check out the cluster on a board in the link below!
> (and you can run it from a solar panel + battery).
> Low power is King.
>
> "72 quad-core ARM servers takes up just one rack unit equivalent of space."
>
> http://www.eetimes.com/electronics-news/4230267/Calxeda-could-create-a-32-bit-microserver-market?cid=NL_EETimesDaily
>
>
> enjoy,
> James
>
>
>
>
>

[gentoo-user] OT: Arm Cluster board

[James]

So Sorry,

I just think that in a few years, Gentoo will be more
about ARM( 64 or 128 bit) than Intel...

Check out the cluster on a board in the link below!
(and you can run it from a solar panel + battery).
Low power is King.

"72 quad-core ARM servers takes up just one rack unit equivalent of space."

http://www.eetimes.com/electronics-news/4230267/Calxeda-could-create-a-32-bit-microserver-market?cid=NL_EETimesDaily


enjoy,
James

Re: [gentoo-user] Wireless network card not loaded on first boot after shutdown

[Canek Peláez Valdés]

On Wed, Nov 2, 2011 at 3:29 AM, Sebastian Beßler
 wrote:
> Hi,
>
> I have a wireless card that works most of the time perfect with the
> ath9k kernel module but not on first boot after a few hours long shutdown.
>
>  * Bringing up interface wlan0
>  *   ERROR: interface wlan0 does not exist
>  *   Ensure that you have loaded the correct kernel module for your hardware
>  * ERROR: net.wlan0 failed to start
>
> A reboot fixes the problem.
> The kernel module loads without error in both cases.
>
> # uname -r
> 3.1.0-gentoo
>
> # lspci -v
> 03:07.0 Network controller: Atheros Communications Inc. Device 002d (rev 01)
>        Subsystem: Atheros Communications Inc. Device 0300
>        Flags: bus master, 66MHz, medium devsel, latency 176, IRQ 21
>        Memory at fdce (32-bit, non-prefetchable) [size=64K]
>        Capabilities: [44] Power Management version 2
>        Kernel driver in use: ath9k
>        Kernel modules: ath9k
>
> Greetings

Does the wireless card needs a firmware? Do you use an initramfs? I
ask since my iwlagn wireless car does, and if I boot using an
initramfs, I need to include the firmware file on it for the card to
work.

Regards.
-- 
Canek Peláez Valdés
Posgrado en Ciencia e Ingeniería de la Computación
Universidad Nacional Autónoma de México

Re: [gentoo-user] Does this drive need a funeral?

[Dale]

James Broadhead wrote:

On 2 November 2011 01:17, Dale  wrote:

Dale wrote:

Hi,

For the first time in my life, I think I have a drive failing on me.  Here
is the info:

  SNIP

What you folks think?  Can I fix it somehow?  I got a good shovel handy
just in case.

Dale

:-)  :-)


Nov  1 19:08:09 localhost kernel: ata4.01: status: { DRDY }
Nov  1 19:08:14 localhost kernel: ata4: link is slow to respond, please be
patient (ready=0)
Nov  1 19:08:19 localhost kernel: ata4: device not ready (errno=-16),
forcing hardreset
Nov  1 19:08:19 localhost kernel: ata4: soft resetting link

I have RMA'd many drives upon seeing errors like this. I eventually
tracked it down to a faulty SATA cable. :-/




Well, this one does it with two different cables and it is IDE.  I have 
to admit tho, I was hoping it would work fine after I brought it down 
here and put it in my rig where it used to be.  I even checked the 
jumper to make sure it was set right.


I love grasping at straws tho.

Dale

:-)  :-)

Re: [gentoo-user] Does this drive need a funeral?

[James Broadhead]

On 2 November 2011 01:17, Dale  wrote:
> Dale wrote:
>>
>> Hi,
>>
>> For the first time in my life, I think I have a drive failing on me.  Here
>> is the info:
>>
>>  SNIP 
>>
>> What you folks think?  Can I fix it somehow?  I got a good shovel handy
>> just in case.
>>
>> Dale
>>
>> :-)  :-)
>>
> Nov  1 19:08:09 localhost kernel: ata4.01: status: { DRDY }
> Nov  1 19:08:14 localhost kernel: ata4: link is slow to respond, please be
> patient (ready=0)
> Nov  1 19:08:19 localhost kernel: ata4: device not ready (errno=-16),
> forcing hardreset
> Nov  1 19:08:19 localhost kernel: ata4: soft resetting link

I have RMA'd many drives upon seeing errors like this. I eventually
tracked it down to a faulty SATA cable. :-/

Re: [gentoo-user] Does this drive need a funeral?

[Dale]

Lorenzo Bandieri wrote:

OK.  For those that have never seen this before, here is what messages looks
like when a hard drive is going belly up:

  (...)

