Re: [SCIENTIFIC-LINUX-USERS] firefox 45.1 crashes

[Stephan Wiesand]

On Apr 29, 2016, at 17:18 , Pat Riehecky wrote:

> On 04/28/2016 01:25 PM, Graham Allan wrote:
>> After the excitement of seeing firefox 45.1 ESR released for SL, we're 
>> getting a handful of reports of frequent crashing.
>> 
>> I've had people try the obvious things (disable plugins esp. flash, create a 
>> new firefox profile, reboot entire workstation) without any obvious 
>> improvement.
>> 
>> Just wondering if anyone else has seen the same thing. Not seeing anything 
>> reported in bugzilla, but it's only been ~1 day.
>> 
>> Graham
> 
> I've not seen any issues running it nearly 100% of the time since the release

Crashes were seen here, and tracked down to the presence of ffmpeg from a 3rd 
party repo, see https://bugzilla.redhat.com/show_bug.cgi?id=1330898

-- Stephan

Re: php 5.6

[Stephan Wiesand]

> On 08 Apr 2016, at 14:40, Stephan Wiesand <stephan.wies...@desy.de> wrote:
> 
>> On 08 Apr 2016, at 14:22, Ralf Farke <fa...@wi.uni-muenster.de> wrote:
>> 
>> Hi,
>> i need php 5.6 for my Owncloud Server 9, installed is 5.4.16 from 
>> @sl-security.
>> I installed the repos epel and webtatic.
>> If i try to install php 5.6w i get the error:
>> php56w-common conflicts with php-common-5.4.16-36.el7_1.x86_64.
>> When i try to delete php 5.4 i get a lot of Software to delete because of 
>> dependencies, so this is not a solution.
>> 
>> Any suggestions?
> 
> rh-php56 from external_products/softwarecollections?

And BTW, https://webtatic.com/packages/php56/ has a recipe for replacing the 
installed php with php56w - "if you know what you're doing"

Re: php 5.6

[Stephan Wiesand]

> On 08 Apr 2016, at 14:22, Ralf Farke  wrote:
> 
> Hi,
> i need php 5.6 for my Owncloud Server 9, installed is 5.4.16 from 
> @sl-security.
> I installed the repos epel and webtatic.
> If i try to install php 5.6w i get the error:
> php56w-common conflicts with php-common-5.4.16-36.el7_1.x86_64.
> When i try to delete php 5.4 i get a lot of Software to delete because of 
> dependencies, so this is not a solution.
> 
> Any suggestions?

rh-php56 from external_products/softwarecollections?

> 
> Greetings,
> 
> Ralf Farke

Re: [SL7.1] iptables-services and iptables packages not sync?

[Stephan Wiesand]

> On 24 Feb 2016, at 07:39, Dirk Hoffmann <hoffm...@cppm.in2p3.fr> wrote:
> 
> Le 23 février 2016 18:19:26 CET, Stephan Wiesand <stephan.wies...@desy.de> a 
> écrit :
>> Hi Dirk,
>> 
>> a bit late, but...
>> 
>> On Feb 10, 2016, at 22:05 , Dirk Hoffmann wrote:
>> 
>>> Dear co-users,
>>> 
>>> Can anyone reproduce the following error on SL7.1?
>>> 
>>> $ sudo yum install iptables-services
>>> [...]
>>> Downloading packages:
>>> iptables-services-1.4.21-16.el7.x86_64.rpm |  49 kB  
>> 00:00
>>> Running transaction check
>>> ERROR with transaction check vs depsolve:
>>> iptables = 1.4.21-16.el7 is needed by
>> iptables-services-1.4.21-16.el7.x86_64
>>> [...]
>>> 
>>> $ rpm -q iptables
>>> iptables-1.4.21-13.el7.x86_64
>>> 
>>> $ sudo yum install iptables-services-1.4.21-13
>>> No package iptables-services-1.4.21-13 available.
>>> 
>>> Any hint what to do?
>> 
>> 
>> my 7.1 mirror consistently has the -13 packages only. The -16 ones only
>> show up in 7.2/7rolling. Looks like somehow those got mixed up.
> 
> On my side, or in SL providers? What can I do to tell you which 
> repository/mirror is actually used by my machines?

yum repolist -v
yum list iptables iptables-services


-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: [SL7.1] iptables-services and iptables packages not sync?

[Stephan Wiesand]

Hi Dirk,

a bit late, but...

On Feb 10, 2016, at 22:05 , Dirk Hoffmann wrote:

> Dear co-users,
> 
> Can anyone reproduce the following error on SL7.1?
> 
> $ sudo yum install iptables-services
> [...]
> Downloading packages:
> iptables-services-1.4.21-16.el7.x86_64.rpm |  49 kB   00:00
> Running transaction check
> ERROR with transaction check vs depsolve:
> iptables = 1.4.21-16.el7 is needed by iptables-services-1.4.21-16.el7.x86_64
> [...]
> 
> $ rpm -q iptables
> iptables-1.4.21-13.el7.x86_64
> 
> $ sudo yum install iptables-services-1.4.21-13
> No package iptables-services-1.4.21-13 available.
> 
> Any hint what to do?  


my 7.1 mirror consistently has the -13 packages only. The -16 ones only show up 
in 7.2/7rolling. Looks like somehow those got mixed up.

Stephan 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: SL7.2 systemd and ordering

[Stephan Wiesand]

Hello Jean-Michel,

> On 15 Jan 2016, at 09:13, Jean-Michel Barbet 
> <jean-michel.bar...@subatech.in2p3.fr> wrote:
> 
> I am fighting with SL72 systemd. I am trying to build kind of an
> appliance based on a LiveCD (created with livecd-creator).
> 
> The idea is to read an USB key, early enough in the startup process, and
> overwrite somes files in /etc with the content of the USB key.
> In particular, there are /etc/sysconfig/network-script/ifcfg-xxx files
> for configuring the network, ssh keys and a /etc/rsyslogd/file.conf
> 
> I did it using a rc-local.service but it is not run early enough and
> the network configuration is not right.

yes, the boot process SL7 inherits from Fedora is aggressively optimized for 
bringing up a login prompt as early as possible, by starting everything in 
parallel that can be. While this is fine for a Laptop or Workstation managed by 
the user, it makes it really hard to run startup tasks before certain other 
services are started :-(

> => Did someone on this list try sth similar ? Ideas on how to have
>   the USB read early ?

The only reliable way I found is to create an extra target as a synchronization 
point, boot into that instead of default.target, and have the last service for 
that target "isolate" default.target.

Let's call your task "localconfig". You'll create a localconfig.service doing 
the work:

---8<---
[Unit]
Description=localconfig: apply some config early
After=basic.target

[Service]
ExecStart=/my/script
Type=oneshot

[Install]
WantedBy=localconfig.target default.target
--->8---

and a localconfig_done.service to run after localconfig is complete:

---8<---
[Unit]
Description=localconfig_done: commence bootup after localconfig
After=basic.target localconfig.service

[Service]
ExecStart=/usr/bin/systemctl isolate default.target
Type=simple
RemainAfterExit=yes

[Install]
WantedBy=localconfig.target default.target
--->8---

and a localconfig.target requiring it:

---8<---
[Unit]
Description=localconfig: synchronization point for early config
Requires=basic.target
After=basic.target
Conflicts=rescue.service rescue.target

Wants=localconfig.service localconfig_done.service
 
[Install]
WantedBy=default.target
--->8---

Now add systemd.unit=localconfig.target to the kernel command line, cross 
fingers and reboot.

> Moreover, I am doubtful on the use of NetworkManager service in this
> situation.

That would make it easier to run your script before the network is configured. 
But it wouldn't use your ifcg- files.

The above may not be perfect or even wrong (I didn't test it), or there may be 
a more elegant solution. If so, rant away ;-)

Still hope it helps,
Stephan

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: a year later - CERN move to Centos - what are we doing?

[Stephan Wiesand]

On Jan 12, 2016, at 21:30 , Mark Stodola wrote:

> Since this is becoming a top-post thread, I will continue the trend.

And I'll break it, since your post is worthy ;-)

[...]

> This is not a new topic, and was discussed at length when RedHat announced 
> the acquisition of CentOS and changes to the build process

Exactly.

[...]

> One of the key features I like about SL is the ability to stay on a specific 
> point release and still receive relevant updates.

Yes, that's the difference, and a pretty significant one.

[...]

>   Arguing for distribution usage/mergers doesn't really do anything 
> productive here in a users' mailing list.

+1. Thanks.

- Stephan

>> --
>> Miles O'Neal
>> CAD Systems Engineer
>> Cirrus Logic | cirrus.com | 1.512.851.4659

Re: a year later - CERN move to Centos - what are we doing?

[Stephan Wiesand]

Will you please stop trolling here. Thanks.

On Jan 12, 2016, at 21:04 , lejeczek wrote:

> hi,
> after my first post I made a move, I should say a smaller rather, I did 
> migrate a small HA cluster from SL7.1 to Centos7.2.
> Instructions to do that I'm sure everybody can easily look up, just one tiny 
> manual intervention was needed above what is already covered by a doc on 
> Centos website.
> But most importantly nothing broke, all the usual servers, web, mail, other 
> net related services including HA carried on seamlessly.
> Like I said earlier, and everybody knows, a lot, a lot is already shared, 
> differences boil down to maybe a philosophy behind each organization 
> responsible for each snip-off, some organizational and administrative 
> processes, protocols.
> Slight advantage seems that Centos offers, but expected as they are closer to 
> the source in the lifecycle supply chain, is higher revision of some rpm 
> packages, I see I get slightly newer kernel for example, etc.
> 
> If I was to voice my opinion out - and scientific devel & other responsible 
> culprits are listening - then I say: go for it, get together, merge userbase, 
> share devel jobs, duties, etc. Merge/share or even better, tell Redhat we 
> want to use their, shared by all, bug reporting system.
> 
> I've decided, I'll be moving over to Centos, gradually but surely.
> Note, one thing to remember if you did SL -> Centos, afterwards, is yum 
> repos, make sure what you have enabled there.
> 
> cheers
> 
> On 12/01/16 09:48, lejeczek wrote:
>> hi everybody,
>> 
>> I've wondered and got curious, what do you guys, gals think about that move?
>> More importantly do you think it's a step we SL users should also consider?
>> CERN mention there were talks between them, Fermilab - what are Fermilab 
>> plans with regards to future releases, with regards to SL in general? (Not 
>> much info on the website.)
>> I personally am just about to trial a migration from SL7 to Centos. I'm 
>> thinking it's inevitable, am I wrong?
>> 
>> best wishes.
>> 

Re: g2c library

[Stephan Wiesand]

Ian,

On Dec 16, 2015, at 14:57 , Ian A Taylor wrote:

[...]

thanks for the details.

