Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[covici]

Alan McKinnon alan.mckin...@gmail.com wrote:

 On 04/08/2015 12:14, cov...@ccs.covici.com wrote:
  Alan McKinnon alan.mckin...@gmail.com wrote:
  
  On 04/08/2015 11:13, cov...@ccs.covici.com wrote:
  Hi.  After upgrading to perl 5.22.0 which was a pain -- had to remove
  all those virtuals -- I am now finding that a small perl program  which
  connects to a mysql database yields the following when it exits:
 
  *** Error in `perl': double free or corruption (!prev): 
  0x01ed8610 ***
  === Backtrace: =
  /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
  /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
  /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(+0x1234d)[0x7f59bb21034d]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI.so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
  /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
  /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]
  /usr/lib64/libperl.so.5.22(+0xda25f)[0x7f59bc81b25f]
  /usr/lib64/libperl.so.5.22(Perl_sv_clear+0x740)[0x7f59bc81bc10]
  /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
  /usr/lib64/libperl.so.5.22(+0xb7298)[0x7f59bc7f8298]
  /usr/lib64/libperl.so.5.22(Perl_mg_free+0x2e)[0x7f59bc7f8a4e]
  /usr/lib64/libperl.so.5.22(Perl_sv_clear+0xae)[0x7f59bc81b57e]
  /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
  /usr/lib64/libperl.so.5.22(Perl_leave_scope+0xd51)[0x7f59bc84a411]
  /usr/lib64/libperl.so.5.22(+0x52c56)[0x7f59bc793c56]
  /usr/lib64/libperl.so.5.22(Perl_my_exit+0x3f)[0x7f59bc798d0f]
  /usr/lib64/libperl.so.5.22(Perl_pp_exit+0x4a)[0x7f59bc857a6a]
  /usr/lib64/libperl.so.5.22(Perl_runops_standard+0x16)[0x7f59bc80f316]
  /usr/lib64/libperl.so.5.22(perl_run+0x2f9)[0x7f59bc79c369]
  perl(main+0x149)[0x400e39]
  /lib64/libc.so.6(__libc_start_main+0xf0)[0x7f59bc3c67b0]
  perl(_start+0x29)[0x400e79]
  === Memory map: 
 
  memory map  ommitted, if it is needed I can reproduce.
 
  So, should I downgrade -- means removing all those virtuals again -- or
  any other ideas would be appreciated.
 
  I  am running the unstable gentoo, if you need more information, I can
  include it.
 
  Thanks in advance for any suggestions.
 
 
 
  did you run emerge @preserved-rebuild, revdep-rebuild and perl-cleaner
  after the upgrade? That stuff's easy to forget.
 
 
  
  I ran both perl-cleaner --reallyall and emerge @preserved-rebuild which
  only rebuilt some haskall stuff.Perl cleaner had some problems, it
  tried to rebuild some python  packages which at the time (before I did
  the complete update_) had some problems because of the 3.3 to 3.4
  change.  So I just  took the emerge line from the perl cleaner, ommitted
  any  package which was a hard blocker andran the rest -- about 186
  packages.  I can try to rerun perl-cleaner if you think that would help
  any, since I now have updated the world.
  
 
 I think running perl-cleaner --all as step 1 is wise.
 It shouldn't need to make any changes, but let's cover all the usual
 bases first, paying particular attention to any DBD/DBI stuff it might
 find that were not installed by portage.
 
 I am also curious why you have blockers and had to fiddle with virtuals
 - the same upgrade here was clean and portage just automatically did
 everything it needed. Do you have any package.* entry for perl stuff?
 
 grep -ir perl /etc/portage

Well, in the past portage did fix all the virtuals, but not this time
and I saw a post on gentoo forums which suggested I unmerge all of them
and then do the update.  But I just did perl first and then ran
perl-cleaner --reallyall and then did the update.  I do have some use
flags with perl i.e.

/etc/portage/package.use:dev-lang/perl  ithreads
/etc/portage/package.use:sys-devel/libperl ithreads
/etc/portage/package.use:www-apache/mod-perl threads
/etc/portage/package.use:net-nntp/inn perl
/etc/portage/package.use:dev-db/postgresql-server perl python

and an extra_econf for innd  which ran OK.



-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

 John Covici
 cov...@ccs.covici.com

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[Andreas K. Huettel]

Am Dienstag, 4. August 2015, 11:13:56 schrieb cov...@ccs.covici.com:
 
 *** Error in `perl': double free or corruption (!prev): 0x01ed8610
 *** === Backtrace: =
 /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
 /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
 /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysq
 l/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mys
 ql/mysql.so(+0x1234d)[0x7f59bb21034d]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI
 .so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
 /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
 /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]

Looks like a bug in dev-perl/DBD-mysql ... you could check on CPAN if anyone 
has already filed a bug for DBD::mysql, if there's known breakage with 5.22 or 
if it's already fixed in a newer version...

-- 
Andreas K. Huettel
Gentoo Linux developer (council, perl, libreoffice)
dilfri...@gentoo.org
http://www.akhuettel.de/

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[covici]

Andreas K. Huettel dilfri...@gentoo.org wrote:

 Am Dienstag, 4. August 2015, 11:13:56 schrieb cov...@ccs.covici.com:
  
  *** Error in `perl': double free or corruption (!prev): 0x01ed8610
  *** === Backtrace: =
  /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
  /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
  /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysq
  l/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mys
  ql/mysql.so(+0x1234d)[0x7f59bb21034d]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI
  .so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
  /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
  /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]
 
 Looks like a bug in dev-perl/DBD-mysql ... you could check on CPAN if anyone 
 has already filed a bug for DBD::mysql, if there's known breakage with 5.22 
 or 
 if it's already fixed in a newer version...
 
 -- 
 Andreas K. Huettel
 Gentoo Linux developer (council, perl, libreoffice)
 dilfri...@gentoo.org
 http://www.akhuettel.de/

Well, you gave me a hint and re-emerging that package has fixed that  --
thanks so much.  I am not sure why, but go figure.

Thanks again.

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

 John Covici
 cov...@ccs.covici.com

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[Alan McKinnon]

On 04/08/2015 13:15, cov...@ccs.covici.com wrote:
 Andreas K. Huettel dilfri...@gentoo.org wrote:
 
 Am Dienstag, 4. August 2015, 11:13:56 schrieb cov...@ccs.covici.com:

 *** Error in `perl': double free or corruption (!prev): 0x01ed8610
 *** === Backtrace: =
 /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
 /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
 /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysq
 l/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mys
 ql/mysql.so(+0x1234d)[0x7f59bb21034d]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI
 .so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
 /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
 /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]

 Looks like a bug in dev-perl/DBD-mysql ... you could check on CPAN if anyone 
 has already filed a bug for DBD::mysql, if there's known breakage with 5.22 
 or 
 if it's already fixed in a newer version...

 -- 
 Andreas K. Huettel
 Gentoo Linux developer (council, perl, libreoffice)
 dilfri...@gentoo.org
 http://www.akhuettel.de/
 
 Well, you gave me a hint and re-emerging that package has fixed that  --
 thanks so much.  I am not sure why, but go figure.


Pick one:

[] Cosmic Rays!
[] Random quantum-level bit flipping!
[] Slight imperfection in cannot-be-perfect disc surface!
[] Random shit in the style of Discworld!
[] Your $DEITY is messing with your head to test your faith!
[] Shit happens sometimes


:-)


-- 
Alan McKinnon
alan.mckin...@gmail.com

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[wraeth]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 04/08/15 21:19, Alan McKinnon wrote:
 
 Pick one:
 
 [] Cosmic Rays! [] Random quantum-level bit flipping! [] Slight
 imperfection in cannot-be-perfect disc surface! [] Random shit in
 the style of Discworld! [] Your $DEITY is messing with your head to
 test your faith! [] Shit happens sometimes

All of the above? :D

- -- 
wraeth wra...@wraeth.id.au
GnuPG Key: B2D9F759
-BEGIN PGP SIGNATURE-
Version: GnuPG v2

iF4EAREIAAYFAlXAoFAACgkQXcRKerLZ91n4cQD/XU9pC37zefUJymeNmT3LpatO
J55Xgl4ra6GM50uFA+kA/37jTKy4UpBxIVn6wyQB4RBoCRW4+7U+IZ5WIillIRYp
=duG0
-END PGP SIGNATURE-

[ SOLVED ] Re: [gentoo-user] smartd and ssmtp not sending emails again.

[Dale]

John Campbell wrote:
 On 08/04/2015 01:35 AM, Dale wrote:
 Howdy,

 I went through this before and I haven't changed anything so not sure
 why it isn't working now.  I decided to test smartd to make sure that it
 would send me a email if a drive developed a problem.  I added -M test
 to the line, restarted smartd and got nothing.  I mean nothing but
 crickets.  Going to post relevant info that I know of.  Here we go:

 Uncommented lines from smartd.conf:


 DEVICESCAN -I 194 -I 231 -I 9

 /dev/sda -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
 test
 /dev/sdb -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
 test
 /dev/sdc -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
 test

 Everything in smartd.conf after the DEVICESCAN is ignored.

 Unless you've stripped out all the comments from smartd.conf it should
 be explained a half dozen lines above your DEVICESCAN line.





Yeppie  That worked.  It did upgrade recently and I had to re-add
the other drive lines back but I don't recall changing that part.  The
upgrade is the reason I tested it.  I wanted to make sure it still
worked, before a drive dies and I'm caught off guard, again.  I figured
it would be something simple that I just didn't notice. 

Thanks much.  :-D

Dale

:-)  :-) 

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[covici]

Alan McKinnon alan.mckin...@gmail.com wrote:

 On 04/08/2015 11:13, cov...@ccs.covici.com wrote:
  Hi.  After upgrading to perl 5.22.0 which was a pain -- had to remove
  all those virtuals -- I am now finding that a small perl program  which
  connects to a mysql database yields the following when it exits:
  
  *** Error in `perl': double free or corruption (!prev): 0x01ed8610 
  ***
  === Backtrace: =
  /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
  /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
  /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(+0x1234d)[0x7f59bb21034d]
  /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI.so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
  /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
  /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]
  /usr/lib64/libperl.so.5.22(+0xda25f)[0x7f59bc81b25f]
  /usr/lib64/libperl.so.5.22(Perl_sv_clear+0x740)[0x7f59bc81bc10]
  /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
  /usr/lib64/libperl.so.5.22(+0xb7298)[0x7f59bc7f8298]
  /usr/lib64/libperl.so.5.22(Perl_mg_free+0x2e)[0x7f59bc7f8a4e]
  /usr/lib64/libperl.so.5.22(Perl_sv_clear+0xae)[0x7f59bc81b57e]
  /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
  /usr/lib64/libperl.so.5.22(Perl_leave_scope+0xd51)[0x7f59bc84a411]
  /usr/lib64/libperl.so.5.22(+0x52c56)[0x7f59bc793c56]
  /usr/lib64/libperl.so.5.22(Perl_my_exit+0x3f)[0x7f59bc798d0f]
  /usr/lib64/libperl.so.5.22(Perl_pp_exit+0x4a)[0x7f59bc857a6a]
  /usr/lib64/libperl.so.5.22(Perl_runops_standard+0x16)[0x7f59bc80f316]
  /usr/lib64/libperl.so.5.22(perl_run+0x2f9)[0x7f59bc79c369]
  perl(main+0x149)[0x400e39]
  /lib64/libc.so.6(__libc_start_main+0xf0)[0x7f59bc3c67b0]
  perl(_start+0x29)[0x400e79]
  === Memory map: 
  
  memory map  ommitted, if it is needed I can reproduce.
  