So, if you see things like this in your log file, rescue data real soon.
  Make a note of this since this could happen on your rig one day.


Thanks, Dale. Actually I've never experienced an hard drive failure,
but your thread made me want to do one as soon as possible :D

(Besides that, "is going belly up" made me burst out laughing)

Lorenzo




Well, I live in the sticks.  I see opossums and especially armadillos 
belly up all the time.  They look like those animal balloons.  lol  
Their little feet stuck up in the air and all.  If you ever run across 
one of those, do NOT poke it with a stick.  It can be bad real fast.


I just wish hard drives were as tough as the armadillos tho.

Dale

:-)  :-)

Re: [gentoo-user] Does this drive need a funeral?

[Dale]

Neil Bothwick wrote:

On Wed, 2 Nov 2011 10:02:03 -0400, Michael Mol wrote:


I think that means you have a nice new door stop.  My first drive to
ever fail that I recall.  I still say windoze screwed up my drive.
  It lasted about 3 days in the windoze rig.

It may also be that the Windows rig's PSU is unhealthy.

Or even the mains supply to that PSU if it is in a different location. I
had a lot of hardware failures after moving house, including several
expensive SCSI drives. They stopped after I bought a UPS.




Well, his rig does start up strange.  You have to push and HOLD the 
power button to get it to stay on.  If you just push it like normal, it 
goes back off again.  Sort of gets on my nerves but I don't know if it 
is a mobo or power supply issue.  Could even be something else too.


Hm.

Dale

:-)  :-)

Re: [gentoo-user] Does this drive need a funeral?

[Pandu Poluan]

On Nov 2, 2011 9:22 PM, "Lorenzo Bandieri" 
wrote:
>
> > Thanks, Dale. Actually I've never experienced an hard drive failure,
> > but your thread made me want to do one as soon as possible :D
>
> Sorry, I meant "made me want to do a backup as soon as possible"
>

I almost thought you're a masochist :-P

Rgds,

Re: [gentoo-user] Does this drive need a funeral?

[Lorenzo Bandieri]

> Thanks, Dale. Actually I've never experienced an hard drive failure,
> but your thread made me want to do one as soon as possible :D

Sorry, I meant "made me want to do a backup as soon as possible"

Re: [gentoo-user] Does this drive need a funeral?

[Lorenzo Bandieri]

> OK.  For those that have never seen this before, here is what messages looks
> like when a hard drive is going belly up:
 (...)
> So, if you see things like this in your log file, rescue data real soon.
>  Make a note of this since this could happen on your rig one day.
>

Thanks, Dale. Actually I've never experienced an hard drive failure,
but your thread made me want to do one as soon as possible :D

(Besides that, "is going belly up" made me burst out laughing)

Lorenzo

Re: [gentoo-user] Does this drive need a funeral?

[Neil Bothwick]

On Wed, 2 Nov 2011 10:02:03 -0400, Michael Mol wrote:

> > I think that means you have a nice new door stop.  My first drive to
> > ever fail that I recall.  I still say windoze screwed up my drive.
> >  It lasted about 3 days in the windoze rig.  
> 
> It may also be that the Windows rig's PSU is unhealthy.

Or even the mains supply to that PSU if it is in a different location. I
had a lot of hardware failures after moving house, including several
expensive SCSI drives. They stopped after I bought a UPS.


-- 
Neil Bothwick

Pedestrians come in two types: Quick or Dead.


signature.asc
Description: PGP signature

Re: [gentoo-user] Consistency checking

[Neil Bothwick]

On Wed, 2 Nov 2011 10:09:13 -0200, Urs Schutz wrote:

> Sometimes I do a 
> > emerge --depclean -p  
> more because portage says that one should do it. I feel
> that it is a lot of work to go through the package list
> manually and do not see the immediate benefit in doing it
> often.

The obvious benefit is that it takes a lot less time to go through the
package list as it is generally shorter when the interval between runs is
shorter.

I have a weekly cron job that runs python-updater, revdep-rebuild,
eix-test-obsolete, emerge -c, emaint -c all, glsa-check etc, all with -p,
and mails me the output. Then I can browse the results at my leisure.


-- 
Neil Bothwick

Ultimate memory manager; Windows, it manages to use it all..


signature.asc
Description: PGP signature

Re: [gentoo-user] Does this drive need a funeral?

[Michael Mol]

On Tue, Nov 1, 2011 at 9:35 PM, Dale  wrote:
> Dale wrote:
> I think that means you have a nice new door stop.  My first drive to ever
> fail that I recall.  I still say windoze screwed up my drive.  It lasted
> about 3 days in the windoze rig.

It may also be that the Windows rig's PSU is unhealthy.