> I am having some trouble running code that utilizes the g2c library
> 
> It doesn't make any difference on whether I explicitly set the library
> path.
> It seems to me that problems are due to incompatibility issues between
> gcc-3 and gcc-4 with regard to libraries.
> 
> Is there a way to match up the gcc version and g2c library


With the advent of gcc4 ("~ since the dawn of time"), g77 was replaced with 
gfortran and libg2c with libgfortran.

Hope this helps,
Stephan

Re: SL 7.1 PXE install: xfs not found

[Stephan Wiesand]

Salut Michel,

as a data point: SL7.1 PXE/kickstart installs work for me, including XFS. We're 
mirroring the distro with rsync too (the tree, not the ISOs) and use the 
kernel/initrd from os/images/pxelinux.

Cheers,
Stephan

On Apr 28, 2015, at 18:43 , Michel Jouvin wrote:

 Thanks Connie!
 
 I am pretty sure there is something wrong on my side but I cannot find what. 
 This morning I downloaded again 7.1 distrib (rsync) and I have still the same 
 problem. But a collegue of mine did the same and it works for him. He sent me 
 his kickstart file and using exactly the same one (except the url to the 
 distrib) I still have the problem...
 
 Installing from 7.0 (with errata) works...
 
 I cannot imagine that this problem is impacted by the firmware... I never saw 
 such a weird problem! I dont understand where the /lib/modules/3.10.0.123 can 
 come from when booting 7.1 vmlinuz+initrd.
 
 Cheers,
 
 Michel
 
 Le 28/04/2015 18:38, Connie Sieh a écrit :
 I will research this.
 
 -- 
 
 Connie J. Sieh
 Computing Services Specialist III
 
 Fermi National Accelerator Laboratory
 630 840 8531 office
 
 http://www.fnal.gov
 cs...@fnal.gov
 
 On Fri, 24 Apr 2015, Michel Jouvin wrote:
 
 I double-checked the kernel version used and the kernel version of the=20
 modules provided by initrd (using lsinitrd, I should have done it=20
 before) with SL 7.1: in fact, they are both the same, 3.10.0.229. But=20
 when PXE booting vmlinuz+initrd, I end up with kernel modules for=20
 3.10.0.123 (version from 7.0) in /lib/modules... How this is possible? I=20
 removed the initrd file checked with lsinitrd to ensure that I was=20
 getting an error when booting about the missing file so I am really=20
 using this initrd file...
 
 I am lost... Any idea is welcome!
 
 Michel
 
 Le 24/04/2015 17:52, Michel Jouvin a =C3=A9crit :
 Hi,
 
 I don't know if this is related in some ways to the other recent=20
 thread about SL 7.1, PXE install. I'm struggling with SL7.1 (and I=20
 have the same pb with CentOS 7.1) when trying to do a PXE install of a=20
 machine using the last versions of images/pxeboot/vmlinuz and=20
 initrd.img. The install fails when trying to set the default=20
 filesystem type to XFS (that I don't use on this machine) because the=20
 xfs module is not found. Looking at console (ALT/F2), I saw that the=20
 kernel version provided by vmlinuz (uname -r) is not matching the=20
 kernel version for which modules are provided in initrd=20
 (/lib/modules).  If I take vmlinuz and initrd from SL7.0 I don't see=20
 the problem (but I am not sure I can install 7.1 booting with=20
 vmlinuz/initrd from 7.0).
 
 Is it expected ? Am I doing a trivial mistake ? Or is there an issue ?
 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Bizarre bug

[Stephan Wiesand]

On Mar 3, 2015, at 18:49 , Stephen John Smoogen wrote:

 On Mar 3, 2015 8:49 AM, P. Larry Nelson lnel...@illinois.edu wrote:
 
 I am seeing a bizarre bug where an SL6.x system hangs on either
 shutdown or reboot at the point where it wants to shutdown the
 loopback interface.
[...]
 [ok, here is where it becomes bizarre]
 Modify /etc/nsswitch.conf to switch the order of files nis to
 nis files for passwd, shadow, and group, as I've always done.
 Reboot.  Boom!  It hangs at loopback interface shutdown!
 
 I want to thank you for giving all the details of your testing. I would
 like to use it as a future example of how to be constructive and helpful to
 other people needing help.

Indeed.

 So have you looked at nscd any? Does having nscd turned on or off alter
 this problem. Also what is in hosts and is the NIS server listed. Thanks

And are you sure it's only passwd/group/shadow you set to nis files? Nothing 
else, in particular not hosts or ethers?

Interesting issue ;-)

Re: SL6 incompatible update of X11

[Stephan Wiesand]

On 2014-11-06, at 0:36, Konstantin Olchanski olcha...@triumf.ca wrote:

 A few days ago an updated linux kernel and updated xorg packages were
 pushed into the SL6 updates. These updates are automatically installed
 by the default yum configuration of SL6.5.
 
 Unfortunately these updates are incompatible with pre-installed X11 video
 drivers for NVIDIA (GeForce 210) and AMD/ATI (AMD E-350/E-450 and socket AM1
 on-board video) from ELREPO.
 
 These are the ELREPO kmod-fglrx and kmod-nvidia packages.
 
 So all computers with these video cards promptly broke.
 
 This incompatibility seems to be well known to the perpetrators (X.org API 
 change, leading to crash of Xorg).
 
 I think such a disruptive update should have been announced a little
 bit more widely and maybe some technical solution could have been implemented
 to avoid breaking X11 outright (i.e. refuse to install new X.org packages
 if known-incompatible NVIDIA or AMD/ATI drivers are loaded).
 
 It looks like corrected drivers are available from ELREPO, but automatic 
 updates
 from ELREPO are normally disabled because they break themselves (newly 
 installed
 package fails to reload the old kernel module resulting in Xorg not starting
 because of mismatch between newly installed userland drivers and old kernel 
 module).
 
 As end result, what could have been planned scheduled maintenance is now an 
 emergency
 patch Wednesday with many computers requiring reboot and many end users 
 disturbed.
 
 I have to fix about 6 computers with AMD/ATI drivers and only (what?) 20 
 computers with NVIDIA drivers.
 
 Please have a nice day.

I'd recommend subscribing -devel - where this batch of updates was announced 
separately when it was provided in testing.

 P.S. To add injury to insult, the super advanced Red Hat kernel module 
 management
 system (dracut) does the super slow (bzip2 -9) rebuilt of initramfs not once,
 but twice - once on install of new driver and second time on removal of old 
 driver.
 What should have taken 5 seconds takes a good 2-3 minutes (/usr/bin/time yum 
 update kmod-nvidia).

To add more fun, the -504 kernel ABI has changes in some agp... interfaces. 
Affects at least the nvidia-304 legacy driver. The 304xx packages ElRepo has 
now seem to be compatible with the -504 kernel, and thus are probably 
incompatible with earlier ones...

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: [SCIENTIFIC-LINUX-USERS] sl65 java-1.6.0-openjdk conflicts with tzdata-java-2014e

[Stephan Wiesand]

On 2014-07-22, at 15:33, Pat Riehecky riehe...@fnal.gov wrote:

 On 07/22/2014 12:56 AM, olli hauer wrote:
 Hi,
 
 I'm running into a strange conflict between  java-1.6.0-openjdk and 
 tzdata-java-2014e.
 Anyone else having a conflict with tzdata or any hints how to solve this 
 conflicts?
 
 
 
 $ rpm -qa --qf '%{NAME} %{VENDOR}\n'  | grep -e tzdata -e openjdk
 java-1.6.0-openjdk-devel Scientific Linux
 java-1.6.0-openjdk Scientific Linux
 tzdata-java Scientific Linux
 java-1.7.0-openjdk-src Scientific Linux
 java-1.6.0-openjdk-src Scientific Linux
 tzdata Scientific Linux
 java-1.7.0-openjdk-devel Scientific Linux
 java-1.7.0-openjdk Scientific Linux
 