  So, should I downgrade -- means removing all those virtuals again -- or
  any other ideas would be appreciated.
  
  I  am running the unstable gentoo, if you need more information, I can
  include it.
  
  Thanks in advance for any suggestions.
  
 
 
 did you run emerge @preserved-rebuild, revdep-rebuild and perl-cleaner
 after the upgrade? That stuff's easy to forget.
 
 

I ran both perl-cleaner --reallyall and emerge @preserved-rebuild which
only rebuilt some haskall stuff.Perl cleaner had some problems, it
tried to rebuild some python  packages which at the time (before I did
the complete update_) had some problems because of the 3.3 to 3.4
change.  So I just  took the emerge line from the perl cleaner, ommitted
any  package which was a hard blocker andran the rest -- about 186
packages.  I can try to rerun perl-cleaner if you think that would help
any, since I now have updated the world.

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

 John Covici
 cov...@ccs.covici.com

Re: [gentoo-user] Re: systemd-224 Look out for new networking behavior [FIXED]

[Franz Fellner]

Fernando Rodriguez wrote:
 On Monday, August 03, 2015 6:41:22 PM walt wrote:
  That line declares *hostname as a constant and then the statement below
  proceeds to assign a value to the 'constant'.  I wonder how many hours
  of frustration have been suffered by student programmers while trying to
  understand the logic behind that.
 
 Because it's not a constant, it's a pointer-to-constant :)
Both of you are right, you can read the declaration in both ways:
hostname is of type pointer to const char.
*hostname is of type const char.

But in this case it is not *hostname, that get's a value assigned, it's 
simply hostname. If you do not set hostname to NULL it stays uninitialised, 
which means its value is what the actual memory is set to - quite undefined.
Correct initialization is really important and should be done consequently so 
it gets an automatism ;) (would avoid issues like this)

 
 const char *hostname; /* pointer to constant char */
 char *const hostname; /* constant pointer to char */
 const char *const hostname; /* constant pointer to constant char */
 
 Is that confusing enough?
 
 -- 
 Fernando Rodriguez
 

[gentoo-user] smartd and ssmtp not sending emails again.

[Dale]

Howdy,

I went through this before and I haven't changed anything so not sure
why it isn't working now.  I decided to test smartd to make sure that it
would send me a email if a drive developed a problem.  I added -M test
to the line, restarted smartd and got nothing.  I mean nothing but
crickets.  Going to post relevant info that I know of.  Here we go:

Uncommented lines from smartd.conf: 


DEVICESCAN -I 194 -I 231 -I 9

/dev/sda -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
test
/dev/sdb -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
test
/dev/sdc -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
test



Other than adding -M test, I don't recall changing anything except the
usual upgrade changes.  From revaliases, with obvious edits of course:


root:me-em...@gmail.com:smtp.gmail.com:465
dale:me-em...@gmail.com:smtp.gmail.com:465
other_user:me-em...@gmail.com:smtp.gmail.com:465 


From ssmtp.conf file, again with obvious edits:


root=postmaster
root=me-em...@gmail.com  #Change to your preferred email address
mailhub=smtp.gmail.com:465  #Could also use port 587 for STARTTLS
rewriteDomain=fireball  #Something to denote your machine's name
FromLineOverride=YES
UseTLS=YES  #Can also try UseSTARTTLS=YES as an alternative
AuthUser=me-em...@gmail.com
AuthPass=secret-stuff #Special characters seem to barf with ssmtp
mailhub=mail
hostname=_HOSTNAME_



From my understanding, smartd passes the info to ssmtp and then it gets
sent from there.  When I test ssmtp, I get this, with obvious edits of
course. 


root@fireball / # echo My first test message | mail -v -s Test for
sSMTP 1 rdalek1...@gmail.com
[-] 220 smtp.gmail.com ESMTP u78sm263385ywu.13 - gsmtp
[-] EHLO _HOSTNAME_
[-] 250 SMTPUTF8
[-] AUTH LOGIN
[-] 334 VXNlcm5hbWU6
[-] cmRhbGVrMTk2N0BnbWFpbC5jb20=
[-] 334 UGFzc3dvcmQ6
[-] 235 2.7.0 Accepted
[-] MAIL FROM:me-em...@gmail.com
[-] 250 2.1.0 OK u78sm263385ywu.13 - gsmtp
[-] RCPT TO:me-em...@gmail.com
[-] 250 2.1.5 OK u78sm263385ywu.13 - gsmtp
[-] DATA
[-] 354  Go ahead u78sm263385ywu.13 - gsmtp
[-] Received: by _HOSTNAME_ (sSMTP sendmail emulation); Tue, 04 Aug
2015 03:28:36 -0500
[-] From: root me-em...@gmail.com
[-] Date: Tue, 04 Aug 2015 03:28:36 -0500
[-] To: me-em...@gmail.com
[-] Subject: Test for sSMTP 1
[-]
[-] My first test message
[-] .
[-] 250 2.0.0 OK 1438676918 u78sm263385ywu.13 - gsmtp
[-] QUIT
[-] 221 2.0.0 closing connection u78sm263385ywu.13 - gsmtp
root@fireball / #


When I check my email, I do get the test message.  It seems to me that
ssmtp is working but that smartd either isn't working or it is not
getting to ssmtp somehow. 

Anyone see a reason this shouldn't be working?  I'd really like a heads
up if a drive is having issues. 

Thanks.

Dale

:-)  :-) 

Re: [gentoo-user] smartd and ssmtp not sending emails again.

[Raffaele BELARDI]

Dale wrote:
 I went through this before and I haven't changed anything so not sure
 why it isn't working now.  I decided to test smartd to make sure that it
 would send me a email if a drive developed a problem.  I added -M test
 to the line, restarted smartd and got nothing.  I mean nothing but
 crickets.  Going to post relevant info that I know of.  Here we go:

(snip)

 When I check my email, I do get the test message.  It seems to me that
 ssmtp is working but that smartd either isn't working or it is not
 getting to ssmtp somehow.

 Anyone see a reason this shouldn't be working?  I'd really like a heads
 up if a drive is having issues.


Any hints from the syslog? Here -M test seems to work fine.

raffaele

Re: [gentoo-user] smartd and ssmtp not sending emails again.

[John Campbell]

On 08/04/2015 01:35 AM, Dale wrote:

Howdy,

I went through this before and I haven't changed anything so not sure
why it isn't working now.  I decided to test smartd to make sure that it
would send me a email if a drive developed a problem.  I added -M test
to the line, restarted smartd and got nothing.  I mean nothing but
crickets.  Going to post relevant info that I know of.  Here we go:

Uncommented lines from smartd.conf:


DEVICESCAN -I 194 -I 231 -I 9

/dev/sda -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
test
/dev/sdb -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
test
/dev/sdc -a -d sat -o on -S on -s (S/../.././12|L/../../6/12) -m root -M
test


Everything in smartd.conf after the DEVICESCAN is ignored.

Unless you've stripped out all the comments from smartd.conf it should 
be explained a half dozen lines above your DEVICESCAN line.

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[Alan McKinnon]

On 04/08/2015 11:13, cov...@ccs.covici.com wrote:
 Hi.  After upgrading to perl 5.22.0 which was a pain -- had to remove
 all those virtuals -- I am now finding that a small perl program  which
 connects to a mysql database yields the following when it exits:
 
 *** Error in `perl': double free or corruption (!prev): 0x01ed8610 ***
 === Backtrace: =
 /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
 /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
 /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(+0x1234d)[0x7f59bb21034d]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI.so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
 /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
 /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]
 /usr/lib64/libperl.so.5.22(+0xda25f)[0x7f59bc81b25f]
 /usr/lib64/libperl.so.5.22(Perl_sv_clear+0x740)[0x7f59bc81bc10]
 /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
 /usr/lib64/libperl.so.5.22(+0xb7298)[0x7f59bc7f8298]
 /usr/lib64/libperl.so.5.22(Perl_mg_free+0x2e)[0x7f59bc7f8a4e]
 /usr/lib64/libperl.so.5.22(Perl_sv_clear+0xae)[0x7f59bc81b57e]
 /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
 /usr/lib64/libperl.so.5.22(Perl_leave_scope+0xd51)[0x7f59bc84a411]
 /usr/lib64/libperl.so.5.22(+0x52c56)[0x7f59bc793c56]
 /usr/lib64/libperl.so.5.22(Perl_my_exit+0x3f)[0x7f59bc798d0f]
 /usr/lib64/libperl.so.5.22(Perl_pp_exit+0x4a)[0x7f59bc857a6a]
 /usr/lib64/libperl.so.5.22(Perl_runops_standard+0x16)[0x7f59bc80f316]
 /usr/lib64/libperl.so.5.22(perl_run+0x2f9)[0x7f59bc79c369]
 perl(main+0x149)[0x400e39]
 /lib64/libc.so.6(__libc_start_main+0xf0)[0x7f59bc3c67b0]
 perl(_start+0x29)[0x400e79]
 === Memory map: 
 
 memory map  ommitted, if it is needed I can reproduce.
 
 So, should I downgrade -- means removing all those virtuals again -- or
 any other ideas would be appreciated.
 
 I  am running the unstable gentoo, if you need more information, I can
 include it.
 
 Thanks in advance for any suggestions.
 


did you run emerge @preserved-rebuild, revdep-rebuild and perl-cleaner
after the upgrade? That stuff's easy to forget.



-- 
Alan McKinnon
alan.mckin...@gmail.com

Re: [gentoo-user] smartd and ssmtp not sending emails again.

[Dale]

Raffaele BELARDI wrote:
 Dale wrote:
 I went through this before and I haven't changed anything so not sure
 why it isn't working now.  I decided to test smartd to make sure that it
 would send me a email if a drive developed a problem.  I added -M test
 to the line, restarted smartd and got nothing.  I mean nothing but
 crickets.  Going to post relevant info that I know of.  Here we go:
 (snip)

 When I check my email, I do get the test message.  It seems to me that
 ssmtp is working but that smartd either isn't working or it is not
 getting to ssmtp somehow.