-- 
:wq

Re: [gentoo-user] Consistency checking

[Urs Schutz]

On Wed, 2 Nov 2011 06:01:28 -0700 (PDT)
Leho Kraav  wrote:

> On Wednesday, November 2, 2011 2:20:02 PM UTC+2, Urs
> Schutz wrote:
> > > eix-test-obsolete -d
> > is easy if it is done often, as there is little
> > corrective action to do on a day to day basis.
> 
> I have been wondering if it's possible to get this
> particular check's output to be machine readable, so it
> could be used in automated scripts. Similar to what qlist
> -IC  does.
> 
> First, I think the different checks need to be
> separatable with cmdline arguments, so one wouldn't have
> to start parsing for some header type thing to know what
> atom list applies to what file.
> 
> After that, the machine readable output needs to be as
> simple as possible, no colors, parenthesis, and whatnot.
> 
> Further thoughts welcome, and suggestions for other ways
> of automating this cleanup. Only thing I know of was
> udept, but that's been dead upstream for a while.
> 
> I should also note that I keep my /etc in git, so
> naturally all automated actions would be commited and
> therefore tracked - no need to fear losing some important
> configuration item, if that is at all applicable here
> even.
> 

Something like > eix -tTc --xml could be a start.

Re: [gentoo-user] Consistency checking

[Leho Kraav]

On Wednesday, November 2, 2011 2:20:02 PM UTC+2, Urs Schutz wrote:
> > eix-test-obsolete -d
> is easy if it is done often, as there is little corrective
> action to do on a day to day basis.

I have been wondering if it's possible to get this particular check's output to 
be machine readable, so it could be used in automated scripts. Similar to what 
qlist -IC  does.

First, I think the different checks need to be separatable with cmdline 
arguments, so one wouldn't have to start parsing for some header type thing to 
know what atom list applies to what file.

After that, the machine readable output needs to be as simple as possible, no 
colors, parenthesis, and whatnot.

Further thoughts welcome, and suggestions for other ways of automating this 
cleanup. Only thing I know of was udept, but that's been dead upstream for a 
while.

I should also note that I keep my /etc in git, so naturally all automated 
actions would be commited and therefore tracked - no need to fear losing some 
important configuration item, if that is at all applicable here even.

Re: [gentoo-user] Consistency checking

[Urs Schutz]

On Tue, 1 Nov 2011 19:00:04 +0100
meino.cra...@gmx.de wrote:

> Urs Schutz  [11-11-01 18:04]:
> > On Tue, 1 Nov 2011 04:45:25 +0100
> > meino.cra...@gmx.de wrote:
> > 
> > > Hi,
> > > 
> > > I know of three commands to check the consistency of a
> > > Gentoo system:
> > > 
> > > 
> > > eix-sync && emerge --color=n -p -v --newuse
> > > --update --deep world emerge -p -v --depclean
> > > revdep-rebuild --ignore -p -v
> > > 
> > > of course, one has to remove the -p and -v flags after
> > > checking the putput of the commands.
> > > 
> > > What else can be checked and should be checked from
> > > time to time or after each update?
> > > 
> > > Thank you very much in advance for any help!
> > > 
> > > Best regards,
> > > mcc
> > > 
snip...
<>
snip... 
> 
> Hi,
> 
> thank you all very much for the checks listed.
> 
> One question remains:
> What sequence of commands is adviced to guarantee
> the best effect on the one hand and on the other hand 
> not to be urged to recompile things -- especially 
> those like libreoffice...-- twice and more?
> 
> Thank you very much in advance for your help!
> Best regards,
> mcc
> 
> 


Portage is clever, and does not build libreoffice twice,
even if you change the sequence. Some binary packages are
faster to install. I'm pretty happy with the prebuilt
version of libreoffice. If you stay with stable packets
there is less rebuilding of heavy packages, as they
normally get less updates.

The sequence depends on what use you have. I would do it
differently on a server 1000 km away, but anyway here is
what I do normally on my laptop:

> eix-sync
then learn about the new packages in portage, look for what
gets updated.

Then 
> emerge -uDpvtN world
to see if there is any blockers, licenses or any other
inconsistencies which need to be resolved manually first
(running a mix of stable and unstable packages here, so
this happens from time to time).

Then portage may work in the background, while I do my
usual work on the laptop (all on one line):
> emerge -uDN world && revdep-rebuild -i && cfg-update -u &&
elogv
When the ventilator noise stops and cpu temperatures drop I
check the yellow and red elogv output lines, and if there is
something indicated by the package maintainers I do it right
away.
I simply delete the green elogv output lines.

Sometimes I do a 
> emerge --depclean -p
more because portage says that one should do it. I feel
that it is a lot of work to go through the package list
manually and do not see the immediate benefit in doing it
often. When I remove packages, I always do
> emerge -uDN world && revdep-rebuild -i && elogv
as I don't trust the state of the system after removing
packages.