 
 $ grep -e tzdata -e openjdk /var/log/yum.log
 Jan 28 11:24:57 Updated: 1:java-1.6.0-openjdk-1.6.0.0-3.1.13.1.el6_5.x86_64
 Jan 28 11:24:59 Updated: 
 1:java-1.6.0-openjdk-devel-1.6.0.0-3.1.13.1.el6_5.x86_64
 Jan 29 03:53:19 Updated: tzdata-2013i-2.el6.noarch
 Jan 29 03:53:20 Updated: tzdata-java-2013i-2.el6.noarch
 Mar 18 05:06:33 Updated: tzdata-2014a-1.el6.noarch
 Mar 18 05:06:34 Updated: tzdata-java-2014a-1.el6.noarch
 Apr 02 06:29:37 Updated: tzdata-java-2014b-1.el6.noarch
 Apr 02 06:29:39 Updated: tzdata-2014b-1.el6.noarch
 Apr 17 06:12:52 Updated: 1:java-1.6.0-openjdk-1.6.0.0-5.1.13.3.el6_5.x86_64
 Apr 17 06:12:54 Updated: 
 1:java-1.6.0-openjdk-devel-1.6.0.0-5.1.13.3.el6_5.x86_64
 Jun 03 16:58:00 Installed: 
 1:java-1.6.0-openjdk-src-1.6.0.0-5.1.13.3.el6_5.x86_64
 Jun 03 17:49:58 Installed: 1:java-1.7.0-openjdk-1.7.0.55-2.4.7.1.el6_5.x86_64
 Jun 03 17:50:03 Installed: 
 1:java-1.7.0-openjdk-src-1.7.0.55-2.4.7.1.el6_5.x86_64
 Jun 03 17:59:49 Installed: 
 1:java-1.7.0-openjdk-devel-1.7.0.55-2.4.7.1.el6_5.x86_64
 Jun 23 21:53:37 Updated: tzdata-2014e-1.el6.noarch
 Jul 17 04:27:17 Updated: 1:java-1.7.0-openjdk-1.7.0.65-2.5.1.2.el6_5.x86_64
 Jul 17 04:27:19 Updated: 
 1:java-1.7.0-openjdk-devel-1.7.0.65-2.5.1.2.el6_5.x86_64
 Jul 17 04:27:23 Updated: 
 1:java-1.7.0-openjdk-src-1.7.0.65-2.5.1.2.el6_5.x86_64
 
 
 $ yum upgrade
 Loaded plugins: downloadonly, security
 Setting up Upgrade Process
 Resolving Dependencies
 -- Running transaction check
 --- Package java-1.6.0-openjdk.x86_64 1:1.6.0.0-5.1.13.3.el6_5 will be 
 updated
 --- Package java-1.6.0-openjdk.x86_64 1:1.6.0.0-6.1.13.4.el6_5 will be an 
 update
  java-1.6.0-openjdk   x86_64 1:1.6.0.0-6.1.13.4.el6_5   a-sl-security  
 41 M
  java-1.6.0-openjdk-devel x86_64 1:1.6.0.0-6.1.13.4.el6_5   a-sl-security  
 14 M
  java-1.6.0-openjdk-src   x86_64 1:1.6.0.0-6.1.13.4.el6_5   a-sl-security  
 39 M
 
 Transaction Summary
 
 Upgrade   3 Package(s)
 
 Total size: 94 M
 y
 Is this ok [y/N]: Downloading Packages:
 Running rpm_check_debug
 Running Transaction Test
 
 
 Transaction Check Error:
   file 
 /usr/lib/jvm/java-1.6.0-openjdk-1.6.0.0.x86_64/jre/lib/zi/Africa/Cairo from 
 install of java-1.6.0-openjdk-1:1.6.0.0-6.1.13.4.el6_5.x86_64 conflicts with 
 file from package tzdata-java-2014e-1.el6.noarch
 
   file 
 /usr/lib/jvm/java-1.6.0-openjdk-1.6.0.0.x86_64/jre/lib/zi/Africa/Casablanca 
 from install of java-1.6.0-openjdk-1:1.6.0.0-6.1.13.4.el6_5.x86_64 conflicts 
 with file from package tzdata-java-2014e-1.el6.noarch
 
 
 
 I am not showing conflicting file within tzdata-java
 
 $ rpm -ql tzdata-java-2014e-1.el6.noarch |grep jdk
 $
 
 Perhaps a 'yum clean all' followed by 'yum reinstall tzdata-java'

Smells like someone replaced 
/usr/lib/jvm/java-1.6.0-openjdk-1.6.0.0.x86_64/jre/lib/zi by a symlink ?


-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: Scientific Linux 7 -- no more IA-32 ?

[Stephan Wiesand]

On 2014-07-08, at 10:19, Jim McCarthy jkmccar...@pacbell.net wrote:

 On Mon, 7 Jul 2014, Connie Sieh wrote:
 
 I note that only X86-64 is available; have I missed something about 
 supported ISAs, or will there also be an IA-32 port/distribution as 
 well?
 
 Yasha Karant
 
 TUV is only releasing X86-64 .
 
 -Connie Sieh
 
 Is this for TUV v7 ALPHA, or is this to become 'the new normal' going
 forward ?
 
 If no more IA-32 support, what would it take to convince the binutils (?)
 development powers-that-be to make available for X86-64 the ld linker option
 -taso (truncated address space option).  Back in the day [1], this option
 existed on Red Hat Linux for DEC Alpha, and the net effect on that 64-bit
 machine was to create an executable in which memory addresses were
 restricted to the lower 32-bits of address space.  Legacy source code that
 used 32-bit (4-byte) integers as pointers to memory addresses could
 therefore be compiled (in gcc, the -Wl,-taso option would pass -taso
 along to the linker), built, and run on the 64-bit machine, albeit without
 taking advantage of the additional memory address space available on the
 64-bit machine (e.g., the DEC Alpha processor family).
 
 Most unfortunately, the ia64 (Itanium) binutils ld linker never had this
 feature that appears to have withered away with Linux for DEC Alpha, nor has
 the X86-64 binutils ld linker had this feature either.  So in my case I've
 been hanging onto IA-32 as my SL platform-of-choice.   But if IA-32 is no
 longer going to be offered, might there be value in resuscitating the
 -taso option for the linker in X86-64 ?  From my perspective this only has
 an upside, for those that want/need it ... is there a hidden downside I
 don't see ?

The toolchain builds ia32 executables (gcc -m32 , ld -m elf_i386).
And unlike ia64, x86_64 runs them without performance penalty.

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: In Place install of SL 6.5 over Centos 6.5 ?

[Stephan Wiesand]

On Jun 25, 2014, at 18:36 , Konstantin Olchanski wrote:

 On Wed, Jun 25, 2014 at 11:15:13AM -0500, James Fait wrote:
 
 I recently received a new server system that has Centos 6.5 installed on it. 
  I would like to change that to a Scientific Linux 6.5 system without having 
 to do a full reinstall, as this has no external media access except for the 
 network.
 
 
 
 If your computer has a USB port (and can boot from USB), you can use my USB 
 installer to do a vanilla or kickstart install SL6.5:
 http://trshare.triumf.ca/~olchansk/linux/SL65-64-USBBOOT/AAA-README-USBBOOT.txt
 (download tarball is two levels up).
 
 If you have infrastructure for network booting (dhcp+tftp+pxelinux), it is 
 trivial
 to network-boot the SL6 installer and install over the network. (I find the 
 speed of
 USB install and network install to be about the same).

You don't even need any that infrastructure. Just retrieve kernel+initrd from 
images/pxeboot using wget, stuff them in /boot, and construct an appropriate 
grub entry. This way you can either start a fully automated kickstart 
installation or an interactive one (for which you need console access).

 I personally recommend a reinstall to: a) avoid creating a mongrel system 
 maybe hard to maintain long term,
 b) removes all doubt about who knows what was running on the computer before 
 you got it and
 generally gives you a clean slate to work with.


Agreed.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: DELL server and hw Raid problem with latest 6.5 kernel

[Stephan Wiesand]

Hi Andras,

On 2014-06-17, at 10:25, Andras Horvath m...@log69.com wrote:

 Hi Everyone,
 
 I've got the following server:
 
 DELL Poweredge T110 II
 01:00.0 RAID bus controller: LSI Logic / Symbios Logic MegaRAID SAS 2008 
 [Falcon] (rev 03)
 System is SL 6.5 64 bit with hw raid 1 setup, so array is seen as sda

is that a Dell card?

 I'm having problem with the latest kernel version for some time now. The 
 previous kernel version boots fine and everything works just well, but the 
 latest kernel (v2.6.32-431.17.1.el6.x86_64) cannot boot and Grub says 
 something like trying to reach blocks outside of partition and that's all 
 the message there is and boot hangs.

If it's really Grub saying this, the kernel hasn't even started. Is /boot a 
separate partition? Size and fs? Have you fsck'ed it? Checked free space and 
inodes? Tried to reinstall the kernel and the initramfs?

Curious,
Stephan

 All the former kernel versions worked fine. Could there have been a change in 
 the modules handling the raid controller?
 
 I've got another issue as well that regards the previous kernel that I'm 
 using now (v2.6.32-431.11.2.el6.x86_64) instead of the latest one. The USB 
 devices seem to lose connection from time to time (UPS and USB disks). And 
 when this happens, they cannot be used anymore. A full reboot is needed 
 before I can connect and use an external USB disk for example. Without 
 reboot, the device just doesn't show up in /dev. I found error messages in 
 the log as below:
 
 kernel: usb 2-1.1: ctrl urb status -75 received
 kernel: usb 2-1: clear tt 1 (8030) error -110
 
 
 Has anyone encountered issues like that? I'd appreciate opinions on how to 
 track further down the problem.
 
 Thanks!
 
 Andras

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: DELL server and hw Raid problem with latest 6.5 kernel

[Stephan Wiesand]

Hi Andras,

On 2014-06-17, at 11:08, Andras Horvath m...@log69.com wrote:

 On Tue, 17 Jun 2014 10:50:17 +0200
 Stephan Wiesand stephan.wies...@desy.de wrote:
 
 I've got the following server:
 
 DELL Poweredge T110 II
 01:00.0 RAID bus controller: LSI Logic / Symbios Logic MegaRAID SAS 2008 
 [Falcon] (rev 03)
 System is SL 6.5 64 bit with hw raid 1 setup, so array is seen as sda
 
 is that a Dell card?
 
 I don't know but it came with the server preinstalled. I can look after this 
 info if it can help.

Looks like an H310. I just booted an R420 with such a card into -431.17.1 and 
it works just fine. Here's the lspci -nn data: 

01:00.0 RAID bus controller [0104]: LSI Logic / Symbios Logic MegaRAID SAS 2008 
[Falcon] [1000:0073] (rev 03)

 I'm having problem with the latest kernel version for some time now. The 
 previous kernel version boots fine and everything works just well, but the 
 latest kernel (v2.6.32-431.17.1.el6.x86_64) cannot boot and Grub says 
 something like trying to reach blocks outside of partition and that's all 
 the message there is and boot hangs.
 
