From bartoszl at ics.p.lodz.pl  Fri Oct  3 13:54:11 2014
From: bartoszl at ics.p.lodz.pl (Bartosz Lis)
Date: Fri, 3 Oct 2014 13:54:11 +0200
Subject: Default configuration of ntp-4.2.6p5-9 is vulnerable to DDoS
 participation
Message-ID: <201410031354.18178.bartoszl@ics.p.lodz.pl>

Hello,

See: https://bugzilla.redhat.com/show_bug.cgi?id=1047854 

"users can disable monitor functionality using 'disable monitor' command in 
the /etc/ntp.conf.  Note that use of 'restrict' command with 'limited' flag 
also enables monitor functionality even when 'disable monitor' command is 
used."

I suggest updating ntp.conf file found in ntp.git project with the following 
patch:

----8<----8<----
--- ntp.conf-orig       2014-10-03 13:35:16.000000000 +0200
+++ ntp.conf    2014-10-03 13:49:43.000000000 +0200
@@ -13,10 +13,15 @@ driftfile /var/lib/ntp/drift
 # ASCII file pub/leap-seconds.
 leapfile /etc/ntp/ntp.leapseconds

+# !! Important !!
+# !! To not participate in NTP based DDoS attacks keep the following line
+# !! and do not use ``limited'' option in ``restrict'' sentences.
+disable monitor
+
 # Permit time synchronization with our time source, but do not
 # permit the source to query or modify the service on this system.
-restrict default kod limited nomodify notrap nopeer noquery
-restrict -6 default kod limited nomodify notrap nopeer noquery
+restrict default kod nomodify notrap nopeer noquery
+restrict -6 default kod nomodify notrap nopeer noquery

 # Permit all access over the loopback interface.  This could
 # be tightened as well, but to do so would effect some of
----8<----8<----

-- 
Bartosz Lis



From j.rekorajski at gmail.com  Sun Oct  5 19:48:14 2014
From: j.rekorajski at gmail.com (=?UTF-8?Q?Jan_R=C4=99korajski?=)
Date: Sun, 5 Oct 2014 19:48:14 +0200
Subject: Default configuration of ntp-4.2.6p5-9 is vulnerable to DDoS
 participation
In-Reply-To: <201410031354.18178.bartoszl@ics.p.lodz.pl>
References: <201410031354.18178.bartoszl@ics.p.lodz.pl>
Message-ID: <CAKY5yHyW9p1D0zs=+tStR53vUvPBQa7o1VRmaHvYvSvEL5rbkQ@mail.gmail.com>

On Fri, Oct 3, 2014 at 1:54 PM, Bartosz Lis <bartoszl at ics.p.lodz.pl> wrote:
> Hello,
>
> See: https://bugzilla.redhat.com/show_bug.cgi?id=1047854
>
> "users can disable monitor functionality using 'disable monitor' command in
> the /etc/ntp.conf.  Note that use of 'restrict' command with 'limited' flag
> also enables monitor functionality even when 'disable monitor' command is
> used."
>
> I suggest updating ntp.conf file found in ntp.git project with the following
> patch:
>

Did you read this bug? Especially comment #3 and followups to it?
'noquery' in restrict line block monlist command. No patching needed.

-- 
Jan R?korajski | SysAdm | PLD/Linux | http://www.pld-linux.org/
baggins<at>pld-linux.org


From glen at delfi.ee  Mon Oct  6 23:32:52 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 00:32:52 +0300
Subject: apache2 restart
Message-ID: <54330A84.2050404@delfi.ee>

apache2 restart has been always unreliable in pld

but in 2.4 it's super annoying. each restart fails with an error

# service httpd restart
Checking Apache 2.4 Web Server configuration.......................[ DONE ]
Stopping Apache 2.4 Web Server service.............................[ DONE ]
Starting Apache 2.4 Web Server service.............................[ FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:80
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:80
no listening sockets available, shutting down
AH00015: Unable to open logs

and the port is in use by the same apache itself. restarting seconds 
later always "fixes":

# service httpd restart
Checking Apache 2.4 Web Server 
configuration......................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service............................................................................[ 
DONE ]


-- 
glen



From hawk at pld-linux.org  Tue Oct  7 07:46:49 2014
From: hawk at pld-linux.org (Marcin Krol)
Date: Tue, 07 Oct 2014 07:46:49 +0200
Subject: apache2 restart
In-Reply-To: <54330A84.2050404@delfi.ee>
References: <54330A84.2050404@delfi.ee>
Message-ID: <54337E49.30007@pld-linux.org>

> Checking Apache 2.4 Web Server configuration.......................[ DONE ]
> Stopping Apache 2.4 Web Server service.............................[ DONE ]
> Starting Apache 2.4 Web Server service.............................[ FAIL ]
> (98)Address already in use: AH00072: make_sock: could not bind to
> address [::]:80
> (98)Address already in use: AH00072: make_sock: could not bind to
> address 0.0.0.0:80
> no listening sockets available, shutting down
> AH00015: Unable to open logs

Not sure if its the case in PLD but in TLD apache init script doesn't 
wait for all processes to stop (and these scripts were same at some 
point, didn't checked if they're same now).

This happens when some process is busy when receiving kill signal, ie. 
handling page which is doing some large database query and it will not 
quit immediately. It also will always happen when some apache process 
has hung. I'm not seeing this behavior more than few times per year and 
in my case its almost always related to database queries.

Perhaps we should add some checks to wait for all apache processes to 
quit in similar way we do in squid init script?

M.