 Anyone see a reason this shouldn't be working?  I'd really like a heads
 up if a drive is having issues.

 Any hints from the syslog? Here -M test seems to work fine.

 raffaele


Sorry.  I forgot to include that.  This is from when I restart smartd.



Aug  4 04:46:45 fireball smartd[3610]: smartd received signal 15: Terminated
Aug  4 04:46:45 fireball smartd[3610]: smartd is exiting (exit status 0)
Aug  4 04:46:45 fireball smartd[7526]: smartd 6.4 2015-06-04 r4109
[x86_64-linux-3.18.7-gentoo] (local build)
Aug  4 04:46:45 fireball smartd[7526]: Copyright (C) 2002-15, Bruce
Allen, Christian Franke, www.smartmontools.org
Aug  4 04:46:45 fireball smartd[7526]: Opened configuration file
/etc/smartd.conf
Aug  4 04:46:45 fireball smartd[7526]: Drive: DEVICESCAN, implied '-a'
Directive on line 27 of file /etc/smartd.conf
Aug  4 04:46:45 fireball smartd[7526]: Configuration file
/etc/smartd.conf was parsed, found DEVICESCAN, scanning devices
Aug  4 04:46:45 fireball smartd[7526]: Device: /dev/sda, type changed
from 'scsi' to 'sat'
Aug  4 04:46:45 fireball smartd[7526]: Device: /dev/sda [SAT], opened
Aug  4 04:46:45 fireball smartd[7526]: Device: /dev/sda [SAT], WDC
WD1600AAJS-00YZCA0, S/N:WD-WCAYU4389620, WWN:5-0014ee-1ad79c1b3,
FW:01.03B01, 160 GB
Aug  4 04:46:45 fireball smartd[7526]: Device: /dev/sda [SAT], found in
smartd database: Western Digital Caviar Blue Serial ATA
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sda [SAT], is SMART
capable. Adding to monitor list.
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdb, type changed
from 'scsi' to 'sat'
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdb [SAT], opened
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdb [SAT], SAMSUNG
HD753LJ, S/N:S1PWJ1KS305193, WWN:5-0024e9-001308c31, FW:1AA01117, 750 GB
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdb [SAT], found in
smartd database: SAMSUNG SpinPoint F1 DT
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdb [SAT], is SMART
capable. Adding to monitor list.
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdc, type changed
from 'scsi' to 'sat'
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdc [SAT], opened
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdc [SAT],
ST3000DM001-1CH166, S/N:W1F4C31Q, WWN:5-000c50-06e7a0a7c, FW:CC29, 3.00 TB
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdc [SAT], found in
smartd database: Seagate Barracuda 7200.14 (AF)
Aug  4 04:46:46 fireball smartd[7526]: Device: /dev/sdc [SAT], is SMART
capable. Adding to monitor list.
Aug  4 04:46:46 fireball smartd[7526]: Monitoring 3 ATA and 0 SCSI devices
Aug  4 04:46:46 fireball smartd[7528]: smartd has fork()ed into
background mode. New PID=7528.
Aug  4 04:46:46 fireball smartd[7528]: file /run/smartd.pid written
containing PID 7528
root@fireball / #


It seems to start and begin monitoring but I don't see anything about it
even trying to email anything. 

Thanks.

Dale

:-)  :-)

Re: [gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[Alan McKinnon]

On 04/08/2015 12:14, cov...@ccs.covici.com wrote:
 Alan McKinnon alan.mckin...@gmail.com wrote:
 
 On 04/08/2015 11:13, cov...@ccs.covici.com wrote:
 Hi.  After upgrading to perl 5.22.0 which was a pain -- had to remove
 all those virtuals -- I am now finding that a small perl program  which
 connects to a mysql database yields the following when it exits:

 *** Error in `perl': double free or corruption (!prev): 0x01ed8610 
 ***
 === Backtrace: =
 /lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
 /lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
 /lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(+0x1234d)[0x7f59bb21034d]
 /usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI.so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
 /usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
 /usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]
 /usr/lib64/libperl.so.5.22(+0xda25f)[0x7f59bc81b25f]
 /usr/lib64/libperl.so.5.22(Perl_sv_clear+0x740)[0x7f59bc81bc10]
 /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
 /usr/lib64/libperl.so.5.22(+0xb7298)[0x7f59bc7f8298]
 /usr/lib64/libperl.so.5.22(Perl_mg_free+0x2e)[0x7f59bc7f8a4e]
 /usr/lib64/libperl.so.5.22(Perl_sv_clear+0xae)[0x7f59bc81b57e]
 /usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
 /usr/lib64/libperl.so.5.22(Perl_leave_scope+0xd51)[0x7f59bc84a411]
 /usr/lib64/libperl.so.5.22(+0x52c56)[0x7f59bc793c56]
 /usr/lib64/libperl.so.5.22(Perl_my_exit+0x3f)[0x7f59bc798d0f]
 /usr/lib64/libperl.so.5.22(Perl_pp_exit+0x4a)[0x7f59bc857a6a]
 /usr/lib64/libperl.so.5.22(Perl_runops_standard+0x16)[0x7f59bc80f316]
 /usr/lib64/libperl.so.5.22(perl_run+0x2f9)[0x7f59bc79c369]
 perl(main+0x149)[0x400e39]
 /lib64/libc.so.6(__libc_start_main+0xf0)[0x7f59bc3c67b0]
 perl(_start+0x29)[0x400e79]
 === Memory map: 

 memory map  ommitted, if it is needed I can reproduce.

 So, should I downgrade -- means removing all those virtuals again -- or
 any other ideas would be appreciated.

 I  am running the unstable gentoo, if you need more information, I can
 include it.

 Thanks in advance for any suggestions.



 did you run emerge @preserved-rebuild, revdep-rebuild and perl-cleaner
 after the upgrade? That stuff's easy to forget.


 
 I ran both perl-cleaner --reallyall and emerge @preserved-rebuild which
 only rebuilt some haskall stuff.Perl cleaner had some problems, it
 tried to rebuild some python  packages which at the time (before I did
 the complete update_) had some problems because of the 3.3 to 3.4
 change.  So I just  took the emerge line from the perl cleaner, ommitted
 any  package which was a hard blocker andran the rest -- about 186
 packages.  I can try to rerun perl-cleaner if you think that would help
 any, since I now have updated the world.
 

I think running perl-cleaner --all as step 1 is wise.
It shouldn't need to make any changes, but let's cover all the usual
bases first, paying particular attention to any DBD/DBI stuff it might
find that were not installed by portage.

I am also curious why you have blockers and had to fiddle with virtuals
- the same upgrade here was clean and portage just automatically did
everything it needed. Do you have any package.* entry for perl stuff?

grep -ir perl /etc/portage


-- 
Alan McKinnon
alan.mckin...@gmail.com

[gentoo-user] upgrade to perl 5.22.0 yields perl double free or corruption

[covici]

Hi.  After upgrading to perl 5.22.0 which was a pain -- had to remove
all those virtuals -- I am now finding that a small perl program  which
connects to a mysql database yields the following when it exits:

*** Error in `perl': double free or corruption (!prev): 0x01ed8610 ***
=== Backtrace: =
/lib64/libc.so.6(+0x71e4b)[0x7f59bc417e4b]
/lib64/libc.so.6(+0x7730e)[0x7f59bc41d30e]
/lib64/libc.so.6(+0x77afb)[0x7f59bc41dafb]
/usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(mysql_db_destroy+0x32)[0x7f59bb206602]
/usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBD/mysql/mysql.so(+0x1234d)[0x7f59bb21034d]
/usr/lib64/perl5/vendor_perl/5.22.0/x86_64-linux-thread-multi/auto/DBI/DBI.so(XS_DBI_dispatch+0xcc9)[0x7f59bb83e7a9]
/usr/lib64/libperl.so.5.22(Perl_pp_entersub+0x49b)[0x7f59bc81663b]
/usr/lib64/libperl.so.5.22(Perl_call_sv+0x36f)[0x7f59bc794b5f]
/usr/lib64/libperl.so.5.22(+0xda25f)[0x7f59bc81b25f]
/usr/lib64/libperl.so.5.22(Perl_sv_clear+0x740)[0x7f59bc81bc10]
/usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
/usr/lib64/libperl.so.5.22(+0xb7298)[0x7f59bc7f8298]
/usr/lib64/libperl.so.5.22(Perl_mg_free+0x2e)[0x7f59bc7f8a4e]
/usr/lib64/libperl.so.5.22(Perl_sv_clear+0xae)[0x7f59bc81b57e]
/usr/lib64/libperl.so.5.22(Perl_sv_free2+0x5d)[0x7f59bc81be9d]
/usr/lib64/libperl.so.5.22(Perl_leave_scope+0xd51)[0x7f59bc84a411]
/usr/lib64/libperl.so.5.22(+0x52c56)[0x7f59bc793c56]
/usr/lib64/libperl.so.5.22(Perl_my_exit+0x3f)[0x7f59bc798d0f]
/usr/lib64/libperl.so.5.22(Perl_pp_exit+0x4a)[0x7f59bc857a6a]
/usr/lib64/libperl.so.5.22(Perl_runops_standard+0x16)[0x7f59bc80f316]
/usr/lib64/libperl.so.5.22(perl_run+0x2f9)[0x7f59bc79c369]
perl(main+0x149)[0x400e39]
/lib64/libc.so.6(__libc_start_main+0xf0)[0x7f59bc3c67b0]
perl(_start+0x29)[0x400e79]
=== Memory map: 

memory map  ommitted, if it is needed I can reproduce.

So, should I downgrade -- means removing all those virtuals again -- or
any other ideas would be appreciated.

I  am running the unstable gentoo, if you need more information, I can
include it.

Thanks in advance for any suggestions.

-- 
Your life is like a penny.  You're going to lose it.  The question is:
How do
you spend it?

 John Covici
 cov...@ccs.covici.com

RE: [gentoo-user] Re: systemd-224 Look out for new networking behavior [FIXED]

[Franz Fellner]

walt wrote:
 On Tue, 04 Aug 2015 08:19:37 +0200
 Franz Fellner alpine.art...@gmail.com wrote:
 
  Fernando Rodriguez wrote:
   On Monday, August 03, 2015 6:41:22 PM walt wrote:
That line declares *hostname as a constant and then the statement
below proceeds to assign a value to the 'constant'.  I wonder how
many hours of frustration have been suffered by student
programmers while trying to understand the logic behind that.
   
   Because it's not a constant, it's a pointer-to-constant :)
  Both of you are right, you can read the declaration in both ways:
  hostname is of type pointer to const char.
  *hostname is of type const char.
  