 If it's really Grub saying this, the kernel hasn't even started. Is /boot a 
 separate partition? Size and fs? Have you fsck'ed it? Checked free space and 
 inodes? Tried to reinstall the kernel and the initramfs?
 
 Since this is a hardver Raid, I can see only a single sda device. I have also 
 only a single root partition, the boot is not a separate one. The size is 1.8 
 TB (2 x 2 TB disks in the array, the raid BIOS of the card does not show any 
 problem). I've got plenty of free space and should not be at the limit of the 
 inodes either I believe. Also, I entirely reinstalled the kernel thinking 
 that something could have gone wrong during the update, but it gives the same 
 result.

It really seems Grub has trouble with your /boot. Maybe due to a filesystem 
inconsistency. Maybe due to a large inode number it can't deal with (does ls 
-i /boot reveal anything unusual?).

Cheers,
Stephan

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: DELL server and hw Raid problem with latest 6.5 kernel

[Stephan Wiesand]

On Jun 17, 2014, at 16:52 , Patrick J. LoPresti wrote:

 On Tue, Jun 17, 2014 at 1:25 AM, Andras Horvath m...@log69.com wrote:
 
 I'm having problem with the latest kernel version for some time now. The 
 previous kernel version boots fine and everything works just well, but the 
 latest kernel (v2.6.32-431.17.1.el6.x86_64) cannot boot and Grub says 
 something like trying to reach blocks outside of partition and that's all 
 the message there is and boot hangs.
 
 This sounds to me like your kernel has some blocks that lie beyond
 what GRUB can read during boot (using the system BIOS). It worked
 before because you got lucky; any time you reinstalled a kernel, you
 were running the risk of some of the new boot image's blocks lying
 outside the bootable range.

Sounds reasonable, but it shouldn't happen in this case (see below).

 If this is correct, checking the inode number will not help. because
 the problem the blocks inside the file itself, not the inode.

Right, if the issue is with the addressable blocks.

 Possible fixes, in increasing order of difficulty:
 
 Copy the kernel and initrd images until you get lucky again

Yes, likely to work after you delete some files larger than those written
in the early days.

 See if your system BIOS has a setting related to booting from large disks

It's a server, and not cheap cr**. And the disk isn't actually large, since 
it's  2 TiB.

 Reinstall grub with the --force-lba option

I'd be surprised if GRUB wouldn't be able to detect availability of LBA on this 
system. I wonder whether there's a way to find out on the GRUB command line.

 Reinstall the system, using an EFI boot partition (have fun)

SL6 boots quite fine from disks with GPT partition tables, even in legacy BIOS 
mode. But again, the logical drive is  2 TiB, so this shouldn't be necessary.

 Reinstall the system, creating a small (500M) /boot partition as the
 first partition on the drive

That's the best choice, and I think it's the one still recommended by TUV.

Current Fedora recommends at least 500MB for /boot.

Still curious,
Stephan

 
 That last is what I have done for years. I tried not doing so for my
 last install on a large RAID -- figuring this is the 21st century --
 and my system failed to boot. I reinstalled with a small /boot
 partition and now it consistently works fine across dozens of
 reinstalls. I do not know whether this is due to a buggy RAID BIOS or
 something else, and I do not care...
 
 Good luck.
 
 - Pat

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: DELL server and hw Raid problem with latest 6.5 kernel

[Stephan Wiesand]

Hi Andras,

On Jun 17, 2014, at 19:02 , Andras Horvath wrote:

 On Tue, 17 Jun 2014 15:52:58 +0200
 Stephan Wiesand stephan.wies...@desy.de wrote:
 On 2014-06-17, at 11:08, Andras Horvath m...@log69.com wrote:
 
 On Tue, 17 Jun 2014 10:50:17 +0200
 Stephan Wiesand stephan.wies...@desy.de wrote:
 
 I've got the following server:
 
 DELL Poweredge T110 II
 01:00.0 RAID bus controller: LSI Logic / Symbios Logic MegaRAID SAS 2008 
 [Falcon] (rev 03)
 System is SL 6.5 64 bit with hw raid 1 setup, so array is seen as sda
 
 is that a Dell card?
 
 I don't know but it came with the server preinstalled. I can look after 
 this info if it can help.
 
 Looks like an H310. I just booted an R420 with such a card into -431.17.1 
 and it works just fine. Here's the lspci -nn data: 
 
 01:00.0 RAID bus controller [0104]: LSI Logic / Symbios Logic MegaRAID SAS 
 2008 [Falcon] [1000:0073] (rev 03)
 
 I'm having problem with the latest kernel version for some time now. The 
 previous kernel version boots fine and everything works just well, but 
 the latest kernel (v2.6.32-431.17.1.el6.x86_64) cannot boot and Grub says 
 something like trying to reach blocks outside of partition and that's 
 all the message there is and boot hangs.
 
 If it's really Grub saying this, the kernel hasn't even started. Is /boot 
 a separate partition? Size and fs? Have you fsck'ed it? Checked free space 
 and inodes? Tried to reinstall the kernel and the initramfs?
 
 Since this is a hardver Raid, I can see only a single sda device. I have 
 also only a single root partition, the boot is not a separate one. The size 
 is 1.8 TB (2 x 2 TB disks in the array, the raid BIOS of the card does not 
 show any problem). I've got plenty of free space and should not be at the 
 limit of the inodes either I believe. Also, I entirely reinstalled the 
 kernel thinking that something could have gone wrong during the update, but 
 it gives the same result.
 
 It really seems Grub has trouble with your /boot. Maybe due to a filesystem 
 inconsistency. Maybe due to a large inode number it can't deal with (does 
 ls -i /boot reveal anything unusual?).
 # ls -i /boot | sort -n
 140811 grub
 141909 extlinux
 177524 System.map-2.6.32-431.5.1.el6.x86_64
 177525 config-2.6.32-431.5.1.el6.x86_64
 177526 symvers-2.6.32-431.5.1.el6.x86_64.gz
 177542 vmlinuz-2.6.32-431.5.1.el6.x86_64
 177582 initramfs-2.6.32-431.11.2.el6.x86_64.img
 178327 System.map-2.6.32-431.11.2.el6.x86_64
 178348 initramfs-2.6.32-358.14.1.el6.x86_64.img
 178452 config-2.6.32-431.11.2.el6.x86_64
 178453 symvers-2.6.32-431.11.2.el6.x86_64.gz
 178458 initramfs-2.6.32-431.5.1.el6.x86_64.img
 178529 vmlinuz-2.6.32-431.11.2.el6.x86_64
 181915 System.map-2.6.32-358.14.1.el6.x86_64
 181916 config-2.6.32-358.14.1.el6.x86_64
 181917 symvers-2.6.32-358.14.1.el6.x86_64.gz
 181918 vmlinuz-2.6.32-358.14.1.el6.x86_64
 191634 efi
 192173 elf-memtest86+-4.10
 192174 memtest86+-4.10
 262276 boot
 
 
 Does it seem high number?

No, but then there are no files for -431.17.1 ... can you install it and try 
again?

Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: DELL server and hw Raid problem with latest 6.5 kernel

[Stephan Wiesand]

On Jun 17, 2014, at 19:20 , Andras Horvath wrote:

 On Tue, 17 Jun 2014 19:08:12 +0200
 Stephan Wiesand stephan.wies...@desy.de wrote:
 On Jun 17, 2014, at 19:02 , Andras Horvath wrote:
 
 On Tue, 17 Jun 2014 15:52:58 +0200
 Stephan Wiesand stephan.wies...@desy.de wrote:
 On 2014-06-17, at 11:08, Andras Horvath m...@log69.com wrote:
 
 On Tue, 17 Jun 2014 10:50:17 +0200
 Stephan Wiesand stephan.wies...@desy.de wrote:
 
 I've got the following server:
 
 DELL Poweredge T110 II
 01:00.0 RAID bus controller: LSI Logic / Symbios Logic MegaRAID SAS 
 2008 [Falcon] (rev 03)
 System is SL 6.5 64 bit with hw raid 1 setup, so array is seen as sda
 
 is that a Dell card?
 
 I don't know but it came with the server preinstalled. I can look after 
 this info if it can help.
 
 Looks like an H310. I just booted an R420 with such a card into -431.17.1 
 and it works just fine. Here's the lspci -nn data: 
 
 01:00.0 RAID bus controller [0104]: LSI Logic / Symbios Logic MegaRAID SAS 
 2008 [Falcon] [1000:0073] (rev 03)
 
 I'm having problem with the latest kernel version for some time now. 
 The previous kernel version boots fine and everything works just well, 
 but the latest kernel (v2.6.32-431.17.1.el6.x86_64) cannot boot and 
 Grub says something like trying to reach blocks outside of partition 
 and that's all the message there is and boot hangs.
 
 If it's really Grub saying this, the kernel hasn't even started. Is 
 /boot a separate partition? Size and fs? Have you fsck'ed it? Checked 
 free space and inodes? Tried to reinstall the kernel and the initramfs?
 
 Since this is a hardver Raid, I can see only a single sda device. I have 
 also only a single root partition, the boot is not a separate one. The 
 size is 1.8 TB (2 x 2 TB disks in the array, the raid BIOS of the card 
 does not show any problem). I've got plenty of free space and should not 
 be at the limit of the inodes either I believe. Also, I entirely 
 reinstalled the kernel thinking that something could have gone wrong 
 during the update, but it gives the same result.
 