From arekm at maven.pl  Tue Oct  7 08:19:46 2014
From: arekm at maven.pl (Arkadiusz =?utf-8?q?Mi=C5=9Bkiewicz?=)
Date: Tue, 7 Oct 2014 08:19:46 +0200
Subject: apache2 restart
In-Reply-To: <54337E49.30007@pld-linux.org>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
Message-ID: <201410070819.47028.arekm@maven.pl>

On Tuesday 07 of October 2014, Marcin Krol wrote:
> > Checking Apache 2.4 Web Server configuration.......................[ DONE
> > ] Stopping Apache 2.4 Web Server service.............................[
> > DONE ] Starting Apache 2.4 Web Server
> > service.............................[ FAIL ] (98)Address already in use:
> > AH00072: make_sock: could not bind to address [::]:80
> > (98)Address already in use: AH00072: make_sock: could not bind to
> > address 0.0.0.0:80
> > no listening sockets available, shutting down
> > AH00015: Unable to open logs
> 
> Not sure if its the case in PLD but in TLD apache init script doesn't
> wait for all processes to stop (and these scripts were same at some
> point, didn't checked if they're same now).

Maybe we should call "httpd -k stop" instead of sending a signal?

But after looking into code it seems to be the same - sends signal without 
waiting AFAIK :/

> M.

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From glen at delfi.ee  Tue Oct  7 09:31:12 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 10:31:12 +0300
Subject: apache2 restart
In-Reply-To: <54337E49.30007@pld-linux.org>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
Message-ID: <543396C0.4020600@delfi.ee>

On 07.10.2014 08:46, Marcin Krol wrote:
> This happens when some process is busy when receiving kill signal, ie. 
> handling page which is doing some large database query and it will not 
> quit immediately. 
no no. the apaches don't have to do anything, totally blank server will 
fail as well.

the faster the machine, the larger chances are that restart will fail. 
(does not happen on slow older machines) so i say it's race condition or 
something.

-- 
glen



From hawk at pld-linux.org  Tue Oct  7 11:42:11 2014
From: hawk at pld-linux.org (Marcin Krol)
Date: Tue, 07 Oct 2014 11:42:11 +0200
Subject: apache2 restart
In-Reply-To: <543396C0.4020600@delfi.ee>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee>
Message-ID: <5433B573.3070805@pld-linux.org>

> no no. the apaches don't have to do anything, totally blank server will
> fail as well.
>
> the faster the machine, the larger chances are that restart will fail.
> (does not happen on slow older machines) so i say it's race condition or
> something.

Just restarted apache few hundred times in loop on my test server which 
basically does nothing now and can't reproduce this. Init script is same 
in PLD and TLD. I'm using prefork MPM on this machine.

M.



From glen at delfi.ee  Tue Oct  7 14:00:51 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 15:00:51 +0300
Subject: apache2 restart
In-Reply-To: <5433B573.3070805@pld-linux.org>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee> <5433B573.3070805@pld-linux.org>
Message-ID: <5433D5F3.7010108@delfi.ee>

On 07.10.2014 12:42, Marcin Krol wrote:
> Just restarted apache few hundred times in loop on my test server 
> which basically does nothing now and can't reproduce this. Init script 
> is same in PLD and TLD. I'm using prefork MPM on this machine.

i have the opposite: 100% failure rate. and also using default config 
(mpm prefork)

perhaps related to apache modules present?

[root at zave ~]# lscpu
Architecture:          i686
CPU op-mode(s):        32-bit
Byte Order:            Little Endian
CPU(s):                4
On-line CPU(s) list:   0-3
Thread(s) per core:    2
Core(s) per socket:    1
Socket(s):             2
Vendor ID:             GenuineIntel
CPU family:            15
Model:                 2
Model name:            Intel(R) Xeon(TM) CPU 2.80GHz
Stepping:              9
CPU MHz:               2791.131
BogoMIPS:              5583.15
Hypervisor vendor:     Linux-VServer
Virtualization type:   container

[root at zave ~]# uname -r
3.4.62.longterm-1

[root at zave ~]# rpm -e apache-base --test
error: Failed dependencies:
         apache(modules-api) = 20120211 is needed by (installed) 
apache-mod_php-5.3.28-4.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_alias-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authz_core-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_mime-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authn_core-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_log_config-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_include-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_auth_basic-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_dir-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_env-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_mime_magic-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_setenvif-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_speling-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_version-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_ldap-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_socache_shmcb-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_filter-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_headers-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_rewrite-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_autoindex-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authz_host-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authz_groupfile-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authz_user-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_negotiation-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authn_file-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_logio-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_authnz_ldap-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_ssl-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_deflate-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_userdir-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_status-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-errordocs-2.4.10-2.i686
         apache-base = 2.4.10-2 is needed by (installed) 
apache-mod_auth-2.4.10-2.i686

[root at zave ~]# service httpd stop; for i in $(seq 1 30); do i="--- $i 
---"; echo "> $i"; service httpd restart; echo "< $i"; done
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
 > --- 1 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 1 ---
 > --- 2 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 2 ---
 > --- 3 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 3 ---
 > --- 4 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 4 ---
 > --- 5 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 5 ---
 > --- 6 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 6 ---
 > --- 7 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 7 ---
 > --- 8 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 8 ---
 > --- 9 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 9 ---
 > --- 10 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 10 ---
 > --- 11 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 11 ---
 > --- 12 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 12 ---
 > --- 13 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 13 ---
 > --- 14 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 14 ---
 > --- 15 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 15 ---
 > --- 16 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 16 ---
 > --- 17 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 17 ---
 > --- 18 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Stopping Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
FAIL ]
(98)Address already in use: AH00072: make_sock: could not bind to 
address [::]:443
(98)Address already in use: AH00072: make_sock: could not bind to 
address 0.0.0.0:443
no listening sockets available, shutting down
AH00015: Unable to open logs
< --- 18 ---
 > --- 19 ---
Checking Apache 2.4 Web Server 
configuration...........................................................................................................................[ 
DONE ]
Apache 2.4 Web Server service is not running.
Starting Apache 2.4 Web Server 
service.................................................................................................................................[ 
DONE ]
< --- 19 ---


-- 
glen



From glen at delfi.ee  Tue Oct  7 14:24:17 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 15:24:17 +0300
Subject: apache2 restart
In-Reply-To: <5433D5F3.7010108@delfi.ee>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee> <5433B573.3070805@pld-linux.org>
 <5433D5F3.7010108@delfi.ee>
Message-ID: <5433DB71.3090802@delfi.ee>

