[gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Duncan]

Mike Frysinger posted <200505251820.02637.vapier@gentoo.org>, excerpted
below,  on Wed, 25 May 2005 18:20:02 -0400:

> can people please try out baselayout-1.11.12-r2+ and see if they notice any 
> regressions ?  the 'best' tests are simply rebooting and seeing if your 
> system comes up :)

I've been running baselayout-1.11.12+ since it went ~amd64.  It has been
very smooth going, far more so than previous versions, where net,
specifically macchanger, wasn't working right without module script
tweaking (that was bugged and eventually fixed).

Hmm..  I /did/ have issues with parallel startup a couple revisions back. 
I had disabled that and haven't reenabled it yet.  I suppose I should do
that and bug it now, if it still fails.  It /was/ working at one point.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman in
http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html


-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Roy Marples]

[-- Attachment #1: Type: text/plain, Size: 784 bytes --]

On Thu, 2005-05-26 at 02:08 -0700, Duncan wrote:
> Hmm..  I /did/ have issues with parallel startup a couple revisions back. 
> I had disabled that and haven't reenabled it yet.  I suppose I should do
> that and bug it now, if it still fails.  It /was/ working at one point.

There may still be issues with parallel startup. It doesn't really
startup in parallel and it's not much faster, so if any bugs do come up
then we'll probably recommend turning it off.

baselayout-1.12.0-alpha3 will have much improved parallel startup thanks
to Paul Pacheco patch (bug #69854) which is essentially a re-write of
the code so we're not that interested in fixing parallel startup in
baselayout-1.11.x at this time.

-- 
Roy Marples <uberlord@gentoo.org>
Gentoo Linux Developer

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Thomas Kirchner]

[-- Attachment #1: Type: text/plain, Size: 409 bytes --]

* On May 25 18:20, Mike Frysinger (gentoo-dev@lists.gentoo.org) wrote:
> so can people please try out baselayout-1.11.12-r2+ and see if they 
> notice any regressions ?  the 'best' tests are simply rebooting and 
> seeing if your system comes up :)

Working great here on ~x86, aside from the initial oops with 1.11.12-r0 
causing X to bomb from the /dev issue.  The other changes have been nice.
Tom

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Marek Więcek]

Mike Frysinger wrote:

> so can people please try out baselayout-1.11.12-r2+ and see if they notice any 
> regressions ?  the 'best' tests are simply rebooting and seeing if your 
> system comes up :)

I have just migrated to baselayout-1.11.12-r4 and everything went
smoothly. Arch x86 with some ~x86 application.

-- 
Marek Więcek

-- 
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Konstantin Kletschke]

I have an ~amd64 installation and while booting the  lvm2  vgs  are  not
activated  anymore,  so  booting  stops  there.  After  logging  in  for
maintenance (root is not on lvm) I type "vgchange -a y" press Ctrl-D and
ehre we go.  The  string  "vgchange"  occures  in  no  file  located  in
/etc/init.d/*. Shouldn't that be the case?

I reported it as a bug  in  bugzilla I have to look up the bugnumber
if interest exists. This happened with  baselayout-1.11.11-r1 and
baselayout-1.11.11-r3 IIRC, may be the error still exists.
Is the new  baselayout-1.11.12-X in my portage tree already? I did not
look up yet. If yes I will test that :)

Regards, Konsti


-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Aron Griffis]

[-- Attachment #1: Type: text/plain, Size: 205 bytes --]

Konsti,

See Mike's message which started this thread.  He addressed lvm2
specifically:

    http://article.gmane.org/gmane.linux.gentoo.devel/28317

Regards,
Aron

--
Aron Griffis
Gentoo Linux Developer


[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Konstantin Kletschke]

Am 2005-05-30 10:44 -0400 schrieb Aron Griffis:

> See Mike's message which started this thread.  He addressed lvm2
> specifically:
> 
>     http://article.gmane.org/gmane.linux.gentoo.devel/28317

You mean I should go on the nerves of the LVM2 guys with that :) ?
Thats OK if the lvm2 specific stuff goes into lvm2 packages...

Regards, Konsti

-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Mike Frysinger]

On Monday 30 May 2005 04:00 pm, Konstantin Kletschke wrote:
> Am 2005-05-30 10:44 -0400 schrieb Aron Griffis:
> > See Mike's message which started this thread.  He addressed lvm2
> > specifically:
> >
> >     http://article.gmane.org/gmane.linux.gentoo.devel/28317
>
> You mean I should go on the nerves of the LVM2 guys with that :) ?
> Thats OK if the lvm2 specific stuff goes into lvm2 packages...

try doing `emerge sync` and then `emerge lvm2` ... you should have a script 
at /lib/rcscripts/addons/lvm2-start.sh

i'll revbump lvm2 before stabilizing baselayout to make sure lvm guys get this 
file
-mike
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Konstantin Kletschke]

Am 2005-05-30 19:28 -0400 schrieb Mike Frysinger:

> try doing `emerge sync` and then `emerge lvm2` ... you should have a script 
> at /lib/rcscripts/addons/lvm2-start.sh

Well, the file is there. It is actually named
/lib/rcscripts/addons/lvm-start.sh though.

This particular error occured on my machine with
baselayout-1.11.12-r2 and therefore I file a bugreport against
baselayout at bugs.gentoo.org #94120.

After that I found two threads here regarding new baselayout versions
and changes in lvm2 handling so Ishowed up here with the issue.

May be some other thing is puzzled here (2005.0 multilib on ~amd64)
but I did not found out yet where to search further.