  But in this case it is not *hostname, that get's a value assigned,
  it's simply hostname. If you do not set hostname to NULL it stays
  uninitialised, which means its value is what the actual memory is set
  to - quite undefined. Correct initialization is really important and
  should be done consequently so it gets an automatism ;) (would avoid
  issues like this)
  
   
   const char *hostname; /* pointer to constant char */
   char *const hostname; /* constant pointer to char */
   const char *const hostname; /* constant pointer to constant char */
   
   Is that confusing enough?
 
 confusing++
 
 Thank you both for being patient enough to teach the ineducable :)
 
 Let me give you one more example of syntax that I find unreasonable,
 and then I'll ask my *real* question, about which I hope you will have
 opinions.
 
 Okay, the statement I referred to above uses this notation:
 
  if (!link-network-hostname)  this notation makes sense to me
  r = sd_dhcp_lease_get_hostname(lease, hostname); this doesn't

The -operator returns the address of the object, in this case of hostname.
If you would just pass hostname the function would receive a _copy_ of the 
object.
hostname is an out-argument, the function writes to it. That is needed 
sometimes
as C only can return one value, if you need to return more things you need to 
pass
them as out-args. But for that to work you need to operate on the actual object 
and
not a copy of it, so you need to pass the address to the actual object.
The declaration of the function of course needs to specify the arg as pointer 
to
the actual type, here pointer to a pointer to char.

 
 In this context does 'hostname' mean a-pointer-to-a-pointer-to-the-
 charstring we actually need?
 
 Doesn't this code seem needlessly complicated?
 
 okay, screed over, thanks for listening
 
 Somewhere I read that there was really only *one* java program ever
 written, and every subsequent java program was written by cut-and-paste
 from the first one.
 
 Is that how professional developers learn the art of programming?
 
 I really would like to hear your opinions on that question because I
 feel it's an important topic.
 
 Thanks guys.
 
 
 

Re: [gentoo-user] Re: crossdev runtime version

[Fernando Rodriguez]

On Tuesday, August 04, 2015 5:16:09 PM walt wrote:
 On Tue, 04 Aug 2015 18:40:49 +0200
 Cor Legemaat c...@cor.za.net wrote:
 
  Hi:
  
  I want to install a mingw64 compiler with =dev-util/mingw64-runtime-
  4.0.1, tried with the cmd:
   crossdev --lenv 'CFLAGS=-march=generic -mtune=generic -O2 -pipe 
   CXXFLAGS=-march=generic -mtune=generic -O2 -pipe' --ex-gdb -t 
   x86_64-w64-mingw32 --ov-output /usr/local/portage-crossdev --l
   4.0.1 --k 4.0.1 -P '-v'
  but cross-x86_64-w64-mingw32/mingw64-runtime-3.2.0-r1 get installed.
  
  What am I doing wrong?
 
 I notice you are using a version of Evolution that is not available
 yet in gentoo.  Are you using gentoo package overlays, or installing
 packages from non-gentoo source repositories?  (I see /usr/local in
 your message :)
 
 Sometimes these little details can give important clues.
 
It should give you 4.0.1 even without the --l 4.0.1 as that's the latest 
unstable version. Are you sure that crossdev went all the way through? I 
usually have to disable the fortran use flag or it fails half way. If you have 
a previous version maybe the 3.2.0 version is from another build. What's the 
output of:

qlop -gvH mingw64-runtime

You could also use crossdev -C to remove it completely and start fresh (after 
cleaning any leftovers in /etc/portage).

What is the contents of /etc/portage/package.keywords/cross-x86_64-w64-mingw32 
after the build completes? It should have ~amd64 amd64 for mingw64-runtime.

Also check that you don't have any cross-x86_64-w64-mingw32 entries anywhere 
else on your /etc/portage/package.* files and directories.

You can also verify that /usr/local/portage-crossdev/cross-x86_64-mingw32 
points to the right directory on /usr/portage just in case.

If all that checks then I at a loss. It installed 4.0.1 for me without any 
arguments but the triplet. You could try:

emerge -pv cross-x86_64-w64-mingw32/mingw64-runtime 
or
emerge -pv =cross-x86_64-w64-mingw32/mingw64-runtime-4.0.1

and see if it gives you any clues.

-- 
Fernando Rodriguez

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

Re: [gentoo-user] Configuring hostapd

[Fernando Rodriguez]

On Tuesday, August 04, 2015 8:18:43 PM Cor Legemaat wrote:
 On Sun, 2015-08-02 at 19:56 -0400, Fernando Rodriguez wrote:
  On Sunday, August 02, 2015 11:12:07 PM Mick wrote:
   On Sunday 02 Aug 2015 22:04:41 Fernando Rodriguez wrote:
On Sunday, August 02, 2015 1:29:50 PM Mick wrote:
 On Sunday 02 Aug 2015 01:50:21 Fernando Rodriguez wrote:
  Hello,
  
  After installing hostapd I can successfully connect to the 
  AP, I can
  get DHCP from it, but I cannot access the network through it 
  (neither
  lan or internet).
 
 This sounds like a (network) routing problem, rather than a 
 hostapd
 issue.

It looks like that, but if I stop iptables completely on the 
router all
unicast traffic still works in the lan (both wired and through 
an external
AP), so if I connect to the hostapd AP with iptables off, 
shouldn't I at
the very least be able to ping the wireless interface on the 
router?

I also tried with only the following rule which enables internet 
access to
all wired workstations and through external AP:

iptables -t nat -A POSTROUTING -o enp0s8 -j MASQUERADE
   
   You should probably specify the local subnet, so that multicast 
   packets are
   not sent out to the Internet, e.g.:
   
   iptables -t nat -A POSTROUTING -o enp0s8 -s 192.168.1.0/24 ! -d
  192.168.1.0/24
   -j MASQUERADE
   
   (Change 192.168.1.0/24 to suit your LAN subnet)
  
  I'm not actually using that rule except as a minimal setup for 
  troubleshooting
  this issue. My actual rules do specify the subnet.
  
   Also have you enabled ip forwarding in your kernel:
   
   sysctl -w net.ipv4.ip_forward=1
  
  Yes, it is an existing router that works perfectly except for the 
  hostapd AP.
  My current setup is as follows:
  
  Internet - Gentoo Router - Switch - AP
  
  Where AP is a wifi router with routing features disabled. Never had 
  problems
  with it. Now I installed hostapd on Gentoo Router and everything 
  else still
  works fine except when I connect to the hostapd AP. Even with only 
  that minimal
  iptable rule or no rules at all.
  
  Thanks,
  
 Probably /dev/random depleated, try enable your hardware rng or sys-
 apps/haveged test with `cat /proc/sys/kernel/random/entropy_avail`
 
 Regards:
 Cor

Thanks. II did get an error about depleted entropy at some point when starting 
hostapd but I went ahead and installed haveged and it still doesn't work. It 
doesn't even work when configured as an open AP. I checked the kernel config 
and 
I had VLAN support disabled. I've rebuilt it but can't reboot right now. Maybe 
it's required even though I'm not using VLANs? 

-- 
Fernando Rodriguez

Re: [gentoo-user] Re: crossdev runtime version

[Cor Legemaat]

On Tue, 2015-08-04 at 17:16 -0700, walt wrote:
 On Tue, 04 Aug 2015 18:40:49 +0200
 Cor Legemaat c...@cor.za.net wrote:
 
  Hi:
  
  I want to install a mingw64 compiler with =dev-util/mingw64-
  runtime-
  4.0.1, tried with the cmd:
   crossdev --lenv 'CFLAGS=-march=generic -mtune=generic -O2 -pipe
   CXXFLAGS=-march=generic -mtune=generic -O2 -pipe' --ex-gdb -t
   x86_64-w64-mingw32 --ov-output /usr/local/portage-crossdev --l
   4.0.1 --k 4.0.1 -P '-v'
  but cross-x86_64-w64-mingw32/mingw64-runtime-3.2.0-r1 get 
  installed.
  
  What am I doing wrong?
 
 I notice you are using a version of Evolution that is not available
 yet in gentoo.  Are you using gentoo package overlays, or installing
 packages from non-gentoo source repositories?  (I see /usr/local in
 your message :)
 
 Sometimes these little details can give important clues.
 
 
The pc with my mail on is a Funtoo current install but that is not the 
one where I try the crossdev on, will check later but I'm pretty sure 
my evolution is in gentoo overlay also 3.13.6. The one where I try the 
crossdev build on is a Gentoo Hardened stable install.

The /usr/local/portage-crossdev is the path to the repository to 
install the cross compiler.

Regards:
Cor

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

Re: [gentoo-user] want to upgrade 50 month old installation

[Felix Miata]

Neil Bothwick composed on 2015-08-04 21:36 (UTC+0100):

 On Tue, 04 Aug 2015 15:32:51 -0400, Felix Miata wrote:

 I've yet to figure out how to get a list of
 all installed packages akin to 'rpm -qa | sort', so I really don't know
 what my starting configuration is.

 qlist -ICv

-bash: qlist: command not found
emerge qlist fails (with unable to parse profile...unsupported EAPI '5')
-- 
The wise are known for their understanding, and pleasant
words are persuasive. Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: [gentoo-user] want to upgrade 50 month old installation

[Felix Miata]

James composed on 2015-08-04 21:07 (UTC):

 Interesting choice:: how do you like your choices, Felix?

Choices are a double edged sword. The more you have, the more power you have,
but the harder to choose, especially while overwhelmed by the unfamiliar.
Your later provided ungrading old installation links are intriguing, but
nevertheless I'm leaning heavily toward starting fresh.

Whether that or upgrade, questions asked and remaining unanswered are leaving
me unable to pick a target, whether latest release, or unknown where best
to go without being encumbered by the systemd adolescent, if there's any
practical point in so doing.

Also there's an as yet unasked question I want to get a handle on before
doing anything else. What I have now has no /dev/fb*, so I'm stuck in 80x25
mode unable to use vga= and apparently with a non-modesetting kernel. I
wouldn't want a new installation also so hamstrung right off the bat
without first knowing what to do about it.

 To the wider list of gentoo hacks::

 Still think we do not need an easier installation semantic? If he decides to
 'upgrade' there will be tons of man-hours spent on this effort. If we had a
 mostly unattended basic installation semantic (proceedure/install) I bet he
 (Felix) would choose that pathway.  Felix, care to comment?

Again it's a question of ability to and interest in dealing with choices.
Among conventional distro installers, only openSUSE's YaST2 power pleases me.
I would say the traditional text-only Debian installer (shared by *buntu) was
worst, if only Anaconda wasn't so horribly horribly unintuitive. Mageia's
isn't too bad if one doesn't mind needing to install minimal and then
pickchoose from urpmi cmdline after setting no-recommends in order to avoid
bloat.