On 07.10.2014 15:00, Elan Ruusam?e wrote:
> i have the opposite: 100% failure rate. and also using default config 
> (mpm prefork)
>
> perhaps related to apache modules present?
>
...
> [root at zave ~]# rpm -e apache-base --test
> error: Failed dependencies:
>         apache(modules-api) = 20120211 is needed by (installed) 
> apache-mod_php-5.3.28-4.i686
>         apache-base = 2.4.10-2 is needed by (installed) 
> apache-mod_alias-2.4.10-2.i686 

i started to experiment, it indeed is apache module related. so are you 
that lucky one that you don't need mod_php in your apaches?

as appears it goes down to these php modules:
Tue Oct  7 15:18:24 2014 php53-pcre-5.3.29-1.i686
Tue Oct  7 15:19:30 2014 php53-session-5.3.29-1.i686
Tue Oct  7 15:19:30 2014 php53-simplexml-5.3.29-1.i686
Tue Oct  7 15:19:30 2014 php53-spl-5.3.29-1.i686

installing or removing these modules (they have dependency loops, so 
packages are all in or all out) the reproduce rate is 100% or 0%. and 
more specifically. disabling session php module (in .ini file) makes 
reproduce rate at 0%.

so what is that module doing that makes apache shutdown to delay leaving 
resource occipied?

-- 
glen



From glen at delfi.ee  Tue Oct  7 14:35:22 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 15:35:22 +0300
Subject: apache2 restart
In-Reply-To: <5433DB71.3090802@delfi.ee>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee> <5433B573.3070805@pld-linux.org>
 <5433D5F3.7010108@delfi.ee> <5433DB71.3090802@delfi.ee>
Message-ID: <5433DE0A.5030706@delfi.ee>

On 07.10.2014 15:24, Elan Ruusam?e wrote:
>
> installing or removing these modules (they have dependency loops, so 
> packages are all in or all out) the reproduce rate is 100% or 0%. and 
> more specifically. disabling session php module (in .ini file) makes 
> reproduce rate at 0%.

so, to be clear:


disable session php module
# php -m > /tmp/php-modules-ok

enable session php module
# php -m > /tmp/php-modules-failing

# diff -u20 /tmp/php-modules-{ok,failing}
--- /tmp/php-modules-ok 2014-10-07 15:33:55.639570602 +0300
+++ /tmp/php-modules-failing    2014-10-07 15:33:53.908706890 +0300
@@ -1,12 +1,13 @@
  [PHP Modules]
  Core
  date
  ereg
  libxml
  pcre
  Reflection
+session
  SimpleXML
  SPL
  standard

  [Zend Modules]

-- 
glen



From hawk at pld-linux.org  Tue Oct  7 15:09:25 2014
From: hawk at pld-linux.org (Marcin Krol)
Date: Tue, 07 Oct 2014 15:09:25 +0200
Subject: apache2 restart
In-Reply-To: <5433DB71.3090802@delfi.ee>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee> <5433B573.3070805@pld-linux.org>
 <5433D5F3.7010108@delfi.ee> <5433DB71.3090802@delfi.ee>
Message-ID: <5433E605.3060000@pld-linux.org>

> as appears it goes down to these php modules:
> Tue Oct  7 15:18:24 2014 php53-pcre-5.3.29-1.i686
> Tue Oct  7 15:19:30 2014 php53-session-5.3.29-1.i686
> Tue Oct  7 15:19:30 2014 php53-simplexml-5.3.29-1.i686
> Tue Oct  7 15:19:30 2014 php53-spl-5.3.29-1.i686
>
> installing or removing these modules (they have dependency loops, so
> packages are all in or all out) the reproduce rate is 100% or 0%. and
> more specifically. disabling session php module (in .ini file) makes
> reproduce rate at 0%.

Have all of them installed and session module enabled (PHP 5.6). Perhaps 
it matters that 99% of my sites are using PHP via CGI/FCGI.

M.


From glen at delfi.ee  Tue Oct  7 15:11:07 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 16:11:07 +0300
Subject: apache2 restart
In-Reply-To: <5433E605.3060000@pld-linux.org>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee> <5433B573.3070805@pld-linux.org>
 <5433D5F3.7010108@delfi.ee> <5433DB71.3090802@delfi.ee>
 <5433E605.3060000@pld-linux.org>
Message-ID: <5433E66B.2050403@delfi.ee>

On 07.10.2014 16:09, Marcin Krol wrote:
> Have all of them installed and session module enabled (PHP 5.6). 
> Perhaps it matters that 99% of my sites are using PHP via CGI/FCGI. 

yes. it does matter. you don't have apache being poisoned by mod_php, 
and php problems do not reach there.

-- 
glen



From glen at delfi.ee  Tue Oct  7 15:17:43 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 07 Oct 2014 16:17:43 +0300
Subject: apache2 restart
In-Reply-To: <54330A84.2050404@delfi.ee>
References: <54330A84.2050404@delfi.ee>
Message-ID: <5433E7F7.3090800@delfi.ee>

On 07.10.2014 00:32, Elan Ruusam?e wrote:
>
>
> but in 2.4 it's super annoying. each restart fails with an error 
this can be mitigated with a delay. experimenting on single host 50000 
usleep worked.

how about new parameter for this and future issues:
http://git.pld-linux.org/?p=packages/apache.git;a=commitdiff;h=00ab488f59c5aacbcab9708a1c70b142520264d2

-- 
glen



From arekm at maven.pl  Tue Oct  7 15:25:39 2014
From: arekm at maven.pl (Arkadiusz =?utf-8?q?Mi=C5=9Bkiewicz?=)
Date: Tue, 7 Oct 2014 15:25:39 +0200
Subject: apache2 restart
In-Reply-To: <5433E7F7.3090800@delfi.ee>
References: <54330A84.2050404@delfi.ee> <5433E7F7.3090800@delfi.ee>
Message-ID: <201410071525.39317.arekm@maven.pl>

On Tuesday 07 of October 2014, Elan Ruusam?e wrote:
> On 07.10.2014 00:32, Elan Ruusam?e wrote:
> > but in 2.4 it's super annoying. each restart fails with an error
> 
> this can be mitigated with a delay. experimenting on single host 50000
> usleep worked.
> 
> how about new parameter for this and future issues:
> http://git.pld-linux.org/?p=packages/apache.git;a=commitdiff;h=00ab488f59c5
> aacbcab9708a1c70b142520264d2

Maybe bugreport to httpd people that "httpd -k stop" should wait for children 
to die before exiting (up to some timeout value) ?