Regards, Konstantin

-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Mike Frysinger]

On Tuesday 31 May 2005 02:53 am, Konstantin Kletschke wrote:
> This particular error occured on my machine with
> baselayout-1.11.12-r2 and therefore I file a bugreport against
> baselayout at bugs.gentoo.org #94120.

are you using lvm2 as your root filesystem ?
-mike
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Konstantin Kletschke]

Am 2005-05-31 08:20 -0400 schrieb Mike Frysinger:

> > baselayout at bugs.gentoo.org #94120.
> 
> are you using lvm2 as your root filesystem ?

No I only have /home, /usr and subdirs under /usr as lvm2 volumes, not
root. Sorry, yes, I forgot to mention...

Konsti

-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Mike Frysinger]

On Tuesday 31 May 2005 08:45 am, Konstantin Kletschke wrote:
> Am 2005-05-31 08:20 -0400 schrieb Mike Frysinger:
> > > baselayout at bugs.gentoo.org #94120.
> >
> > are you using lvm2 as your root filesystem ?
>
> No I only have /home, /usr and subdirs under /usr as lvm2 volumes, not
> root. Sorry, yes, I forgot to mention...

did you properly `etc-update` then ?

your /etc/conf.d/rc must have RC_VOLUME_ORDER set to at least 'lvm'

i assume at boot you never see a message like 'Setting up the Logical Volume 
Manager' before the 'Checking all filesystems' message ?
-mike
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Konstantin Kletschke]

Am 2005-05-31 09:18 -0400 schrieb Mike Frysinger:

> did you properly `etc-update` then ?

Ooops!

> your /etc/conf.d/rc must have RC_VOLUME_ORDER set to at least 'lvm'

ARGH!! I apologize! I always watch the diff output while etc-updating
exactly, even on baselayout or something else importand stuff, but this
time RC_VOLUME_ORDER did not made it into my new /etc/conf.d/rc!

> i assume at boot you never see a message like 'Setting up the Logical Volume 
> Manager' before the 'Checking all filesystems' message ?

No, was completely missing :)


Thanks for your help and tips!

Kind Regards, Konsti

-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Mike Frysinger]

On Tuesday 31 May 2005 06:49 pm, Konstantin Kletschke wrote:
> Am 2005-05-31 09:18 -0400 schrieb Mike Frysinger:
> > did you properly `etc-update` then ?
>
> Ooops!

someone (i think johnm) mentioned this to me before ... i'm going to add a 
small patch so that if RC_VOLUME_ORDER is unset, it'll default to "raid evms 
lvm dm" ... i'll drop it with 1.11.13 or 1.11.14

that way people who are bad and dont etc-update dont end up with bjorked 
systems
-mike
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Konstantin Kletschke]

Am 2005-05-31 18:57 -0400 schrieb Mike Frysinger:

> small patch so that if RC_VOLUME_ORDER is unset, it'll default to "raid evms 
> lvm dm" ... i'll drop it with 1.11.13 or 1.11.14

Yes and may be add a boot warning meanwhile that the RC_VOLUME_ORDER is
completely unset. I for example would have seen that :)

Konsti

-- 
GPG KeyID EF62FCEF
Fingerprint: 13C9 B16B 9844 EC15 CC2E  A080 1E69 3FDA EF62 FCEF
-- 
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: baselayout-1.11.12-r2 request for testers

[Kevin Bryan]

Mike Frysinger <vapier <at> gentoo.org> writes:
> 
> yes, it's finally that time ... after months of hearing us say 'we want to get 
> new baselayout stable asap', we're serious
> 
> so can people please try out baselayout-1.11.12-r2+ and see if they notice any 
> regressions ?  the 'best' tests are simply rebooting and seeing if your 
> system comes up :)
> 


I've been running the 1.11.x series since .8 with no problems (except the
parallel thing that was mentioned before).  Good work!  

I just wanted to share some of the coolness of the new scripts:

Let's say you have a wireless network that uses BlueSocket login.  It's rather
annoying to have to load a web browser before you can start using the
connection.  I wrote a little perl script to submit the form and call it from
postup().  (Email me if you want the script.)

A second bit of logic I added to postup() was to merge the settings from my
wireless card into the bonding interface.  This works as follows: 

    local ESSID=$( wireless_get_essid eth1 )
    if [ "${IFACE}" == "bond0" ] ; then
        echo "ESSID: $ESSID" 
        if [ -z "$(eval echo \${config_${ESSID}})" ] || 
           [ "$(eval echo \${config_${ESSID}})" == "dhcp" ] ; then
            dhcpcd -t 5 bond0
        else
            echo $(eval echo "\${config_${ESSID}}")
            ifconfig ${IFACE} $(eval echo "\${config_${ESSID}}")
            (route -n | grep -q "^0.0.0.0") && route del default
            route add $(eval echo "\${routes_${ESSID}}")
        fi
    fi

Note that I also have these lines for the bond0 interface:
slaves_bond0="eth0 eth1"
config_bond0=( "null" ) # I'll configure this later

The astute reader will notice that I got a little lazy and didn't use the
dhcpcd_xxx variables for that condition, but that wouldn't be too hard to hack
in there.

The end result of this is I can seemlessly go back and forth between the wired
and wireless worlds without losing network connections.  It's great to be on
wireless, start a download, plug into the wired, and then get an immediate speed
boost!

Here's another hint for wireless users:  Test your card's ability to respond to
iwconfig <card> txpower <val> commands.  Then put whatever the max value is into
the /etc/conf.d/wireless preassociate() function.  This should have a noticable
effect on your ability to connect to networks.

Thanks for making this all this possible!

--Kevin



-- 
gentoo-dev@gentoo.org mailing list