The Gentoo instructions look competent enough to do well for most of the
people it's designed for, if only they aren't trying to do as currently I,
avoid systemd.

I really should have followed up on my installation 50 months ago at *least*
3 years ago. I have no recollection what stopped me, unless it was a naive
choice to put it on one of my oldest slowest machines with nv11 instead of
newer Intel or ATI and bunches more CPU power. It could also at least in part
be a result of space required exceeding what I'm used to. Most of my test
installations are in 4.8G / partitions that wind up 80% full or less. This
original is on 4.8G, has only 26% free, apparently has no Xorg or KDE, and no
qlist to figure what *is* installed.

 If we (gentoo) had a simple installation semantic, this sort of problem
 would most likely disappear; so the wider community could delve into other
 technical support issues.. YMMV.

I get the feeling Gentoo isn't a right choice for people who need a simple
installation semantic.
-- 
The wise are known for their understanding, and pleasant
words are persuasive. Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

[gentoo-user] Re: want to upgrade 50 month old installation

[James]

Grant Edwards grant.b.edwards at gmail.com writes:


 I've tried that pathway.  Many times.  The mostly unattended
 installers all install things I don't want, pick options I don't like,
 and end up configured to do things the way the authors of the
 installer wanted to do things rather than the way I want to do things.

Um, we can think out of the box for a new and cool installation semantic.
Just look at blueness's posting (Gentoo Reference System) on www.gentoo.org
as a new, and useful approach to installs for established gentoo admins.



  If we (gentoo) had a simple installation semantic, this sort of
  problem would most likely disappear; so the wider community could
  delve into other technical support issues.. YMMV.

 There are tons of options for a simple installation semantic if
 that's what people want.  I don't see any benefit in turning Gentoo
 into yet another me too one-click installation trying to compete
 with RedHat and Ubuntu.


Non-sequitur argument. Just because we'd have an *optional installer* does
not mean anyone would have to use it. Folks can still install the way they
like, including using ansible as Stefan does. Currently you have to
spin your own ansible setup, but it'd not be that difficult for a gentoo
reference install, based on ansible either. More options are better, imho.
No you, as an astute user, can choose any installation semantic, including
rolling your own.  I'm curious to see Felix's responses.


James

Re: [gentoo-user] want to upgrade 50 month old installation

[Felix Miata]

Neil Bothwick composed on 2015-08-04 18:44 (UTC+0100):

 On Tue, 04 Aug 2015 13:12:42 -0400, Felix Miata wrote:
 
 6-# emerge portage
  This produced a longish warning:
 !!! /etc/make.profile is not a symlink and will probably prevent most
 merges. !!! It should point into a profile within /usr/portage/profiles/
 !!! (You can safely ignore this message when syncing. It's harmless.)
 !!! If you have just changed your profile configuration, you should
 revert !!! backto the previous configuration. Due to your current
 profile being !!! invalid, allowed actions are limited to --help,
 --info, --sync, and !!! --version.
 
 So, /etc/make.profile exists, but it's a symlink to a non-existant
 ../usr/portage/profiles/default/linux/x86/10.0/desktop. Is all I need
 to do to be able to proceed to change the symlink to point
 to ...x86/13.0/des...? Any suggestions or words of wisdom?
 
 The message says it's not a symlink, not that it points nowhere. It may
 be that your cloning method dereferenced it when copying. Just reset it
 with eselect profile list followed by eselect profile set.

I think we have a n00b communication failure here. :-p

These are the current states of source and target
(post-emerge --sync, emerge portage, and eselect profile set 6 on 
target):

Source:
# uname -a  out
Linux kt400 2.6.37-gentoo-r4 #1 Sun May 15 19:32:50 EDT 2011 i686 AMD 
Athlon(tm) XP 2000+ AuthenticAMD GNU/Linux
# mount | grep ' / '  out
/dev/sda7 on / type ext3 (rw,noatime)
# blkid /dev/sda29  out
/dev/sda29: LABEL=gentoon UUID=eb3b5ce7-1675-4356-a508-ba6c30e590e0 
SEC_TYPE=ext2 TYPE=ext3 
# ls -l /etc/mak* | grep -v *conf.1* out
-rw-r--r-- 1 root root 670 May 16  2011 /etc/make.conf
-rw-r--r-- 1 root root 421 Apr 26  2011 /etc/make.conf.01
-rw-r--r-- 1 root root 544 May 15  2011 /etc/make.conf.12
-rw-r--r-- 1 root root 588 May 15  2011 /etc/make.conf.13
-rw-r--r-- 1 root root 677 May 15  2011 /etc/make.conf.14
-rw-r--r-- 1 root root 698 May 16  2011 /etc/make.conf.15
-rw-r--r-- 1 root root 670 May 16  2011 /etc/make.conf.16
-rw-r--r-- 1 root root 421 Apr 26  2011 /etc/make.conf.catalyst
lrwxrwxrwx 1 root root  40 Apr 26  2011 /etc/make.globals - 
../usr/share/portage/config/make.globals
lrwxrwxrwx 1 root root  54 May 16  2011 /etc/make.profile - 
../usr/portage/profiles/default/linux/x86/10.0/desktop
# ls -l /usr/portage/profiles/default/linux/x86/10.0/desktop/* out
-rw-r--r-- 1 portage portage2 Oct 22  2009 
/usr/portage/profiles/default/linux/x86/10.0/desktop/eapi
-rw-r--r-- 1 portage portage   34 Aug  6  2009 
/usr/portage/profiles/default/linux/x86/10.0/desktop/parent
/usr/portage/profiles/default/linux/x86/10.0/desktop/gnome:
total 2
-rw-r--r-- 1 portage portage  2 Mar 29  2010 eapi
-rw-r--r-- 1 portage portage 43 Mar 29  2010 parent
/usr/portage/profiles/default/linux/x86/10.0/desktop/kde:
total 2
-rw-r--r-- 1 portage portage  2 Mar 29  2010 eapi
-rw-r--r-- 1 portage portage 41 Mar 29  2010 parent

Target:
# uname -a  out
Linux kt400 2.6.37-gentoo-r4 #1 Sun May 15 19:32:50 EDT 2011 i686 AMD 
Athlon(tm) XP 2000+ AuthenticAMD GNU/Linux
# mount | grep ' / '
/dev/sda29 on / type ext3 (rw,noatime)
# blkid /dev/sda29
/dev/sda29: LABEL=gentoon UUID=eb3b5ce7-1675-4356-a508-ba6c30e590e0 
TYPE=ext3 
# ls -l /etc/mak* | grep -v *conf.1*
-rw-r--r-- 1 root root 670 May 16  2011 /etc/make.conf
-rw-r--r-- 1 root root 421 Apr 26  2011 /etc/make.conf.01
-rw-r--r-- 1 root root 544 May 15  2011 /etc/make.conf.12
-rw-r--r-- 1 root root 588 May 15  2011 /etc/make.conf.13
-rw-r--r-- 1 root root 677 May 15  2011 /etc/make.conf.14
-rw-r--r-- 1 root root 698 May 16  2011 /etc/make.conf.15
-rw-r--r-- 1 root root 670 May 16  2011 /etc/make.conf.16
-rw-r--r-- 1 root root 421 Apr 26  2011 /etc/make.conf.catalyst
lrwxrwxrwx 1 root root  40 Apr 26  2011 /etc/make.globals - 
../usr/share/portage/config/make.globals
lrwxrwxrwx 1 root root  58 Aug  4 13:30 /etc/make.profile - 
../usr/portage/profiles/default/linux/x86/13.0/desktop/kde
# ls -l /usr/portage/profiles/default/linux/x86/13.0/desktop/kde/*
-rw-r--r-- 1 root root2 Mar 19  2014 
/usr/portage/profiles/default/linux/x86/13.0/desktop/kde/eapi
-rw-r--r-- 1 root root   41 Jan 18  2013 
/usr/portage/profiles/default/linux/x86/13.0/desktop/kde/parent
/usr/portage/profiles/default/linux/x86/13.0/desktop/kde/systemd:
total 2
-rw-r--r-- 1 root root  2 Mar 19  2014 eapi
-rw-r--r-- 1 root root 40 Oct 19  2013 parent

In case it might be useful, .bash_history:
Up until I started today's thread:
http://fm.no-ip.com/Tmp/Linux/G/bash_history-kt400N.txt
From back when I installed 4 years ago, annotated at the time:
http://fm.no-ip.com/Tmp/Linux/G/bash_history.05
-- 
The wise are known for their understanding, and pleasant
words are persuasive. Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: [gentoo-user] Re: want to upgrade 50 month old installation

[Dale]

James wrote:
 Felix Miata mrmazda at earthlink.net writes:


 1-I just find upgrade processes more enjoyable than inital installations and
 their follow-up tedium getting from defaults back to the way I like things  
 to  work.

 Now that I've seen several thread responses subsequent to this one, I'm
 leaning towared just doing a fresh installation, but I'm curious about what
 would happen by trying, and how long it really would take.
 Interesting choice:: how do you like your choices, Felix?


 To the wider list of gentoo hacks::

 Still think we do not need an easier installation semantic? If he decides to
 'upgrade' there will be tons of man-hours spent on this effort. If we had a
 mostly unattended basic installation semantic (proceedure/install) I bet he
 (Felix) would choose that pathway.  Felix, care to comment?


 If we (gentoo) had a simple installation semantic, this sort of problem
 would most likely disappear; so the wider community could delve into other
 technical support issues.. YMMV.


 James


For me, it wouldn't matter if Gentoo had a installer or not.  It still
would be faster to do a fresh install even without a installer.  So, it
doesn't matter really.  Most of the install time is waiting on a
compile, especially on a older and slower machine. 

Dale

:-)  :-) 

[gentoo-user] Re: want to upgrade 50 month old installation

[Grant Edwards]

On 2015-08-04, James wirel...@tampabay.rr.com wrote:
 Felix Miata mrmazda at earthlink.net writes:


 1-I just find upgrade processes more enjoyable than inital installations and
 their follow-up tedium getting from defaults back to the way I like things  
 to  work.


 Now that I've seen several thread responses subsequent to this one, I'm
 leaning towared just doing a fresh installation, but I'm curious about what
 would happen by trying, and how long it really would take.

 Interesting choice:: how do you like your choices, Felix?

 To the wider list of gentoo hacks::

 Still think we do not need an easier installation semantic? If he decides to
 'upgrade' there will be tons of man-hours spent on this effort. If we had a
 mostly unattended basic installation semantic (proceedure/install) I bet he
 (Felix) would choose that pathway.

I've tried that pathway.  Many times.  The mostly unattended
installers all install things I don't want, pick options I don't like,
and end up configured to do things the way the authors of the
installer wanted to do things rather than the way I want to do things.