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From glen at delfi.ee  Tue Oct  7 15:31:50 2014
From: glen at delfi.ee (=?UTF-8?B?RWxhbiBSdXVzYW3DpGU=?=)
Date: Tue, 07 Oct 2014 16:31:50 +0300
Subject: apache2 restart
In-Reply-To: <201410071525.39317.arekm@maven.pl>
References: <54330A84.2050404@delfi.ee> <5433E7F7.3090800@delfi.ee>
 <201410071525.39317.arekm@maven.pl>
Message-ID: <5433EB46.6080204@delfi.ee>

On 07.10.2014 16:25, Arkadiusz Mi?kiewicz wrote:
> On Tuesday 07 of October 2014, Elan Ruusam?e wrote:
>> On 07.10.2014 00:32, Elan Ruusam?e wrote:
>>> but in 2.4 it's super annoying. each restart fails with an error
>> this can be mitigated with a delay. experimenting on single host 50000
>> usleep worked.
>>
>> how about new parameter for this and future issues:
>> http://git.pld-linux.org/?p=packages/apache.git;a=commitdiff;h=00ab488f59c5
>> aacbcab9708a1c70b142520264d2
> Maybe bugreport to httpd people that "httpd -k stop" should wait for children
> to die before exiting (up to some timeout value) ?

were you able to reproduce this?

you sure some childs still exist when -k stop exits?

50000 usec is too small for me to look around in the system, don't know 
even how to strace this, as attaching strace to httpd will fix race 
condition.


-- 
glen



From qboosh at pld-linux.org  Tue Oct  7 18:22:54 2014
From: qboosh at pld-linux.org (Jakub Bogusz)
Date: Tue, 7 Oct 2014 18:22:54 +0200
Subject: [packages/rpm] use correct string returned by file for Perl
 scripts
In-Reply-To: <9d486499dc83722e4bd7ee4a05a5e7025b46b2c0_refs_heads_master@pld-linux.org>
References: <e5486255e7ecf6559dfb45a5638ff8d9fbc545c2_refs_heads_master@pld-linux.org>
 <9d486499dc83722e4bd7ee4a05a5e7025b46b2c0_refs_heads_master@pld-linux.org>
Message-ID: <20141007162254.GA6435@mail>

On Tue, Oct 07, 2014 at 05:59:44PM +0200, draenog wrote:
> +--- lib/rpmfc.c.orig	2014-10-07 15:53:03.185631053 +0100
> ++++ lib/rpmfc.c	2014-10-07 16:02:32.487591097 +0100
> +@@ -593,7 +593,7 @@
> + 
> +   { "Desktop Entry",		RPMFC_DESKTOP_FILE|RPMFC_INCLUDE },
> + 
> +-  { "perl script text",		RPMFC_PERL|RPMFC_INCLUDE },
> ++  { "Perl script",		RPMFC_PERL|RPMFC_INCLUDE },
> +   { "Perl5 module source text", RPMFC_PERL|RPMFC_MODULE|RPMFC_INCLUDE },
> + 
> +   { "PHP script text",		RPMFC_PHP|RPMFC_INCLUDE },

I think it depends on file/magic version.
IMO it's better to include both strings.


-- 
Jakub Bogusz    http://qboosh.pl/


From draenog at pld-linux.org  Tue Oct  7 18:33:55 2014
From: draenog at pld-linux.org (Kacper Kornet)
Date: Tue, 7 Oct 2014 18:33:55 +0200
Subject: [packages/rpm] use correct string returned by file for Perl
 scripts
In-Reply-To: <20141007162254.GA6435@mail>
References: <e5486255e7ecf6559dfb45a5638ff8d9fbc545c2_refs_heads_master@pld-linux.org>
 <9d486499dc83722e4bd7ee4a05a5e7025b46b2c0_refs_heads_master@pld-linux.org>
 <20141007162254.GA6435@mail>
Message-ID: <20141007163355.GA28400@camk.edu.pl>

On Tue, Oct 07, 2014 at 06:22:54PM +0200, Jakub Bogusz wrote:
> On Tue, Oct 07, 2014 at 05:59:44PM +0200, draenog wrote:
> > +--- lib/rpmfc.c.orig	2014-10-07 15:53:03.185631053 +0100
> > ++++ lib/rpmfc.c	2014-10-07 16:02:32.487591097 +0100
> > +@@ -593,7 +593,7 @@
> > + 
> > +   { "Desktop Entry",		RPMFC_DESKTOP_FILE|RPMFC_INCLUDE },
> > + 
> > +-  { "perl script text",		RPMFC_PERL|RPMFC_INCLUDE },
> > ++  { "Perl script",		RPMFC_PERL|RPMFC_INCLUDE },
> > +   { "Perl5 module source text", RPMFC_PERL|RPMFC_MODULE|RPMFC_INCLUDE },
> > + 
> > +   { "PHP script text",		RPMFC_PHP|RPMFC_INCLUDE },

> I think it depends on file/magic version.
> IMO it's better to include both strings.

Indeed, on Ac file-4.23 it returns the old string. I haven't dig
deep enough in file history previously. Fixed in next commit. 
 