 It really seems Grub has trouble with your /boot. Maybe due to a 
 filesystem inconsistency. Maybe due to a large inode number it can't deal 
 with (does ls -i /boot reveal anything unusual?).
 # ls -i /boot | sort -n
 140811 grub
 141909 extlinux
 177524 System.map-2.6.32-431.5.1.el6.x86_64
 177525 config-2.6.32-431.5.1.el6.x86_64
 177526 symvers-2.6.32-431.5.1.el6.x86_64.gz
 177542 vmlinuz-2.6.32-431.5.1.el6.x86_64
 177582 initramfs-2.6.32-431.11.2.el6.x86_64.img
 178327 System.map-2.6.32-431.11.2.el6.x86_64
 178348 initramfs-2.6.32-358.14.1.el6.x86_64.img
 178452 config-2.6.32-431.11.2.el6.x86_64
 178453 symvers-2.6.32-431.11.2.el6.x86_64.gz
 178458 initramfs-2.6.32-431.5.1.el6.x86_64.img
 178529 vmlinuz-2.6.32-431.11.2.el6.x86_64
 181915 System.map-2.6.32-358.14.1.el6.x86_64
 181916 config-2.6.32-358.14.1.el6.x86_64
 181917 symvers-2.6.32-358.14.1.el6.x86_64.gz
 181918 vmlinuz-2.6.32-358.14.1.el6.x86_64
 191634 efi
 192173 elf-memtest86+-4.10
 192174 memtest86+-4.10
 262276 boot
 
 
 Does it seem high number?
 
 No, but then there are no files for -431.17.1 ... can you install it and try 
 again?
 # ls -i /boot | sort -n
 140811 grub
 141909 extlinux
 177524 System.map-2.6.32-431.5.1.el6.x86_64
 177525 config-2.6.32-431.5.1.el6.x86_64
 177526 symvers-2.6.32-431.5.1.el6.x86_64.gz
 177542 vmlinuz-2.6.32-431.5.1.el6.x86_64
 177582 initramfs-2.6.32-431.11.2.el6.x86_64.img
 178317 System.map-2.6.32-431.17.1.el6.x86_64
 178318 config-2.6.32-431.17.1.el6.x86_64
 178326 symvers-2.6.32-431.17.1.el6.x86_64.gz
 178327 System.map-2.6.32-431.11.2.el6.x86_64
 178348 initramfs-2.6.32-358.14.1.el6.x86_64.img
 178452 config-2.6.32-431.11.2.el6.x86_64
 178453 symvers-2.6.32-431.11.2.el6.x86_64.gz
 178458 initramfs-2.6.32-431.5.1.el6.x86_64.img
 178527 vmlinuz-2.6.32-431.17.1.el6.x86_64
 178529 vmlinuz-2.6.32-431.11.2.el6.x86_64
 179738 initramfs-2.6.32-431.17.1.el6.x86_64.img
 181915 System.map-2.6.32-358.14.1.el6.x86_64
 181916 config-2.6.32-358.14.1.el6.x86_64
 181917 symvers-2.6.32-358.14.1.el6.x86_64.gz
 181918 vmlinuz-2.6.32-358.14.1.el6.x86_64
 191634 efi
 192173 elf-memtest86+-4.10
 192174 memtest86+-4.10
 262276 boot


Inode numbers look innocious. I do wonder why you have a /boot/boot though, and 
about the extlinux. Are you sure you're using Grub to boot the server?

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: libotf not installed by SL6.5 installer?

[Stephan Wiesand]

On Mar 22, 2014, at 06:11 , Denice wrote:

 On Fri, 21 Mar 2014, Denice wrote:
 
 On Fri, 21 Mar 2014, Konstantin Olchanski wrote:
 
 I see an odd problem with installing SL 6.5. (I use my USB installer, see 
 my other message about it).
 The installation works okey, boots into linux okey, login to root okey, but 
 emacs does
 not work because libotf is not installed. This is very strange because 
 libotf is listed
 as a dependancy for the emacs package. yum install libotf fixes emacs.
 Does anybody else see this?
 
 Looks like an old problem - I saw it with SL 6.2 or 6.3 I believe. It should 
 be gone with 6.5
 
 2 different packages at one time provided libotf.so.*:
 libotf
 openmpi-psm
 
 openmpi-psm is gone in later SL versions, so the problem doesn't
 occur - at least I haven't seen it for some time.
 
 nope - I just made a query, and the issue is still there but with a
 different RPM name:
 
 $ cd  /triumfcs/mirror/scientificlinux.org/6.5/x86_64/os/Packages
 $ rpm -qp *.rpm --qf='[%{NAME} %{PROVIDES}\n]' |grep libotf.so.
 compat-openmpi libotf.so.0
 compat-openmpi libotf.so.0()(64bit)
 compat-openmpi-psm libotf.so.0()(64bit)
 libotf libotf.so.0
 libotf libotf.so.0()(64bit)
 
 So I guess whether you trip on this depends on what you are trying
 to pull in with an installation.

And the problem is new in 6.5. Earlier compat-openmpi packages don't have that 
problem.

NB the old BZ for openmpi was #768457 .

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: [SCIENTIFIC-LINUX-USERS] fc20 changed system logging

[Stephan Wiesand]

On 2014-01-31, at 15:51, Pat Riehecky riehe...@fnal.gov wrote:

 This documentation suggests rsyslog and journald coexist in the 7 Beta.

They do. By default, rsyslog is installed by default as usual, journald stores 
in /run (= tmpfs) only, and I like it.

NB the systemd authors have stated very clearly that they're not going to 
implement remote logging. Rsyslog won't go away.

 https://access.redhat.com/site/documentation/en-US/Red_Hat_Enterprise_Linux/7-Beta/html-single/Migration_Planning_Guide/index.html#_new_logging_framework
 
 On 01/31/2014 12:25 AM, ToddAndMargo wrote:
 Hi All,
 
 Reference: 
 http://docs.fedoraproject.org/en-US/Fedora/20/html/Release_Notes/sect-Release_Notes-Changes_for_Sysadmin.html
 
 I just noticed FC20 changed system logging.  That is going to
 take some getting use to.  RHEL 7 perhaps?
 
 -T
 
 
 2.8.1. Syslog removed from default installation
 syslog is no longer included in default installations. journald
 logging serves most use cases as well as, or better than, syslogd.
 
 Users accustomed to checking /var/log/messages for system logs should 
 instead use journalctl.
 
 journalctl command examples:
 
 newjournalctl   |  old messages
 +-
 journalctl  |  less /var/log/messages
 journalctl -f   |  tail -f /var/log/messages
 journalctl --unit named.service |  grep named /var/log/messages
 journalctl -b   |  Shows logs from current boot,
|  no simple equivalent.

Re: conflict x86_64 != i686

[Stephan Wiesand]

On 2014-01-30, at 12:02, Adrian Sevcenco adrian.sevce...@cern.ch wrote:

 Hi! On a SL system installed with only x86_64 packages i try to install
 a emi middleware component which have also i686 dependencies .. the
 problem is that i see this:
 [root@grid04 yum.repos.d]# yum --disablerepo=xrootd* install db4.i686
 ...snip...
 Error:  Multilib version problems found. This often means that the root
 ...snip...
 Protected multilib versions: db4-4.7.25-17.el6.i686 !=
 db4-4.7.25-18.el6_4.x86_64
 
 the problem is that i cannot uninstall the x86_64 as all the system will
 be uninstalled ...
 
 Any idea about this?

Looks like your system is already 6.5, but the SL repository you're using is 
6.4.

 Thanks!
 Adrian
 
 

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: SL6.4/5 network just stops working

[Stephan Wiesand]

On Dec 5, 2013, at 18:51 , Orion Poplawski wrote:

 I'm seeing some very strange behavior on one of our storage servers recently, 
 and am wondering if anyone else has been experiencing similar issues.  I 
 think it may be related to InfiniBand somehow, but not sure.  Unfortunately 
 there are no error messages in the logs of any kind.  But network traffic out 
 of one or more interfaces just stops, or some traffic (ping e.g.) will work, 
 but ssh/tcp won't.
 
 Seen with both 2.6.32-431 and 2.6.32-358.23.2, and I think 2.6.32-220.23.1.


Not observed here, including on ~160 systems with IB. But then we have no 
systems running -431 yet, few running -358.23.2, and none running kernels as 
old as -220.x.y. Most are on -358.x.y.

Thought it might still be a useful data point.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Superblock last mount time (date now date) is in the future.

[Stephan Wiesand]

Hi Bill,

On Oct 30, 2013, at 17:45 , Bill Askew wrote:

 Hi everyone
 
 I am running SL6.2 64bit on a Lenovo T61p.  We don't always set the date 
 to the current date and sometimes the date is in the past.  If the year is 
 2010 - 2012 I get the following message at boot up.
 
 Checking filesystem
 /dev/mapper/VolGroup-lv_root: Superblock last mount time (Wed Oct 30 
 12:11:30 2013,
  now = Sat Oct 30 06:27:50 2010) is in the future.
 
 /dev/mapper/VolGroup-lv_root: UNEXPECTED INCONSISTENCY: RUN fsck MANUALLY.
 
 [FAILED]
 *** An error occurred during the file system check.
 *** Dropping you to shell; the system will reboot
 *** when you leave the shell.
 Give root password for maintenance
 (or type Control-D to continue):
 
 This looks to me like a bug in fsck.  I can work around the boot up 
 failure by modifying rc.sysinit to run fsck with the –y option.  
 Alternatively I have created /fsckoptions with –y (this gets removed by 
 rc.sysinit after boot up).

I think it's supposed to be a feature. I vaguely remember some BZs about this 
happening during the first boot after installation in the early EL6 days (maybe 
even during beta), but not what they did to remedy the problem.

 I would appreciate any other suggestions on how to work around this 
 problem.

I'm not sure at all that it helps, nor that it won't have any negative effects, 
but I'm wondering whether setting the mount count to 0 using tune2fs would 
prevent this check from happening. And it should be possible to set the last 
mount time with debugfs. I've never tried doing any of this though. It's just 
how I would try to tackle this problem if I had to.