 If we (gentoo) had a simple installation semantic, this sort of
 problem would most likely disappear; so the wider community could
 delve into other technical support issues.. YMMV.

There are tons of options for a simple installation semantic if
that's what people want.  I don't see any benefit in turning Gentoo
into yet another me too one-click installation trying to compete
with RedHat and Ubuntu.

-- 
Grant Edwards   grant.b.edwardsYow! Let's all show human
  at   CONCERN for REVERAND MOON's
  gmail.comlegal difficulties!!

[gentoo-user] Re: systemd-224 Look out for new networking behavior [FIXED]

[walt]

On Tue, 04 Aug 2015 08:19:37 +0200
Franz Fellner alpine.art...@gmail.com wrote:

 Fernando Rodriguez wrote:
  On Monday, August 03, 2015 6:41:22 PM walt wrote:
   That line declares *hostname as a constant and then the statement
   below proceeds to assign a value to the 'constant'.  I wonder how
   many hours of frustration have been suffered by student
   programmers while trying to understand the logic behind that.
  
  Because it's not a constant, it's a pointer-to-constant :)
 Both of you are right, you can read the declaration in both ways:
 hostname is of type pointer to const char.
 *hostname is of type const char.
 
 But in this case it is not *hostname, that get's a value assigned,
 it's simply hostname. If you do not set hostname to NULL it stays
 uninitialised, which means its value is what the actual memory is set
 to - quite undefined. Correct initialization is really important and
 should be done consequently so it gets an automatism ;) (would avoid
 issues like this)
 
  
  const char *hostname; /* pointer to constant char */
  char *const hostname; /* constant pointer to char */
  const char *const hostname; /* constant pointer to constant char */
  
  Is that confusing enough?

confusing++

Thank you both for being patient enough to teach the ineducable :)

Let me give you one more example of syntax that I find unreasonable,
and then I'll ask my *real* question, about which I hope you will have
opinions.

Okay, the statement I referred to above uses this notation:

 if (!link-network-hostname)  this notation makes sense to me
 r = sd_dhcp_lease_get_hostname(lease, hostname); this doesn't

In this context does 'hostname' mean a-pointer-to-a-pointer-to-the-
charstring we actually need?

Doesn't this code seem needlessly complicated?

okay, screed over, thanks for listening

Somewhere I read that there was really only *one* java program ever
written, and every subsequent java program was written by cut-and-paste
from the first one.

Is that how professional developers learn the art of programming?

I really would like to hear your opinions on that question because I
feel it's an important topic.

Thanks guys.

Re: [gentoo-user] want to upgrade 50 month old installation

[Neil Bothwick]

On Tue, 4 Aug 2015 23:48:16 +0300, Andrew Savchenko wrote:

   2-clone the existing partition to a larger one to be the upgrade
   target 3-boot the target
   4-note that there exists no /etc/portage/  
  
  How did you clone it? It appears parts are missing.  
 
 In the old days make.conf and other files were not in /etc/portage,
 but in /etc. At least non-optional stuff.

I know, but

1) It is unlikely that /etc/portage would be missing
2) The move of make.conf to there was a few years ago
3) The fact the make.profile was not a symlink points to the cloning not
producing a clone.


-- 
Neil Bothwick

I'm not a complete idiot - several parts are missing.


pgpWn0_7Q1M0K.pgp
Description: OpenPGP digital signature

[gentoo-user] Re: crossdev runtime version

[walt]

On Tue, 04 Aug 2015 18:40:49 +0200
Cor Legemaat c...@cor.za.net wrote:

 Hi:
 
 I want to install a mingw64 compiler with =dev-util/mingw64-runtime-
 4.0.1, tried with the cmd:
  crossdev --lenv 'CFLAGS=-march=generic -mtune=generic -O2 -pipe 
  CXXFLAGS=-march=generic -mtune=generic -O2 -pipe' --ex-gdb -t 
  x86_64-w64-mingw32 --ov-output /usr/local/portage-crossdev --l
  4.0.1 --k 4.0.1 -P '-v'
 but cross-x86_64-w64-mingw32/mingw64-runtime-3.2.0-r1 get installed.
 
 What am I doing wrong?

I notice you are using a version of Evolution that is not available
yet in gentoo.  Are you using gentoo package overlays, or installing
packages from non-gentoo source repositories?  (I see /usr/local in
your message :)

Sometimes these little details can give important clues.

[gentoo-user] Re: want to upgrade 50 month old installation

[James]

Felix Miata mrmazda at earthlink.net writes:


 1-I just find upgrade processes more enjoyable than inital installations and
 their follow-up tedium getting from defaults back to the way I like things  
 to  work.


 Now that I've seen several thread responses subsequent to this one, I'm
 leaning towared just doing a fresh installation, but I'm curious about what
 would happen by trying, and how long it really would take.

Interesting choice:: how do you like your choices, Felix?


To the wider list of gentoo hacks::

Still think we do not need an easier installation semantic? If he decides to
'upgrade' there will be tons of man-hours spent on this effort. If we had a
mostly unattended basic installation semantic (proceedure/install) I bet he
(Felix) would choose that pathway.  Felix, care to comment?


If we (gentoo) had a simple installation semantic, this sort of problem
would most likely disappear; so the wider community could delve into other
technical support issues.. YMMV.


James

Re: [gentoo-user] Re: want to upgrade 50 month old installation

[Neil Bothwick]

On Tue, 4 Aug 2015 20:59:47 + (UTC), Grant Edwards wrote:

  No way on 32-bit Athlon. I have Athlon-XP. Even with distcc to
  Core2Duo it takes about 10 days of compilation time to build all
  stuff, I'm not counting time to fix all failures here. Well, I have  
  3000 packages installed...  
 
 He's going to have to compile all the user-space stuff either way
 (upgrade or fresh install), so how long that takes is moot.

Except that with an upgrade the old versions are still there and,
usually, usable while compiling the new.


-- 
Neil Bothwick

Adolescence, n.: The stage between puberty and adultery.


pgplWKRcfs5S7.pgp
Description: OpenPGP digital signature

[gentoo-user] crossdev runtime version

[Cor Legemaat]

Hi:

I wand to install a mingw64 compiler with =dev-util/mingw64-runtime-
4.0.1, tried with the cmd:
 crossdev --lenv 'CFLAGS=-march=generic -mtune=generic -O2 -pipe 
 CXXFLAGS=-march=generic -mtune=generic -O2 -pipe' --ex-gdb -t 
 x86_64-w64-mingw32 --ov-output /usr/local/portage-crossdev --l 4.0.1 
 --k 4.0.1 -P '-v'
but cross-x86_64-w64-mingw32/mingw64-runtime-3.2.0-r1 get installed.

What am I doing wrong?

Regards:
Cor

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

Re: [gentoo-user] Configuring hostapd

[Cor Legemaat]

On Sun, 2015-08-02 at 19:56 -0400, Fernando Rodriguez wrote:
 On Sunday, August 02, 2015 11:12:07 PM Mick wrote:
  On Sunday 02 Aug 2015 22:04:41 Fernando Rodriguez wrote:
   On Sunday, August 02, 2015 1:29:50 PM Mick wrote:
On Sunday 02 Aug 2015 01:50:21 Fernando Rodriguez wrote:
 Hello,
 
 After installing hostapd I can successfully connect to the 
 AP, I can
 get DHCP from it, but I cannot access the network through it 
 (neither
 lan or internet).

This sounds like a (network) routing problem, rather than a 
hostapd
issue.
   
   It looks like that, but if I stop iptables completely on the 
   router all
   unicast traffic still works in the lan (both wired and through 
   an external
   AP), so if I connect to the hostapd AP with iptables off, 
   shouldn't I at
   the very least be able to ping the wireless interface on the 
   router?
   
   I also tried with only the following rule which enables internet 
   access to
   all wired workstations and through external AP:
   
   iptables -t nat -A POSTROUTING -o enp0s8 -j MASQUERADE
  
  You should probably specify the local subnet, so that multicast 
  packets are
  not sent out to the Internet, e.g.:
  
  iptables -t nat -A POSTROUTING -o enp0s8 -s 192.168.1.0/24 ! -d
 192.168.1.0/24
  -j MASQUERADE
  
  (Change 192.168.1.0/24 to suit your LAN subnet)
 
 I'm not actually using that rule except as a minimal setup for 
 troubleshooting
 this issue. My actual rules do specify the subnet.
 
  Also have you enabled ip forwarding in your kernel:
  
  sysctl -w net.ipv4.ip_forward=1
 
 Yes, it is an existing router that works perfectly except for the 
 hostapd AP.
 My current setup is as follows:
 
 Internet - Gentoo Router - Switch - AP
 
 Where AP is a wifi router with routing features disabled. Never had 
 problems
 with it. Now I installed hostapd on Gentoo Router and everything 
 else still
 works fine except when I connect to the hostapd AP. Even with only 
 that minimal
 iptable rule or no rules at all.
 
 Thanks,
 
Probably /dev/random depleated, try enable your hardware rng or sys-
apps/haveged test with `cat /proc/sys/kernel/random/entropy_avail`

Regards:
Cor

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

[gentoo-user] Re: want to upgrade 50 month old installation

[Grant Edwards]

On 2015-08-04, Felix Miata mrma...@earthlink.net wrote:

 That's right, May 2011, my first and only Gentoo installation, 32 bit on an
 old Athlon, which means no sse2, and kernel 2.6.37. It coexists in multiboot
 on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
 it rather than installing fresh,

Can we ask why?

 if it's doable.

It probably is (for some degnerate value of doable).

My gut feeling is that a fresh install is going to be a _lot_ easier
and faster.  A fresh install will take a couple hours. An upgrade will
take somewhere between a couple days and a couple weeks.

-- 
Grant Edwards   grant.b.edwardsYow! Is it 1974?  What's
  at   for SUPPER?  Can I spend
  gmail.commy COLLEGE FUND in one
   wild afternoon??

Re: [gentoo-user] want to upgrade 50 month old installation

[Dale]

Felix Miata wrote:
 That's right, May 2011, my first and only Gentoo installation, 32 bit on an
 old Athlon, which means no sse2, and kernel 2.6.37. It coexists in multiboot
 on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
 it rather than installing fresh, if it's doable. My initial steps have been:

 1-scan through:
   a: https://wiki.gentoo.org/wiki/Upgrading_Gentoo
   b: https://wiki.gentoo.org/wiki/Handbook:X86/Working/Portage
 2-clone the existing partition to a larger one to be the upgrade target
 3-boot the target
 4-note that there exists no /etc/portage/
 5-# emerge --sync
   which warned I need to emerge portage before doing anything else
 6-# emerge portage
   This produced a longish warning:
 !!! /etc/make.profile is not a symlink and will probably prevent most merges.
 !!! It should point into a profile within /usr/portage/profiles/
 !!! (You can safely ignore this message when syncing. It's harmless.)
 !!! If you have just changed your profile configuration, you should revert
 !!! backto the previous configuration. Due to your current profile being
 !!! invalid, allowed actions are limited to --help, --info, --sync, and
 !!! --version.