-- 
  Kacper


From n3npq at me.com  Tue Oct  7 18:34:22 2014
From: n3npq at me.com (Jeffrey Johnson)
Date: Tue, 07 Oct 2014 12:34:22 -0400
Subject: [packages/rpm] use correct string returned by file for Perl
 scripts
In-Reply-To: <20141007162254.GA6435@mail>
References: <e5486255e7ecf6559dfb45a5638ff8d9fbc545c2_refs_heads_master@pld-linux.org>
 <9d486499dc83722e4bd7ee4a05a5e7025b46b2c0_refs_heads_master@pld-linux.org>
 <20141007162254.GA6435@mail>
Message-ID: <AEEBFF9F-0B42-445C-84AC-E45C2C5245EA@me.com>


On Oct 7, 2014, at 12:22 PM, Jakub Bogusz wrote:

> On Tue, Oct 07, 2014 at 05:59:44PM +0200, draenog wrote:
>> +--- lib/rpmfc.c.orig	2014-10-07 15:53:03.185631053 +0100
>> ++++ lib/rpmfc.c	2014-10-07 16:02:32.487591097 +0100
>> +@@ -593,7 +593,7 @@
>> + 
>> +   { "Desktop Entry",		RPMFC_DESKTOP_FILE|RPMFC_INCLUDE },
>> + 
>> +-  { "perl script text",		RPMFC_PERL|RPMFC_INCLUDE },
>> ++  { "Perl script",		RPMFC_PERL|RPMFC_INCLUDE },
>> +   { "Perl5 module source text", RPMFC_PERL|RPMFC_MODULE|RPMFC_INCLUDE },
>> + 
>> +   { "PHP script text",		RPMFC_PHP|RPMFC_INCLUDE },
> 
> I think it depends on file/magic version.

Yes.

> IMO it's better to include both strings.
> 

If you want to control for content in *.rpm tags, a private magic file
just for rpm is needed.

Dialects of magic strings magic should not be compiled into RPM,
that will never succeed.

hth

73 de Jeff



From hawk at pld-linux.org  Tue Oct  7 20:18:47 2014
From: hawk at pld-linux.org (Marcin Krol)
Date: Tue, 07 Oct 2014 20:18:47 +0200
Subject: apache2 restart
In-Reply-To: <5433E66B.2050403@delfi.ee>
References: <54330A84.2050404@delfi.ee> <54337E49.30007@pld-linux.org>
 <543396C0.4020600@delfi.ee> <5433B573.3070805@pld-linux.org>
 <5433D5F3.7010108@delfi.ee> <5433DB71.3090802@delfi.ee>
 <5433E605.3060000@pld-linux.org> <5433E66B.2050403@delfi.ee>
Message-ID: <54342E87.3090206@pld-linux.org>

> yes. it does matter. you don't have apache being poisoned by mod_php,
> and php problems do not reach there.

I have mod_php too on some hosts (including test one), but mod_php is 
used by just few sites.

M.


From glen at pld-linux.org  Wed Oct  8 19:13:05 2014
From: glen at pld-linux.org (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Wed, 08 Oct 2014 20:13:05 +0300
Subject: rpm5 lua
Message-ID: <543570A1.3080400@pld-linux.org>

rpm 4.5, all cool:
$ rpm -q rpm; rpm -E '%{lua:print(os.getenv("TMPDIR") or "/tmp")}'
rpm-4.5-70.i686
/home/glen/tmp


rpm 5:
d# rpm -q rpm; rpm -E '%{lua:print(os.getenv("TMPDIR") or "/tmp")}'
rpm-5.4.13-7.i686
*** Error in `rpm': malloc(): memory corruption: 0x082b3fc8 ***
...<does not return>

strace tail of the above command:

writev(3, [{"*** Error in `", 14}, {"rpm", 3}, {"': ", 3}, {"malloc(): 
memory corruption", 27}, {": 0x", 4}, {"08d06000", 8}, {" ***\n", 5}], 
7*** Error in `rpm': malloc(): memory corruption: 0x08d06000 ***
) = 64
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 
0) = 0xb76ff000
open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 4
fstat64(4, {st_mode=S_IFREG|0644, st_size=12682, ...}) = 0
mmap2(NULL, 12682, PROT_READ, MAP_PRIVATE, 4, 0) = 0xb6986000
close(4)                                = 0
open("/lib/libgcc_s.so.1", O_RDONLY|O_CLOEXEC) = 4
read(4, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\220 
\0\0004\0\0\0"..., 512) = 512
futex(0xb73d9420, FUTEX_WAIT_PRIVATE, 2, NULL <does not return>


... seems to be related to the "or str" statement, and the length of the 
inline string, <3 is ok, >=4 is epic fail:

# rpm -E '%{lua:print(os.getenv("TMPDIR") or "4231")}'
*** Error in `rpm': malloc(): memory corruption: 0x0997afb8 ***
^C
# rpm -E '%{lua:print(os.getenv("TMPDIR") or "423")}'
423
# rpm -E '%{lua:print(os.getenv("TMPDIR") or "asdf")}'
*** Error in `rpm': malloc(): memory corruption: 0x083bfe28 ***
^C
# rpm -E '%{lua:print(os.getenv("TMPDIR") or "bla")}'
bla
#

-- 
glen



From glen at pld-linux.org  Wed Oct  8 19:41:33 2014
From: glen at pld-linux.org (=?UTF-8?B?RWxhbiBSdXVzYW3DpGU=?=)
Date: Wed, 08 Oct 2014 20:41:33 +0300
Subject: rpm5 lua
In-Reply-To: <543570A1.3080400@pld-linux.org>
References: <543570A1.3080400@pld-linux.org>
Message-ID: <5435774D.1080509@pld-linux.org>

On 08.10.2014 20:13, Elan Ruusam?e wrote:
>
> ... seems to be related to the "or str" statement, and the length of 
> the inline string, <3 is ok, >=4 is epic fail:
>
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "4231")}'
> *** Error in `rpm': malloc(): memory corruption: 0x0997afb8 ***
> ^C
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "423")}'
> 423
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "asdf")}'
> *** Error in `rpm': malloc(): memory corruption: 0x083bfe28 ***
> ^C
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "bla")}'
> bla
> #
and as you already guessed, the boundary on 64bit system is 7 chars:

? rpm -E '%{lua:print(os.getenv("TMPDIR") or "blahbla")}'
/home/users/glen/tmp

? rpm -E '%{lua:print(os.getenv("TMPDIR") or "blahblah")}'
*** Error in `rpm': free(): invalid next size (fast): 0x0000000000696a50 ***
======= Backtrace: =========
/lib64/libc.so.6[0x3dec2735b8]
/lib64/libc.so.6[0x3dec2789ae]
/lib64/libc.so.6[0x3dec2796d6]
/usr/lib64/librpmio-5.4.so[0x3df0cacb93]
/usr/lib64/librpmio-5.4.so(expandMacros+0xa1)[0x3df0cad0c1]
/usr/lib64/librpmio-5.4.so(rpmExpand+0x108)[0x3df0cad268]
/usr/lib64/librpm-5.4.so[0x3df2c2d170]
/lib64/libpopt.so.0[0x3ded8029f1]
/lib64/libpopt.so.0[0x3ded802981]
/lib64/libpopt.so.0(poptGetNextOpt+0x389)[0x3ded804699]
/usr/lib64/librpm-5.4.so(rpmcliInit+0x17f)[0x3df2c2d78f]
rpm[0x40277d]
/lib64/libc.so.6(__libc_start_main+0xf5)[0x3dec221c15]
rpm[0x403ada]
...


-- 
glen



From n3npq at me.com  Wed Oct  8 22:06:41 2014
From: n3npq at me.com (Jeffrey Johnson)
Date: Wed, 08 Oct 2014 16:06:41 -0400
Subject: rpm5 lua
In-Reply-To: <543570A1.3080400@pld-linux.org>
References: <543570A1.3080400@pld-linux.org>
Message-ID: <83E99CC6-AF05-45F7-91BC-B85B60D8BEAD@me.com>

Yup ... there were numerous problems with rpm + lua-5.2.x.

All fixed about a month ago in CVS, will be in rpm-5.4.16 somewhen.

73 de Jeff

On Oct 8, 2014, at 1:13 PM, Elan Ruusam?e wrote:

> rpm 4.5, all cool:
> $ rpm -q rpm; rpm -E '%{lua:print(os.getenv("TMPDIR") or "/tmp")}'
> rpm-4.5-70.i686
> /home/glen/tmp
> 
> 
> rpm 5:
> d# rpm -q rpm; rpm -E '%{lua:print(os.getenv("TMPDIR") or "/tmp")}'
> rpm-5.4.13-7.i686
> *** Error in `rpm': malloc(): memory corruption: 0x082b3fc8 ***
> ...<does not return>
> 
> strace tail of the above command:
> 
> writev(3, [{"*** Error in `", 14}, {"rpm", 3}, {"': ", 3}, {"malloc(): memory corruption", 27}, {": 0x", 4}, {"08d06000", 8}, {" ***\n", 5}], 7*** Error in `rpm': malloc(): memory corruption: 0x08d06000 ***
> ) = 64
> mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb76ff000
> open("/etc/ld.so.cache", O_RDONLY|O_CLOEXEC) = 4
> fstat64(4, {st_mode=S_IFREG|0644, st_size=12682, ...}) = 0
> mmap2(NULL, 12682, PROT_READ, MAP_PRIVATE, 4, 0) = 0xb6986000
> close(4)                                = 0
> open("/lib/libgcc_s.so.1", O_RDONLY|O_CLOEXEC) = 4
> read(4, "\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\3\0\1\0\0\0\220 \0\0004\0\0\0"..., 512) = 512
> futex(0xb73d9420, FUTEX_WAIT_PRIVATE, 2, NULL <does not return>
> 
> 
> ... seems to be related to the "or str" statement, and the length of the inline string, <3 is ok, >=4 is epic fail:
> 
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "4231")}'
> *** Error in `rpm': malloc(): memory corruption: 0x0997afb8 ***
> ^C
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "423")}'
> 423
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "asdf")}'
> *** Error in `rpm': malloc(): memory corruption: 0x083bfe28 ***
> ^C
> # rpm -E '%{lua:print(os.getenv("TMPDIR") or "bla")}'
> bla
> #
> 
> -- 
> glen
> 
> _______________________________________________
> pld-devel-en mailing list
> pld-devel-en at lists.pld-linux.org
> http://lists.pld-linux.org/mailman/listinfo/pld-devel-en



From arekm at maven.pl  Fri Oct 10 17:01:02 2014
From: arekm at maven.pl (Arkadiusz =?utf-8?q?Mi=C5=9Bkiewicz?=)
Date: Fri, 10 Oct 2014 17:01:02 +0200
Subject: linux kernel binutils vs FSF binutils
Message-ID: <201410101701.02692.arekm@maven.pl>


From Fedora binutils.spec:

# Note - the Linux Kernel binutils releases are too unstable and contain too
# many controversial patches so we stick with the official FSF version
# instead.


Do we want to follow?

Our current binutils hang (assembler hangs) when building gcc 4.9.1 for 
example and inddeed - there was no release for long time.

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From qboosh at pld-linux.org  Fri Oct 10 17:45:23 2014
From: qboosh at pld-linux.org (Jakub Bogusz)
Date: Fri, 10 Oct 2014 17:45:23 +0200
Subject: linux kernel binutils vs FSF binutils
In-Reply-To: <201410101701.02692.arekm@maven.pl>
References: <201410101701.02692.arekm@maven.pl>
Message-ID: <20141010154523.GA10113@mail>

On Fri, Oct 10, 2014 at 05:01:02PM +0200, Arkadiusz Mi?kiewicz wrote:
> 
> From Fedora binutils.spec:
> 
> # Note - the Linux Kernel binutils releases are too unstable and contain too
> # many controversial patches so we stick with the official FSF version
> # instead.
> 
> 
> Do we want to follow?

How many patches does Fedora include against GNU binutils releases?
How many of them are included in HJL binutils?

> Our current binutils hang (assembler hangs) when building gcc 4.9.1 for 
> example and inddeed - there was no release for long time.

Have you tried with GNU binutils release?