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: afs client startup script error

[Stephan Wiesand]

Hi Sean,

On Sep 17, 2013, at 14:16 , Sean Murray wrote:

 As I read it the afs startup script in /etc/rc.d/init.d/afs does not permit 
 the
 case of a cell not being in the CellServDB file even if you enable 
 ENABLE_AFSDB
 to append the -afsdb command line switch.

yes, that check has been there ~forever.

If it's a problem for you, you can either provide a CellServDB with your cell 
included or remove the check from the init script.

If you think this should all be done differently, please let us know how 
exactly you think it should work. That's rather a change for SL7 though.

Best regards,
Stephan

 It looks like the line 213 will block it from working ?
 CELL gets defined as the dnsname correctly but then the script refuses to 
 start
 as the CELL is not in the CellServDB, causing the script to bomb out on line 
 218
 
 Offending function is this :
 
 195 check_thiscell() {
 196 CELL=
 197 if [ -s /usr/vice/etc/ThisCell ] ; then
 198 CELL=$(cat /usr/vice/etc/ThisCell)
 199 if [ -n $CELL ]; then
 200 grep -w -q -s $CELL /usr/vice/etc/CellServDB
 201 if [ $? -ne 0 ] ; then
 202 echo afs: $CELL from ThisCell is not in CellServDB.  Not 
 starting. 12
 203 return 1
 204 fi
 205 return 0
 206 fi
 207 fi
 208 CELL=`dnsdomainname`;
 209 if [ $? -ne 0 -o -z $CELL ] ; then
 210 echo afs: No cell defined. Trying to figure it out failed. Not 
 starting. 12
 211 return 1
 212 fi
 213 grep -w -q -s $CELL /usr/vice/etc/CellServDB
 214 if [ $? -ne 0 ] ; then
 215 echo afs: No cell defined. Trying to figure it out failed. Not 
 starting. 12
 216 return 1
 217 fi
 218 echo $CELL  /usr/vice/etc/ThisCell  chmod 0644 
 /usr/vice/etc/ThisCell
 219 if [ $? -ne 0 ] ; then
 220 echo afs: No cell defined. Trying to configure $CELL failed. 
 Not starting. 12
 221 return 1
 222 fi
 223 echo afs: Warning: No cell was defined. Autoconfigured $CELL. 12
 224 }
 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: [SCIENTIFIC-LINUX-USERS] kernel 2.6.32-358.6.1.el6 updates removes openafs

[Stephan Wiesand]

On Apr 26, 2013, at 15:31 , Pat Riehecky riehe...@fnal.gov wrote:

 On 04/25/2013 04:52 PM, Joseph Thomas Szep wrote:
 Hi,
 
 We tried to install the lastest kernel update kernel-2.6.32-358.6.1.el6 and 
 due to
 an odd chain of dependencies, all openafs packages were removed.
 
 On our systems, we only keep 2 kernels (as opposed to the default of 3).  So 
 we have
 kernels:
 
 # rpm -q kernel
 kernel-2.6.32-279.19.1.el6.x86_64
 kernel-2.6.32-358.2.1.el6.x86_64
 
 and openafs packages:
 
 openafs.x86_64   1.6.2-0.144.sl6 
 @sl/6.2
 openafs-client.x86_641.6.2-0.144.sl6 
 @sl/6.2
 openafs-krb5.x86_64  1.6.2-0.144.sl6 
 @sl/6.2
 openafs-module-tools.x86_64  1.6.2-0.144.sl6 
 @sl-security/6.2
 kmod-openafs.noarch  1.6.2-4.SL64.el6
 @sl-security/6.2
 kmod-openafs-279.x86_64  1.6.2-0.144.sl6.279 
 @sl-security/6.2
 kmod-openafs-358.x86_64  1.6.2-0.144.sl6.358.0.1 
 @sl-security/6.2
 
 When the kernel-2.6.32-358.6.1 packages install (due to installonly_limit=2 
 in our yum.conf), the
 279 kernel is removed and that triggers the removal of kmod-openafs-279.  
 That seems to trigger
 the removal of kmod-openafs and THAT triggers the removal of 
 kmod-openafs-358.
 
 All this seems to trigger the removal of openafs-client and that leads to 
 local afs-depend
 packages to be removed (usrlocalITbin).  The output from a yum upgrade 
 follows:
 
 # yum update kernel
 Loaded plugins: aliases, priorities, product-id, protectbase, 
 refresh-packagekit, subscription-manager
 Updating Red Hat repositories.
 1427 packages excluded due to repository priority protections
 0 packages excluded due to repository protections
 Setting up Update Process
 Resolving Dependencies
 -- Running transaction check
 --- Package kernel.x86_64 0:2.6.32-358.6.1.el6 will be installed
 -- Processing Dependency: kernel-firmware = 2.6.32-358.6.1.el6 for 
 package: kernel-2.6.32-358.6.1.el6.x86_64
 -- Running transaction check
 --- Package kernel-firmware.noarch 0:2.6.32-358.2.1.el6 will be updated
 --- Package kernel-firmware.noarch 0:2.6.32-358.6.1.el6 will be an update
 -- Finished Dependency Resolution
 -- Running transaction check
 --- Package kernel.x86_64 0:2.6.32-279.19.1.el6 will be erased
 -- Processing Dependency: kernel(do_settimeofday) = 0x5603cf43 for package: 
 kmod-openafs-279-1.6.2-0.144.sl6.279.x86_64
 -- Running transaction check
 --- Package kmod-openafs-279.x86_64 0:1.6.2-0.144.sl6.279 will be erased
 -- Processing Dependency: kmod-openafs-279 for package: 
 kmod-openafs-1.6.2-4.SL64.el6.noarch
 -- Running transaction check
 --- Package kmod-openafs.noarch 0:1.6.2-4.SL64.el6 will be erased
 -- Processing Dependency: openafs-kernel = 1.6 for package: 
 openafs-client-1.6.2-0.144.sl6.x86_64
 -- Running transaction check
 --- Package openafs-client.x86_64 0:1.6.2-0.144.sl6 will be erased
 -- Processing Dependency: openafs-client = 1.6 for package: 
 kmod-openafs-358-1.6.2-0.144.sl6.358.0.1.x86_64
 -- Processing Dependency: openafs-client for package: 
 usrlocalITbin-6.0-el6.bucs.1.noarch
 -- Running transaction check
 --- Package kmod-openafs-358.x86_64 0:1.6.2-0.144.sl6.358.0.1 will be erased
 --- Package usrlocalITbin.noarch 0:6.0-el6.bucs.1 will be erased
 -- Finished Dependency Resolution
 
 Dependencies Resolved
 
 
  Package Arch  Version   
  Repository   Size
 
 Installing:
  kernel  x86_642.6.32-358.6.1.el6
  sl-security  26 M
 Removing:
  kernel  x86_642.6.32-279.19.1.el6   
  @sl-security/6.2113 M
 Updating for dependencies:
  kernel-firmware noarch2.6.32-358.6.1.el6
  sl-security  11 M
 Removing for dependencies:
  kmod-openafsnoarch1.6.2-4.SL64.el6  
  @sl-security/6.20.0
  kmod-openafs-279x86_641.6.2-0.144.sl6.279   
  @sl-security/6.21.3 M
  kmod-openafs-358x86_64
 1.6.2-0.144.sl6.358.0.1@sl-security/6.21.3 M
  openafs-client  x86_641.6.2-0.144.sl6   
  @sl/6.2 2.4 M
  

Re: Problems with zfs on linux

[Stephan Wiesand]

Hello,

On Apr 19, 2013, at 18:12 , Olivier Mauras wrote:

 On 2013-04-19 17:29, Fabrice BOYRIE wrote:
 [...]
 Bigger one: problem with selinux
 When I mount zfs volume, I've the following errors:
 SELinux: initialized (dev zfs, type zfs), not configured for labeling
 and even root can't write on the disk
   
 I've modified selinux-policy srpm adding the following patch
 policy-zfs.patch 
 
 diff -Nur nsaserefpolicy/policy/modules/kernel/filesystem.te
 serefpolicy-3.7.19/policy/modules/kernel/filesystem.te
 --- aserefpolicy/policy/modules/kernel/filesystem.te   2010-04-13
 20:44:37.0 +0200
 +++ serefpolicy-3.7.19/policy/modules/kernel/filesystem.te   2013-04-19
 17:30:43.952120437 +0200
 @@ -21,6 +21,7 @@
  
  # Use xattrs for the following filesystem types.
  # Requires that a security xattr handler exist for the filesystem.
 +fs_use_xattr zfs gen_context(system_u:object_r:fs_t,s0);
  fs_use_xattr btrfs gen_context(system_u:object_r:fs_t,s0);
  fs_use_xattr encfs gen_context(system_u:object_r:fs_t,s0);
  fs_use_xattr ext2 gen_context(system_u:object_r:fs_t,s0);
 
 With this patch applied, selinux seems working (I can write and i can use
 chcon). But at next update, it will breaks. And I don't understand enough
 selinux to make a specific module.
 
   How solves this problems ?
 
 Thanks in advance
 
 Fabrice BOYRIE
 
 Hello Fabrice,
 
 While the patch is simple, the filesystem module is quite complicated and it 
 would require quite some work to make a standalone module only for ZFS.
 Sadly for now i think that it's simpler to patch the actual package than 
 anything else

depending on your definition of simple, mounting with fscontext= may 
actually be simpler. And it will work across policy updates.

 and as long as the upstream vendor doesn't explicitely support ZFS in their 
 SELinux rule, you/we'll have to continue use a patched package.

If you believe the above patch is sufficient (I don't quite get the Requires 
that a security xattr handler exist for the filesystem part), filing a BZ with 
TUV would probably make sense.