 So, /etc/make.profile exists, but it's a symlink to a non-existant
 ../usr/portage/profiles/default/linux/x86/10.0/desktop. Is all I need to do
 to be able to proceed to change the symlink to point to ...x86/13.0/des...?
 Any suggestions or words of wisdom?


First, you are going to have a interesting few days, at least.  It would
be faster and easier to start fresh.  Honestly.  If you just have to or
want to for a learning experience, cool. 

See if eselect exists.  If it does, try this:

eselect profile list

If that works, pick whatever profile is closest to what you use and set
it.  That *should* take care of your first problem.  You got lots more
coming I bet.

If that doesn't work, then you have to link it the old fashioned way. 
Link the directory for the profile in
/usr/portage/profiles/your-profile to /etc/make.profile and then see
if it is happy. 

Also, since this is going to be uphill all the way, I'd use the latest
unstable portage excluding the  version.  At least that way, portage
will help solve some problems, if it can. 

I suspect this thread could get long and interesting.  o_O

Dale

:-)  :-) 

Re: [gentoo-user] want to upgrade 50 month old installation

[Felix Miata]

Grant Edwards composed on 2015-08-04 17:20 (UTC):

 Felix Miata wrote:

 That's right, May 2011, my first and only Gentoo installation, 32 bit on an
 old Athlon, which means no sse2, and kernel 2.6.37. It coexists in multiboot
 on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
 it rather than installing fresh,

 Can we ask why?

Because, assuming it's feasible, I can? :-)

1-I just find upgrade processes more enjoyable than inital installations and
their follow-up tedium getting from defaults back to the way I like things to
work.

2-From one installation to the next, I typically forget installation choices
that in hindsight I would not have made.

 if it's doable.

 It probably is (for some degnerate value of doable).

 My gut feeling is that a fresh install is going to be a _lot_ easier

For some degenerate value of easier. :-)

 and faster.  A fresh install will take a couple hours. An upgrade will
 take somewhere between a couple days and a couple weeks.

Seriously, more than a day?

Now that I've seen several thread responses subsequent to this one, I'm
leaning towared just doing a fresh installation, but I'm curious about what
would happen by trying, and how long it really would take.

Skipping or after attemping upgrade, I'd chroot from an existing, probably
openSUSE rather than Fedora, because I have Tumbleweed all the way back to
11.2 to choose from. Would there be any particular advantage to picking a
particular one to use, with/without systemd, or a kernel version close, or
newer, or older, than that which will be emerged?

I like that eselect list currently offers a kde sans systemd sans plasma
option. Ultimately what I'd like to do is get Gentoo on at least one of my
much faster systems, but only after enough experience with it to have a
respectable shot at putting Trinity on it instead of any of the more popular
DEs.  This machine is a guinea pig for familiarization purposes.
-- 
The wise are known for their understanding, and pleasant
words are persuasive. Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: [gentoo-user] Re: want to upgrade 50 month old installation

[Mick]

On Tuesday 04 Aug 2015 18:20:40 Grant Edwards wrote:
 On 2015-08-04, Felix Miata mrma...@earthlink.net wrote:
  That's right, May 2011, my first and only Gentoo installation, 32 bit on
  an old Athlon, which means no sse2, and kernel 2.6.37. It coexists in
  multiboot on one HD with 12 installations of Fedora and openSUSE. I'd
  like to upgrade it rather than installing fresh,
 
 Can we ask why?
 
  if it's doable.
 
 It probably is (for some degnerate value of doable).
 
 My gut feeling is that a fresh install is going to be a _lot_ easier
 and faster.  A fresh install will take a couple hours. An upgrade will
 take somewhere between a couple days and a couple weeks.

+1

Back up your /var/lib/portage/world and /etc, then use a LiveCD to follow the 
Gentoo handbook.  After you download and untar a stage 3 filesystem you can 
copy back your /var/lib/portage/world, build a new kernel and

 emerge -uaDv world


You can use your old config files in your /etc back up to make any quick edits 
necessary on your new installation. 
-- 
Regards,
Mick


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

[gentoo-user] want to upgrade 50 month old installation

[Felix Miata]

That's right, May 2011, my first and only Gentoo installation, 32 bit on an
old Athlon, which means no sse2, and kernel 2.6.37. It coexists in multiboot
on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
it rather than installing fresh, if it's doable. My initial steps have been:

1-scan through:
a: https://wiki.gentoo.org/wiki/Upgrading_Gentoo
b: https://wiki.gentoo.org/wiki/Handbook:X86/Working/Portage
2-clone the existing partition to a larger one to be the upgrade target
3-boot the target
4-note that there exists no /etc/portage/
5-# emerge --sync
which warned I need to emerge portage before doing anything else
6-# emerge portage
This produced a longish warning:
!!! /etc/make.profile is not a symlink and will probably prevent most merges.
!!! It should point into a profile within /usr/portage/profiles/
!!! (You can safely ignore this message when syncing. It's harmless.)
!!! If you have just changed your profile configuration, you should revert
!!! backto the previous configuration. Due to your current profile being
!!! invalid, allowed actions are limited to --help, --info, --sync, and
!!! --version.

So, /etc/make.profile exists, but it's a symlink to a non-existant
../usr/portage/profiles/default/linux/x86/10.0/desktop. Is all I need to do
to be able to proceed to change the symlink to point to ...x86/13.0/des...?
Any suggestions or words of wisdom?
-- 

The wise are known for their understanding, and pleasant
words are persuasive. Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: [gentoo-user] want to upgrade 50 month old installation

[Neil Bothwick]

On Tue, 04 Aug 2015 13:12:42 -0400, Felix Miata wrote:

 2-clone the existing partition to a larger one to be the upgrade target
 3-boot the target
 4-note that there exists no /etc/portage/

How did you clone it? It appears parts are missing.

 5-# emerge --sync
   which warned I need to emerge portage before doing anything else
 6-# emerge portage
   This produced a longish warning:
 !!! /etc/make.profile is not a symlink and will probably prevent most
 merges. !!! It should point into a profile within /usr/portage/profiles/
 !!! (You can safely ignore this message when syncing. It's harmless.)
 !!! If you have just changed your profile configuration, you should
 revert !!! backto the previous configuration. Due to your current
 profile being !!! invalid, allowed actions are limited to --help,
 --info, --sync, and !!! --version.
 
 So, /etc/make.profile exists, but it's a symlink to a non-existant
 ../usr/portage/profiles/default/linux/x86/10.0/desktop. Is all I need
 to do to be able to proceed to change the symlink to point
 to ...x86/13.0/des...? Any suggestions or words of wisdom?

The message says it's not a symlink, not that it points nowhere. It may
be that your cloning method dereferenced it when copying. Just reset it
with eselect profile list followed by eselect profile set.

Beware that you are likely to have many blockers, circular dependencies
and USE flag changes to deal with.


-- 
Neil Bothwick

Top Oxymorons Number 45: Resident alien


pgp5UbaCDAiG4.pgp
Description: OpenPGP digital signature

Re: [gentoo-user] want to upgrade 50 month old installation

[Alan Grimes]

50 months?!?!??!  that's like a fresh install Jeez, have you even
finished the burn-in testing on that?

My machine is more than 60 months old and I only recently completed the
burn-in test, (the mobo failed, wasn't giving me the PCI-E channels I
needed).

My home directory is about eleven years old...


-- 
IQ is a measure of how stupid you feel.

Powers are not rights.

Re: [gentoo-user] want to upgrade 50 month old installation

[Daniel Frey]

On 08/04/2015 11:30 AM, Felix Miata wrote:
 Grant Edwards composed on 2015-08-04 17:20 (UTC):
 
 and faster.  A fresh install will take a couple hours. An upgrade will
 take somewhere between a couple days and a couple weeks.
 
 Seriously, more than a day?
 

Oh sure, it's possible.

In late 2013 I upgraded some old installs installed in 2007/2008. They
were left stagnant for a long time due to some really finicky software
(tried an upgrade and it broke so many things on that software it wasn't
funny.)

However, this process took almost a week to iron the bugs out.

Portage had a bit of a fit and refused to do anything. There were so
many changes with core things (like udev, python, perl, and numerous
others) that it just crapped out.

I found that I had to do it in little pieces at a time and portage got
in my way constantly. I wish there was a setting to just forcibly
compile a package and then manually deal with breakage afterward with
something like revdep-rebuild, rather than trying to solve problems it
can't deal with beforehand. It would have been a lot easier.

So after upgrading some core items which took several hours of figuring
out what to remove, what to upgrade, and what to switch to (in some
cases) then I could install a new kernel and boot, after that I still
had portage getting in the way and wound up installing packages manually
instead of emerging world.

I would've just started fresh if the software package I was using
actually had an installer that worked. I had to do a lot of tricks to
get it installed initially and didn't want to repeat that process.

Dan

[gentoo-user] Re: want to upgrade 50 month old installation

[Grant Edwards]

On 2015-08-04, Andrew Savchenko birc...@gentoo.org wrote:
 On Tue, 4 Aug 2015 17:20:40 + (UTC) Grant Edwards wrote:
 On 2015-08-04, Felix Miata mrma...@earthlink.net wrote:
 
  That's right, May 2011, my first and only Gentoo installation, 32 bit on an
  old Athlon, which means no sse2, and kernel 2.6.37. It coexists in 
  multiboot
  on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
  it rather than installing fresh,
 
 Can we ask why?
 
  if it's doable.
 
 It probably is (for some degnerate value of doable).
 
 My gut feeling is that a fresh install is going to be a _lot_ easier
 and faster.  A fresh install will take a couple hours.

 With all userspace software?

No.  I'm just talking about the basic OS stuff.

 No way on 32-bit Athlon. I have Athlon-XP. Even with distcc to
 Core2Duo it takes about 10 days of compilation time to build all
 stuff, I'm not counting time to fix all failures here. Well, I have
 3000 packages installed...

He's going to have to compile all the user-space stuff either way
(upgrade or fresh install), so how long that takes is moot.

-- 
Grant Edwards   grant.b.edwardsYow! I wish I was on a
  at   Cincinnati street corner
  gmail.comholding a clean dog!

Re: [gentoo-user] want to upgrade 50 month old installation

[Stefan G. Weichinger]

Am 04.08.2015 um 20:30 schrieb Felix Miata:

 1-I just find upgrade processes more enjoyable than inital installations and
 their follow-up tedium getting from defaults back to the way I like things to
 work.

been there, done that ... several times.
Loads of work but also a big chunk of learning possible, sure.