-- 
Jakub Bogusz    http://qboosh.pl/


From arekm at maven.pl  Fri Oct 10 21:34:27 2014
From: arekm at maven.pl (Arkadiusz =?iso-8859-2?q?Mi=B6kiewicz?=)
Date: Fri, 10 Oct 2014 21:34:27 +0200
Subject: linux kernel binutils vs FSF binutils
In-Reply-To: <20141010154523.GA10113@mail>
References: <201410101701.02692.arekm@maven.pl> <20141010154523.GA10113@mail>
Message-ID: <201410102134.27312.arekm@maven.pl>

On Friday 10 of October 2014, Jakub Bogusz wrote:
> On Fri, Oct 10, 2014 at 05:01:02PM +0200, Arkadiusz Mi?kiewicz wrote:
> > From Fedora binutils.spec:
> > 
> > # Note - the Linux Kernel binutils releases are too unstable and contain
> > too # many controversial patches so we stick with the official FSF
> > version # instead.
> > 
> > 
> > Do we want to follow?
> 
> How many patches does Fedora include against GNU binutils releases?

http://pkgs.fedoraproject.org/cgit/binutils.git/plain/binutils.spec

> How many of them are included in HJL binutils?

No idea.

> > Our current binutils hang (assembler hangs) when building gcc 4.9.1 for
> > example and inddeed - there was no release for long time.
> 
> Have you tried with GNU binutils release?

I didn't try but FC has 4.9.1 and it builds ada (hjl binutils assembler 
"hangs" on ada stuff waiting for some stdin data that it never gets or 
something like that).

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From arekm at maven.pl  Thu Oct 16 20:09:53 2014
From: arekm at maven.pl (Arkadiusz =?utf-8?q?Mi=C5=9Bkiewicz?=)
Date: Thu, 16 Oct 2014 20:09:53 +0200
Subject: kde konsole underline behaviour
Message-ID: <201410162009.53382.arekm@maven.pl>


Is anyone else seeing such behaviour under kde konsole:

 http://ixion.pld-linux.org/~arekm/kde-4.14.2-konsole-bug.avi

?

How to reproduce:
vim /etc/passwd
:set cursorline

move cursor up and down

Trying to track it down. Currently I would blame konsole but...

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From ed at yen.ipipan.waw.pl  Thu Oct 16 20:18:50 2014
From: ed at yen.ipipan.waw.pl (=?utf-8?B?xYF1a2FzeiBNYcWba28=?=)
Date: Thu, 16 Oct 2014 20:18:50 +0200
Subject: kde konsole underline behaviour
In-Reply-To: <201410162009.53382.arekm@maven.pl>
References: <201410162009.53382.arekm@maven.pl>
Message-ID: <1770615.MruCsgJGll@laptok>

Dnia czwartek, 16 pa?dziernika 2014 20:09:53 Arkadiusz Mi?kiewicz pisze:
> Is anyone else seeing such behaviour under kde konsole:
> 
>  http://ixion.pld-linux.org/~arekm/kde-4.14.2-konsole-bug.avi
> 
> ?
> 
> How to reproduce:
> vim /etc/passwd
> 
> :set cursorline
> 
> move cursor up and down
> 
> Trying to track it down. Currently I would blame konsole but...

No, it's OK here.
-- 
?ukasz Ma?ko                                                            _o)
Lukasz.Masko(at)ipipan.waw.pl                                           /\\
Registered Linux User #61028                                           _\_V
Ubuntu: staroafryka?skie s?owo oznaczaj?ce "Nie umiem zainstalowa? Debiana"



From arekm at maven.pl  Thu Oct 16 21:04:05 2014
From: arekm at maven.pl (Arkadiusz =?utf-8?q?Mi=C5=9Bkiewicz?=)
Date: Thu, 16 Oct 2014 21:04:05 +0200
Subject: kde konsole underline behaviour
In-Reply-To: <1770615.MruCsgJGll@laptok>
References: <201410162009.53382.arekm@maven.pl> <1770615.MruCsgJGll@laptok>
Message-ID: <201410162104.05987.arekm@maven.pl>

On Thursday 16 of October 2014, ?ukasz Ma?ko wrote:
> Dnia czwartek, 16 pa?dziernika 2014 20:09:53 Arkadiusz Mi?kiewicz pisze:
> > Is anyone else seeing such behaviour under kde konsole:
> >  http://ixion.pld-linux.org/~arekm/kde-4.14.2-konsole-bug.avi
> > 
> > ?
> > 
> > How to reproduce:
> > vim /etc/passwd
> > 
> > :set cursorline
> > 
> > move cursor up and down
> > 
> > Trying to track it down. Currently I would blame konsole but...
> 
> No, it's OK here.

Do you youse konsole color scheme with black background? If not try one (white 
on black for example)

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From ed at yen.ipipan.waw.pl  Thu Oct 16 21:21:17 2014
From: ed at yen.ipipan.waw.pl (=?utf-8?B?xYF1a2FzeiBNYcWba28=?=)
Date: Thu, 16 Oct 2014 21:21:17 +0200
Subject: kde konsole underline behaviour
In-Reply-To: <201410162104.05987.arekm@maven.pl>
References: <201410162009.53382.arekm@maven.pl> <1770615.MruCsgJGll@laptok>
 <201410162104.05987.arekm@maven.pl>
Message-ID: <3750853.8FWWRTyRhE@laptok>

Dnia czwartek, 16 pa?dziernika 2014 21:04:05 Arkadiusz Mi?kiewicz pisze:
> On Thursday 16 of October 2014, ?ukasz Ma?ko wrote:
> > Dnia czwartek, 16 pa?dziernika 2014 20:09:53 Arkadiusz Mi?kiewicz pisze:
> > > Is anyone else seeing such behaviour under kde konsole:
> > >  http://ixion.pld-linux.org/~arekm/kde-4.14.2-konsole-bug.avi
> > > 
> > > ?
> > > 
> > > How to reproduce:
> > > vim /etc/passwd
> > > 
> > > :set cursorline
> > > 
> > > move cursor up and down
> > > 
> > > Trying to track it down. Currently I would blame konsole but...
> > 
> > No, it's OK here.
> 
> Do you youse konsole color scheme with black background? If not try one
> (white on black for example)