Regards,
Stephan

 Regards,
 Olivier
 
  

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: [SCIENTIFIC-LINUX-USERS] Samba4 rpm

[Stephan Wiesand]

On Feb 8, 2013, at 17:01 , Paul Robert Marino wrote:

 And just as an aside last I looked at systemd I didn't think it was fully 
 baked yet. Many of the init scripts were ported in a haphazard rushed manner 
 as a result they don't all work the way they really should. Also I question 
 some of the logic for servers. Some of the capabilities of systemd are great 
 for desktops but scary on a server. I go to great lengthens to ensure that if 
 someone plugs something in to one of my boxes nothing happens and now here is 
 a service control system built around the premise of doing exactly what I 
 want to prevent.
 Further more the CLI is clunky and not the most functional or initiative set 
 of commands for sysadmins who don't use X11.

I happen to like systemd. It feels a bit strange, but it's not bad at all. And 
with SL7, it's going to be upon us. If you have problems with it, file bugs 
against F18 *now* and against EL7 beta as soon as possible.

Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Solution - Installation of 5.8 breaks LibreOffice 3.5

[Stephan Wiesand]

On Jan 26, 2013, at 21:21 , Larry Linder wrote:

 On Saturday 26 January 2013 11:43 am, Larry Linder wrote:
 Down loaded SL 5.8, created DVD's and updated test system not a fresh
 install. The only causality is LibreOffice 3.5.
 When you open a LibreOffice file and try to edit a file the key board is
 disconnected.  Once you type to  a LOffice 3.5 document,  display gets
 dumb, and it eventually asks you to terminate the program.  Did SL 5.8
 install a new KDE for me?
 
 Everything else appears to work -
 
 I noticed that I now have KDE 4 as a desktop.
 
 Tried to install a new version of LibreOffice 3.6 to see if it would behave
 differently.   KDE complains that it cannot launch /usr/bin/libreOffice3.6
 and quits.
 Looked on Net and there is so much noise about KDE and this subject - Had
 to quit but never found a solution but KDE developer closed bug report.
 
 Is there a way to regress back to SL5.7 from SL 5.8 , that was on this
 system?
 
 Fortunately this is only one system and not many.
 
 Thanks
 Larry Linder
 
 You have to install an export into your .bashrc to get the problem resolved.
 For a global solution you could add it to /etc/bashrc.
 
 What was tried.
 export  OOO_FORCE_DESKTOP=noneWorks
 export OOO_FORCE_DESKTOP=kde4 does not work
 export OOO_FORCE_DESKTOP= kde4 libreofficedoes not work
 export OOO_FORCE_DESKTOP=kde4 soffice may work on open office
 
 used yum to remove all the openOffice and LibreOffice and reloaded 
 libreOffice3.5 that had previously worked and quit after system was updated 
 to SL 5.8.

Larry,

wherever your KDE is from, it's not from SL5. Nor is your LibreOffice.

All this is really simply off topic.

Regards,
Stephan

 libreOffice3.5 and libreOffice 3.6 both now work.
 The side effect of this export is that a window pops up asking for a driver.  
  
 Close it and it does not seem to effect operation.   
 
 This command must change the input stream for libreOffice but not the 
 interrupt.   The reason I say that is that once you type a character in any 
 kind of document KDE freezes and even a vi of .bashrc will not function till 
 libreOffice is terminated.
 
 Larry Linder

Re: C++11, the devtools package and PUIAS

[Stephan Wiesand]

Hi Jan,

On Jan 17, 2013, at 16:12 , Jan Kundrát wrote:

 Hi,
 I'd like to get a decent support for the C++11 language on SL6. RH ships the 
 devtoolset [1] package for just this purpose; the PUIAS repo [2] has the 
 RPMs. However, I am not much familiar with the PUIAS, so I would like to ask 
 here for references -- is that a repository which is known to be good?

I've cherry-picked packages from their repos and have been happy with those. 
There's always some trust involved when you install software from yet another 
source, but I don't consider this one higher risk than the others we draw from.

Back then when the EL clones had problems with the rebuild of nfs-utils in 6.1, 
PUIAS was bug-compatible with the others including SL ;-)

 Any plans to make devtools available as a part of the SL?

Connie said she was working on releasing them a while ago, but it seems that 
hasn't happened. Connie, any news?

I also managed to rebuild them myself, but the procedure was a bit odd and I 
would much rather use packages rebuilt by someone more experienced with that. 
Next time a user asks for them, and they're not available from SL, I'd probably 
pick them from PUIAS now.

Best regards,
Stephan


 With kind regards,
 Jan
 
 [1] https://access.redhat.com/knowledge/docs/Red_Hat_Developer_Toolset/
 [2] http://springdale.math.ias.edu/data/puias/DevToolset/6.3/

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: java vulnerability

[Stephan Wiesand]

On Jan 17, 2013, at 18:15 , Connie Sieh wrote:

 On Thu, 17 Jan 2013, Ken Teh wrote:
 
 What's the status of the java package that's installed on SL6x? 
 java-1.6.0-openjdk.  Is it vulnerable to this java security flaw that made 
 the national news this week?  Cyber is advising us to remove it but a lot of 
 packages depend on it.  The biggie is LibreOffice.
 
 I thought that the biggest issue was with Java 7 and not Java 6.


That's what I thought. In any case, removing the browser plugin (icedtea-web 
with openjdk) seems to be the most important step, and advisable wherever 
feasible. LibreOffice shouldn't depend on that.

A related question: Does anyone know whether openjdk6 will continue to be 
supported after the Oracle JDK6 end of service life?

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Security ERRATA Low: selinux-policy enhancement update on SL5.x, SL6.x i386/x86_64

[Stephan Wiesand]

This is the second selinux-policy update within a couple of days, and both were 
promoted from enhancement to security. That's a major pain in the rear. Why 
do these have to land on all systems?

- Stephan

On Nov 21, 2012, at 17:40 , Pat Riehecky wrote:

 Synopsis: Low: selinux-policy enhancement update
 Issue date: 2012-11-19
 
 This update adds the following enhancements:
 
 * An SELinux policy for openshift packages has been added
 
 This update has been placed in the security tree to avoid selinux
 related problems.
 
 
 SL6.x
 
 SRPMS:
 selinux-policy-3.7.19-155.el6_3.8.src.rpm
 
 i386:
 selinux-policy-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-doc-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-minimum-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-mls-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-targeted-3.7.19-155.el6_3.8.noarch.rpm
 
 
 x86_64:
 selinux-policy-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-doc-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-minimum-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-mls-3.7.19-155.el6_3.8.noarch.rpm
 selinux-policy-targeted-3.7.19-155.el6_3.8.noarch.rpm

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: User proc uses all RAM+swap = kernel panic - shouldn't OS not allow?

[Stephan Wiesand]

On Sep 13, 2012, at 18:59 , Chris Schanzle wrote:

 In our experience, if memory is allocated and never touched, it's like you 
 never allocated it at all (with respect to swap).  Allocated but untouched 
 pages will not be swapped.


Right, but they do count as committed. Thus, once overcommitment of virtual 
memory is turned off, such allocations may fail even if they would not have 
caused any problems. Unfortunately, the kernel can't know.

Having overcommitment turned off, and sufficient swap space (much of which is 
never used in this case), is probably the sane setup Winnie us looking for.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: User proc uses all RAM+swap = kernel panic - shouldn't OS not allow?

[Stephan Wiesand]

Hello Winnie,

On Sep 13, 2012, at 16:01 , Winnie Lacesso wrote:

 Several times over past few years I've seen user processes go mad 
 (programming error)  use all RAM, then all swap (as ganglia so vividly 
 shows), then the box ends up at a kernel panic.
 (Server OS is SL5.x 64-bit BTW)

we rarely see panics in these cases. The box just becomes unusable. Which 
effectively makes no difference though.

 What's puzzling is, shouldn't the OS by default not allow users to do 
 something bad enough to cause grief to the OS?
 
 Possibly some sort of tuning can fix this, but one expects that, out of 
 the box, this should not be needed, users just can't bring OS down.

There are plenty of ways to bring a *x system down, or make it deny service to 
users, by making mistakes in userland. Just think of the classic fork bomb.

 In the past on SL4 I did see oom come into play when box too 
 loaded (killing the mysqld process for instance)  am wondering
 why this isn't happening on SL5 with badly behaved user processes.

By any chance, were your SL4 systems mostly 32-bit, and your SL5 systems are 
mostly 64-bit? As much as I do advocate using 64-bit, I have to admit that the 
x86-64 kernel seems to be handling OOM situations much worse than x86 used to. 
And I think it started with SL3 already.

 Grateful for advice!


The one way I know about to reliably prevent these problems is to use syctl to 
change the value of vm.overcommit_ratio, and possibly adapt 
vm.overcommit_memory. Both are documented in proc(5).

The problem with this approach is that there's more and more software making 
very generous use of virtual address space without ever using what was 
allocated. The current Maple and Oracle's Java come to mind.

Having sufficient swap space does help. We used to set aside only 2GB for swap 
even on systems with much more RAM, because they weren't supposed to swap/page 
much at all. But it turns out that having the recommended amount makes systems 
much more resilient to memory hogs.

Hope this helps,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: SL 6.3: wrong permissions for infiniband devices

[Stephan Wiesand]

Hi Francesco,

On Aug 24, 2012, at 10:01 , Francesco Minafra wrote:

 seems that with a change made in upstream to the rdma package (no
 more udev rules)
 the infiniband devices get created by the kernel with the wrong permissions.
[...]
 This prevents non-root users from using the infiniband.
[...]
 See the discussion at:
 https://www.centos.org/modules/newbb/viewtopic.php?topic_id=38586
 
 Can someone with access to TUV bug archive check if this has been fixed 
 already?


let me do that for you... visit bugzilla.redhat.com ... enter rdma in the 
search field ... and in the result list, there is 
https://bugzilla.redhat.com/show_bug.cgi?id=834428

No special access required. But thanks for bringing this up - the issue may 
affect our site as well.

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: [SCIENTIFIC-LINUX-USERS] Password required in single-user mode?