 2-From one installation to the next, I typically forget installation choices
 that in hindsight I would not have made.

That's what software like ansible is good for.

It really pays off to learn to use such a tool ... even when you only
use basic features (like I do currently).

Writing your first roles and playbooks takes more time than simply
setting up stuff manually. But it really helps as soon as you roll that
out to the second, third, X-th server.

 Seriously, more than a day?

yes :-)

Re: [gentoo-user] want to upgrade 50 month old installation

[Felix Miata]

Dale composed on 2015-08-04 12:41 (UTC-0500):

 First, you are going to have a interesting few days, at least.  It would
 be faster and easier to start fresh.  Honestly.  If you just have to or
 want to for a learning experience, cool. 

 See if eselect exists.  If it does, try this:

 eselect profile list

 If that works, pick whatever profile is closest to what you use and set
 it.  That *should* take care of your first problem. 

No complaint from selecting 3, then 6.

 You got lots more coming I bet.

It already seems to be telling me don't. Man portage works, but portage
--help produces not found. :-P I've yet to figure out how to get a list of
all installed packages akin to 'rpm -qa | sort', so I really don't know what
my starting configuration is. Startx doesn't work, which looks like maybe
because /usr/bin/X* doesn't exist, and /etc/X11 is rather sparse.

 If that doesn't work, then you have to link it the old fashioned way. 
 Link the directory for the profile in
 /usr/portage/profiles/your-profile to /etc/make.profile and then see
 if it is happy. 

 Also, since this is going to be uphill all the way, I'd use the latest
 unstable portage excluding the  version.  At least that way, portage
 will help solve some problems, if it can. 

http://mirrors.us.kernel.org/gentoo/distfiles/portage-2.2.8.tar.bz2 wouldn't
be close enough, or is that what you're suggesting?
http://mirrors.us.kernel.org/gentoo/snapshots/ has a lot to pick from.

 I suspect this thread could get long and interesting.  o_O

At least for now, I'd like to not try to go past 20121221 in order to avoid
systemd. So far I've not found a procedure lending itself to that except to
install via http://mirrors.us.kernel.org/gentoo/releases/x86/20121221/ , but
the machine doesn't boot USB and only has a CD reader. Neither of those
normally matter, since my preferred installation method is HTTP initiated by
Grub loading an installation kernel and initrd, but I've yet to locate
Gentoo's for that AFAICT, if it even has any such thing. I guess the chroot
to untar methodology on
https://wiki.gentoo.org/wiki/Handbook:X86/Installation/Stage obviates any
such need? On http://mirrors.us.kernel.org/gentoo/releases/x86/autobuilds/ I
don't see anything that looks like a way to get to 20121221 if not already 
there.


Neil Bothwick composed on 2015-08-04 18:44 (UTC+0100):

 How did you clone it? It appears parts are missing.

I used the word clone a bit loosely. I did rsync -av after a fresh mkfs.ext3.
What's missing on clone is missing on source too. Difference in used below I
suppose is mostly on account of having already run emerge --sync?

Filesystem   1K-blocks  Used Available Use% Mounted on
/dev/sda3   396623256772119369  69% /disks/boot
/dev/sda7  4875929   3410712   1219425  74% /disks/ogentoo
/dev/sda29 6501216   3689976   2483516  60% /
/dev/sda10 4837465   3365041   1226632  74% /disks/evergreen
/dev/sda12 3250579   1302593   1784125  43% /home
/dev/sda13 1625241248895   1294417  17% /usr/local
-- 
The wise are known for their understanding, and pleasant
words are persuasive. Proverbs 16:21 (New Living Translation)

 Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata  ***  http://fm.no-ip.com/

Re: [gentoo-user] want to upgrade 50 month old installation

[Neil Bothwick]

On Tue, 04 Aug 2015 15:32:51 -0400, Felix Miata wrote:

 I've yet to figure out how to get a list of
 all installed packages akin to 'rpm -qa | sort', so I really don't know
 what my starting configuration is.

qlist -ICv


-- 
Neil Bothwick

Top Oxymorons Number 36: Alone together


pgpQ_PbTnP2vV.pgp
Description: OpenPGP digital signature

[gentoo-user] Re: want to upgrade 50 month old installation

[Grant Edwards]

On 2015-08-04, Felix Miata mrma...@earthlink.net wrote:
 Grant Edwards composed on 2015-08-04 17:20 (UTC):

 My gut feeling is that a fresh install is going to be a _lot_ easier

 For some degenerate value of easier. :-)

 and faster.  A fresh install will take a couple hours. An upgrade will
 take somewhere between a couple days and a couple weeks.

 Seriously, more than a day?

Probably.  There have beens some major changes in the past 4 years.
The last time I tried up upgrade a system that was more than a year
old, it took a couple days.

Portage was unable to resolve a lot of conflects and blockers. I hand
to uninstall a _lot_ of stuff to get to the point where portage could
be convinced to do any upgrades at all.  It would have been way faster
to do a fresh install.

 Now that I've seen several thread responses subsequent to this one,
 I'm leaning towared just doing a fresh installation, but I'm curious
 about what would happen by trying, and how long it really would take.

Give it a try and let us know how it goes.

-- 
Grant Edwards   grant.b.edwardsYow! I'm having an
  at   EMOTIONAL OUTBURST!!  But,
  gmail.comuh, WHY is there a WAFFLE
   in my PAJAMA POCKET??

Re: [gentoo-user] Re: want to upgrade 50 month old installation

[Andrew Savchenko]

On Tue, 4 Aug 2015 17:20:40 + (UTC) Grant Edwards wrote:
 On 2015-08-04, Felix Miata mrma...@earthlink.net wrote:
 
  That's right, May 2011, my first and only Gentoo installation, 32 bit on an
  old Athlon, which means no sse2, and kernel 2.6.37. It coexists in multiboot
  on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
  it rather than installing fresh,
 
 Can we ask why?
 
  if it's doable.
 
 It probably is (for some degnerate value of doable).
 
 My gut feeling is that a fresh install is going to be a _lot_ easier
 and faster.  A fresh install will take a couple hours.

With all userspace software? No way on 32-bit Athlon. I have
Athlon-XP. Even with distcc to Core2Duo it takes about 10 days
of compilation time to build all stuff, I'm not counting time to
fix all failures here. Well, I have 3000 packages installed...

 An upgrade will
 take somewhere between a couple days and a couple weeks.
 

Best regards,
Andrew Savchenko


pgpIgLqQq7xXs.pgp
Description: PGP signature

Re: [gentoo-user] want to upgrade 50 month old installation

[Jeremi Piotrowski]

On Tue, 4 Aug 2015, Daniel Frey wrote:
 I found that I had to do it in little pieces at a time and portage got
 in my way constantly. I wish there was a setting to just forcibly
 compile a package and then manually deal with breakage afterward with
 something like revdep-rebuild, rather than trying to solve problems it
 can't deal with beforehand. It would have been a lot easier.

You can do that, just go into /usr/portage and treat the whole thing like
a BSD ports system. When you're in a package directory execute

ebuild package.ebuild merge

and portage will do it's thing, only without checking dependencies. So
you'll probably get configure failures, but once a package is built that
way portage will treat it as though it were installed normally.

Re: [gentoo-user] want to upgrade 50 month old installation

[Andrew Savchenko]

On Tue, 4 Aug 2015 18:44:47 +0100 Neil Bothwick wrote:
 On Tue, 04 Aug 2015 13:12:42 -0400, Felix Miata wrote:
 
  2-clone the existing partition to a larger one to be the upgrade target
  3-boot the target
  4-note that there exists no /etc/portage/
 
 How did you clone it? It appears parts are missing.

In the old days make.conf and other files were not in /etc/portage,
but in /etc. At least non-optional stuff.

Best regards,
Andrew Savchenko


pgpk_ozEwD__p.pgp
Description: PGP signature

[gentoo-user] Re: want to upgrade 50 month old installation

[James]

Felix Miata mrmazda at earthlink.net writes:


 That's right, May 2011, my first and only Gentoo installation, 32 bit on an
 old Athlon, which means no sse2, and kernel 2.6.37. It coexists in multiboot
 on one HD with 12 installations of Fedora and openSUSE. I'd like to upgrade
 it rather than installing fresh, if it's doable.
 Any suggestions or words of wisdom?


Hello Felix. You might want to look at these (2) resources if you are
still intent on the upgrade pathway of an old gentoo installation::


[1] http://blog.siphos.be/2015/01/old-gentoo-system-not-a-problem/


[2]
https://blog.jolexa.net/2009/03/gentoo-tips-to-upgrade-your-
really-old-installation/


hth,
James

Re: [gentoo-user] Re: systemd-224 Look out for new networking behavior [FIXED]

[Mike Gilbert]

On Tue, Aug 4, 2015 at 7:56 PM, walt w41...@gmail.com wrote:
 Let me give you one more example of syntax that I find unreasonable,
 and then I'll ask my *real* question, about which I hope you will have
 opinions.

 Okay, the statement I referred to above uses this notation:

  if (!link-network-hostname)  this notation makes sense to me
  r = sd_dhcp_lease_get_hostname(lease, hostname); this doesn't

 In this context does 'hostname' mean a-pointer-to-a-pointer-to-the-
 charstring we actually need?

 Doesn't this code seem needlessly complicated?

Nope, looks like standard C to me. If you want a function to update an
argument, you need to pass a pointer to said argument. If you want to
update a pointer, you need to pass a pointer to a pointer.

[gentoo-user] ipset needs to patch the kernel?

[Meino . Cramer]

Hi,

this morning I was trying to emerge ipset (in order to use with
sidmat) and got this instead of the executable:


 Unpacking source...
 Unpacking ipset-6.20.1.tar.bz2 to 
 /var/tmp/portage/net-firewall/ipset-6.20.1/work
 Source unpacked in /var/tmp/portage/net-firewall/ipset-6.20.1/work
 Preparing source in 
 /var/tmp/portage/net-firewall/ipset-6.20.1/work/ipset-6.20.1 ...
 * Sorry, but you have to patch kernel sources with the following patch:
 *  # cd /usr/src/linux
 *  # patch -i 
/var/tmp/portage/net-firewall/ipset-6.20.1/work/ipset-6.20.1/netlink.patch -p1
 * You should recompile and run new kernel to avoid runtime errors.
 * ERROR: net-firewall/ipset-6.20.1::gentoo failed (prepare phase):
 *   Unpatched kernel
 * 

I am runnung Linux 4.1.4 vanilla and asking myself, whether this patch
is still be needed and whether there are other ways to accomplish what
ipset would do ...

I dont like the idea of patching the kernel in order to get some minor
user land tools to run...

Are there any other ways to acchieve the same ?

Best regards,
Meino