> eix-test-obsolete -d
is easy if it is done often, as there is little corrective
action to do on a day to day basis.

> eclean-dist -d -p
This is really fast and painless, if you do not need old
package sources.

> glsa-check -t all
should find no problems if you have an up to date
installation.

> eselect news list
and
> eselect news read xx
is always a good thing to do and helps you to prevent the
big problems with a evolving Gentoo installation.

If there is a day without portage updates for your packages
you could do
> emerge -uDN  --complete-graph  --with-bdeps=y world
or
> rkhunter
or
> smartctl --all /dev/sda
or
> freshclam && clamscan
or build a new kernel, just to have some fun...

Sometimes I get a little jealous when I see my nephew
installing programs and updating applications in Ubuntu...

Regards
Urs

Re: [gentoo-user] Unable to login to gentoo

[YoYo Siska]

On Tue, Nov 01, 2011 at 04:54:06PM -0500, Dale wrote:
> Michael Schreckenbauer wrote:
> > Am Dienstag, 1. November 2011, 23:00:57 schrieb Alan McKinnon:
> >> On Tue, 01 Nov 2011 14:49:30 -0500
> >>
> >> Dale  wrote:
> >>> Vishnupradeep wrote:
>  On Tue, Nov 1, 2011 at 2:48 PM, Mick 
>  >  wrote:
>   emerge -1aDv app-portage/gentoolkit
> 
>   revdep-rebuild -v --ask
> 
>  i can't use emerge command as it is shown as command not found. can
>  u tell the correct path of emerge.
> >>> /usr/bin/emerge
> >>>
> >>> Little bit of learning here:
> >>>
> >>> root@smoker / # which emerge
> >>> /usr/bin/emerge
> >>> root@smoker / #
> >>>
> >>> Now you know where it is and how I found out where it is.  Trick is
> >>> remembering a command you rarely use.
> >> That won't work. The $PATH is broken so the shell can't find ls and
> >> emerge.
> >>
> >> Well, it won't find which either :-)
> >>
> >> Solution: You run which which and tell the OP which directory contains
> >> which so he can run /path/towhich emerge to find out where emerge is.
> > Or just use "type -a", because that's a builtin.
> > ~ $ type -a emerge
> > emerge is /usr/bin/emerge
> >
> > Best,
> > Michael
> >
> >
> >
> >
> 
> 
> That is better.  Now to remember that command.  ;-)
> 
> Dale

Except that both won't work in OP's case, because both look for the
executable in directories in PATH which is the OP's problem ;)

yoyo

[gentoo-user] Wireless network card not loaded on first boot after shutdown

[Sebastian Beßler]

Hi,

I have a wireless card that works most of the time perfect with the
ath9k kernel module but not on first boot after a few hours long shutdown.

 * Bringing up interface wlan0
 *   ERROR: interface wlan0 does not exist
 *   Ensure that you have loaded the correct kernel module for your hardware
 * ERROR: net.wlan0 failed to start

A reboot fixes the problem.
The kernel module loads without error in both cases.

# uname -r
3.1.0-gentoo

# lspci -v
03:07.0 Network controller: Atheros Communications Inc. Device 002d (rev 01)
Subsystem: Atheros Communications Inc. Device 0300
Flags: bus master, 66MHz, medium devsel, latency 176, IRQ 21
Memory at fdce (32-bit, non-prefetchable) [size=64K]
Capabilities: [44] Power Management version 2
Kernel driver in use: ath9k
Kernel modules: ath9k

Greetings

Sebastian Beßler



signature.asc
Description: OpenPGP digital signature

Re: [gentoo-user] Unable to login to gentoo

[czernitko]

Well, if nothing else, you could use binfmt support in kernel to execute
64bit code on 32bit OS, as same as you can execute ARM or other arch
binaries. On the other hand, booting 64bit live system would be much
easier, faster and better approach, I guess :)

If you don't want to mess with chroot from Fedora, you can use Gentoo amd64
Live DVD, which can be downloaded from

http://gentoo.osuosl.org//releases/amd64/11.0/

Best regards,
Peter

2011/11/1 Sebastian Beßler 

> Am 01.11.2011 20:51, schrieb Dale:
>
> > Don't boot a 32 bit OS and chroot into a 64 bit one. That can be done
> > but it requires some additional commands.
>
> I always thought that is impossibl because the 32bit libs and kernel are
> unable to execute 64bit code. You could do the other way round with no
> probs but a 64bit cpu in 32bit mode with 32bit libs and kernel?
> I really would like to know more about that.
>
> Greetings
>
> Sebastian Beßler
>
>