Nope, still works correctly.
-- 
?ukasz Ma?ko                                                            _o)
Lukasz.Masko(at)ipipan.waw.pl                                           /\\
Registered Linux User #61028                                           _\_V
Ubuntu: staroafryka?skie s?owo oznaczaj?ce "Nie umiem zainstalowa? Debiana"



From glen at delfi.ee  Sun Oct 19 16:42:34 2014
From: glen at delfi.ee (=?UTF-8?Q?Elan_Ruusam=C3=A4e?=)
Date: Sun, 19 Oct 2014 17:42:34 +0300
Subject: versioned /usr/share/pythonX.X
Message-ID: <CAHiWvWH3hSOrB1c=b5rmNn7qUDS7vRSKgw0pY9ZPDYk69y9Wow@mail.gmail.com>

http://git.pld-linux.org/gitweb.cgi?p=packages/python3.git;a=commitdiff;h=cc48ca6b2b540b5fe1a092dc486516764e980356

this requires a bit explanation and discussion

we used to rebuild whole tree of packages if python X.Y.Z version changed
of Y.


-- 
glen


From arekm at maven.pl  Sun Oct 19 18:05:56 2014
From: arekm at maven.pl (Arkadiusz =?utf-8?q?Mi=C5=9Bkiewicz?=)
Date: Sun, 19 Oct 2014 18:05:56 +0200
Subject: versioned /usr/share/pythonX.X
In-Reply-To: <CAHiWvWH3hSOrB1c=b5rmNn7qUDS7vRSKgw0pY9ZPDYk69y9Wow@mail.gmail.com>
References: <CAHiWvWH3hSOrB1c=b5rmNn7qUDS7vRSKgw0pY9ZPDYk69y9Wow@mail.gmail.com>
Message-ID: <201410191805.56102.arekm@maven.pl>

On Sunday 19 of October 2014, Elan Ruusam?e wrote:
> http://git.pld-linux.org/gitweb.cgi?p=packages/python3.git;a=commitdiff;h=c
> c48ca6b2b540b5fe1a092dc486516764e980356
> 
> this requires a bit explanation
It is now just like it was in 3.3. In 3.3 it was:

+static char share_python[] = "share/python" VERSION;

> and discussion

-- 
Arkadiusz Mi?kiewicz, arekm / ( maven.pl | pld-linux.org )


From jajcus at jajcus.net  Mon Oct 20 08:39:37 2014
From: jajcus at jajcus.net (Jacek Konieczny)
Date: Mon, 20 Oct 2014 08:39:37 +0200
Subject: versioned /usr/share/pythonX.X
In-Reply-To: <CAHiWvWH3hSOrB1c=b5rmNn7qUDS7vRSKgw0pY9ZPDYk69y9Wow@mail.gmail.com>
References: <CAHiWvWH3hSOrB1c=b5rmNn7qUDS7vRSKgw0pY9ZPDYk69y9Wow@mail.gmail.com>
Message-ID: <5444AE29.90908@jajcus.net>

On 19/10/14 16:42, Elan Ruusam?e wrote:
> http://git.pld-linux.org/gitweb.cgi?p=packages/python3.git;a=commitdiff;h=cc48ca6b2b540b5fe1a092dc486516764e980356
> 
> this requires a bit explanation and discussion
> 
> we used to rebuild whole tree of packages if python X.Y.Z version changed
> of Y.

We still need to rebuild everything so the .pyc and .so files match the
current Python. Only '.py' can be shared, but then we have problem of
.pyc for current Python being written on runtime, when a module is
imported by uid=0 and this pyc version was not included in the package.

Greets,
	Jacek


From glen at delfi.ee  Thu Oct 23 09:22:44 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Thu, 23 Oct 2014 10:22:44 +0300
Subject: www.pld-users.org
Message-ID: <5448ACC4.1040309@delfi.ee>

www.pld-users.org

the site is empty, and displays permission errors

what happened to it? why it's reporting errors to anonymous end-user. 
such option should not be enabled in production!

-- 
glen



From draenog at pld-linux.org  Tue Oct 28 14:18:26 2014
From: draenog at pld-linux.org (Kacper Kornet)
Date: Tue, 28 Oct 2014 14:18:26 +0100
Subject: Inconsistent ownerwship and mode bits for /etc/cron.d
Message-ID: <20141028131826.GB32435@camk.edu.pl>

In all crons with packages on ftp (cronie, fcron, hc-cron) the directory
/etc/cron.d belongs to group crond and has mode 750. 

However the directory is also provided by package filesystem, and there
it belongs to group root and has mode 755.

So there are two solutions:

1) Change gid and mode for /etc/cron.d in filesystem.spec. But then
crond group has to be created also in filesystem.spec.

2) Change only the mode and add %verify(not group).

-- 
  Kacper


From glen at delfi.ee  Tue Oct 28 15:49:50 2014
From: glen at delfi.ee (=?ISO-8859-1?Q?Elan_Ruusam=E4e?=)
Date: Tue, 28 Oct 2014 16:49:50 +0200
Subject: Inconsistent ownerwship and mode bits for /etc/cron.d
In-Reply-To: <20141028131826.GB32435@camk.edu.pl>
References: <20141028131826.GB32435@camk.edu.pl>
Message-ID: <544FAD0E.2050805@delfi.ee>

On 28.10.2014 15:18, Kacper Kornet wrote:
> In all crons with packages on ftp (cronie, fcron, hc-cron) the directory
> /etc/cron.d belongs to group crond and has mode 750.
>
> However the directory is also provided by package filesystem, and there
> it belongs to group root and has mode 755.
>
> So there are two solutions:
>
> 1) Change gid and mode for /etc/cron.d in filesystem.spec. But then
> crond group has to be created also in filesystem.spec.
>
> 2) Change only the mode and add %verify(not group).

see old thread:

http://lists.pld-linux.org/mailman/pipermail/pld-devel-en/2014-May/023896.html

should continue where it left off!


-- 
glen