[Stephan Wiesand]

Hi,

On Aug 23, 2012, at 15:18 , Pat Riehecky riehe...@fnal.gov wrote:

 On 08/18/2012 03:57 PM, David Sommerseth wrote:
 Hi,
 
 I've been running Scientific Linux since the 6.0 days, and single-user mode 
 have basically behaved how I have expected it those few times I needed it.  
 As I usually set up my boxes root accounts with passwords disabled, 
 single-user mode needs to be without root password.
 
 Today, after having upgraded from 6.3, I needed to enter single-user mode at 
 boot.  And I was asked for a password at boot time.  Is this change 
 intentional?
 
 # cat /etc/redhat-release
 Scientific Linux release 6.3 (Carbon)
 # rpm -qa | grep -i sl_password_for_singleuser | wc -l
 0
 # grep SINGLE /etc/sysconfig/init
 SINGLE=/sbin/sushell
 
 If this change was intentional, how can I go back to the old behaviour?  I 
 double checked the behaviour with an old VM with SL6.1, and that behaves as 
 expected.
 
 
 kind regards,
 
 David Sommerseth
 
 Hi David,
 
 The behavior shouldn't have changed.  You've provided just about all the 
 relevant details in your email, so there isn't really anything I want to ask 
 for more information.

but I do :-) How did you end up in single user mode? Did you pass single to 
the kernel, or was there a problem with a filesystem and the system decided to 
drop you to a shell?

In the latter case, rc.sysinit in SL  6.3 unconditionally ran sulogin, which 
may explain some of the complaints we had on this list, and this case.

As of 6.3 the rcS-emergency task is started instead. It still defaults to 
sulogin, but it's configurable. An EMERGENCY=/sbin/sushell in 
/etc/sysconfig/init should do.

Regards,
Stephan

 Can I have you try setting /etc/sysconfig/init = SINGLE to /sbin/sulogin 
 rebooting and setting it back to /sbin/sushell?  Perhaps something got 
 'stuck' wrong
 
 /sbin/sushell is a shell script, so can I have you verify its contents?  Mine 
 looks like:
 
 #!/bin/bash
 
 [ -z $SUSHELL ]  SUSHELL=/bin/bash
 
 exec $SUSHELL
 
 
 Pat

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15732 Zeuthen, Germany

Re: server crashing out of memory

[Stephan Wiesand]

On Jul 17, 2012, at 19:22 , Orion Poplawski wrote:

 Our SL6.2 KVM and nfs/backup server has been crashing frequently recently 
 (starting around Fri 13th - yikes!) with Kernel panic - Out of memory and no 
 killable processes.  The server has 48GB ram, 2GB swap, only about 15GB 
 dedicated to VM guests.  I've tried bumping up vm.min_free_kbytes to 262144 
 to no avail.  Nothing strange is getting written to the logs before the crash.
 
 Happening with both 2.6.32-220.23.1 and 2.6.32-279.1.1.
 
 Anyone else seeing this?

Not on our KVM servers (which don't have any other duties though), which have 
been running -220.23.1 for three weeks.

  Any other ideas?

Is swap space sufficient?

Have you modified vm.overcommit_* ? Doing so may help turning the panics into 
allocation failures that can be handled.

Do any slab pools keep growing, to an unusual size?

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Problems installing Open64 on SL 6.2

[Stephan Wiesand]

Hi,

On Jun 30, 2012, at 19:05 , Jeff Layton wrote:

 Hello,
 
 I'm trying to install Open64 on my SL6.2 system and I keep getting failed 
 dependencies which I'm having a hard time deciphering. I'm trying the AMD 
 version of Open64 (4.5.1) but I've also tried Open64 5.0 from the Open64 
 site. Both give me the same failed dependencies:

despite the name and the rpm %arch, this seems to be 32-bit software, and you 
probably haven't installed the required 32-bit compatibility packages. Try yum 
install instead of rpm -i.

Regards,
Stephan
 
 
 [root@test1 RPMS]# rpm -ivh x86_open64-4.5.1-1.x86_64.rpm
 error: Failed dependencies:
ld-linux.so.2 is needed by x86_open64-4.5.1-1.x86_64
ld-linux.so.2(GLIBC_2.3) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6 is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.0) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.1) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.11) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.1.3) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.2) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.2.3) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.2.4) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.3) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.3.4) is needed by x86_open64-4.5.1-1.x86_64
libc.so.6(GLIBC_2.7) is needed by x86_open64-4.5.1-1.x86_64
libdl.so.2 is needed by x86_open64-4.5.1-1.x86_64
libdl.so.2(GLIBC_2.0) is needed by x86_open64-4.5.1-1.x86_64
libdl.so.2(GLIBC_2.1) is needed by x86_open64-4.5.1-1.x86_64
libm.so.6 is needed by x86_open64-4.5.1-1.x86_64
libm.so.6(GLIBC_2.0) is needed by x86_open64-4.5.1-1.x86_64
libm.so.6(GLIBC_2.1) is needed by x86_open64-4.5.1-1.x86_64
libpthread.so.0 is needed by x86_open64-4.5.1-1.x86_64
libpthread.so.0(GLIBC_2.0) is needed by x86_open64-4.5.1-1.x86_64
libpthread.so.0(GLIBC_2.1) is needed by x86_open64-4.5.1-1.x86_64
libpthread.so.0(GLIBC_2.2) is needed by x86_open64-4.5.1-1.x86_64
libpthread.so.0(GLIBC_2.3.2) is needed by x86_open64-4.5.1-1.x86_64
libpthread.so.0(GLIBC_2.3.4) is needed by x86_open64-4.5.1-1.x86_64
 
 
 Any help is greatly appreciated!
 
 Thanks!
 
 Jeff

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: emacs is not working

[Stephan Wiesand]

Hello,

On May 14, 2012, at 11:50 , vivek chalotra wrote:

 Hello all,
 
 I have just installed slc6.2 in my hp laptop but its giving an error while 
 opening emacs. pls look into the matter. below is the error:
 
 [nidhi@localhost Desktop]$ emacs chap1
 emacs: error while loading shared libraries: libotf.so.0: cannot open shared 
 object file: No such file or directory
 [nidhi@localhost Desktop]$ emacs chap1
 [1] 3637
 [nidhi@localhost Desktop]$ emacs: error while loading shared libraries: 
 libotf.so.0: cannot open shared object file: No such file or directory
 ^C
 [1]+  Exit 127emacs chap1
 [nidhi@localhost Desktop]$ emacs
 emacs: error while loading shared libraries: libotf.so.0: cannot open shared 
 object file: No such file or directory
 [nidhi@localhost Desktop]$ emacs
 emacs   emacs-23.1  emacs-23.1-nox  emacsclient
 [nidhi@localhost Desktop]$ emacs test
 emacs: error while loading shared libraries: libotf.so.0: cannot open shared 
 object file: No such file or directory

that's upstream Bug #768457, which is not public for reasons beyond me. As of 
6.2, openmpi provides libotf.so.0 even though the libs are not in a location 
searched by default (and probably wouldn't work with emacs anyway).

The fix is to yum install libotf.

Regards
Stephan

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: File size diff on local disk vs NFS share

[Stephan Wiesand]

On May 3, 2012, at 20:11 , Florian Philipp wrote:

 Am 03.05.2012 20:01, schrieb aurfalien:
 Hi all,
 
 I never really paid attention to this but a file on an NFS mount is
 showing 64M in size, but when copying the file to a local drive, it
 shows 2.5MB in size.
 
 My NFS server is hardware Raided with a volume stripe size of 128K
 were the volume size is 20TB, my local disk is about 500GB.
 
 Is this due to my stripe size?
 
 Nuggets are appreciated.
 
 - aurf
 
 Is it a sparse file? What exact command did you use to calculate the
 file size?

A sparse file will grow when copied, not shrink.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Security ERRATA Low: ImageMagick on SL5.x i386/x86_64

[Stephan Wiesand]

On Mar 21, 2012, at 22:24 , riehe...@fnal.gov wrote:

 Synopsis:Low: ImageMagick security and bug fix update
 Issue Date:  2012-02-21
 CVE Numbers: CVE-2010-4167

heads up: the issue described in 
https://www.centos.org/modules/newbb/viewtopic.php?topic_id=36328 caused us 
some trouble.

If you have the suspicion that you users rely on convert for converting 
postscript to image formats (it seems quite common), you may want to hold this 
one or roll back.

Regards,
Stephan

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: [SCIENTIFIC-LINUX-USERS] Security ERRATA Low: ImageMagick on SL5.x i386/x86_64

[Stephan Wiesand]

On Mar 23, 2012, at 14:26 , Pat Riehecky wrote:

 On 03/23/2012 08:12 AM, Stephan Wiesand wrote:
 On Mar 21, 2012, at 22:24 , riehe...@fnal.gov wrote:
 
 Synopsis:Low: ImageMagick security and bug fix update
 Issue Date:  2012-02-21
 CVE Numbers: CVE-2010-4167
 heads up: the issue described in 
 https://www.centos.org/modules/newbb/viewtopic.php?topic_id=36328 caused us 
 some trouble.
 
 If you have the suspicion that you users rely on convert for converting 
 postscript to image formats (it seems quite common), you may want to hold 
 this one or roll back.
 
 Regards,
  Stephan
 
 
 There does not seem to be an associated upstream bug, at least in my 
 searching.  Can someone confirm I didn't overlook it (and possibly get a bug 
 opened upstream to notify them of the issue)?

I can't find it either. Given that the issue is discussed in the knowledge 
base, the bug probably exists but isn't public.

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: pxeboot kernel problem (additional)

[Stephan Wiesand]

On Mar 5, 2012, at 15:01 , Stephen Berg (Contractor) wrote:

 On 03/05/2012 07:37 AM, Stephen Berg (Contractor) wrote:
 I'm testing a way to install/upgrade some remote systems.  What I'm doing is 
 hand jamming a change in /boot/grub/grub.conf to point to /boot/vmlinuz and 
 /boot/initrd.img from the /image/pxeboot directory off the install DVD, 
 tried both 6.1 and 6.2.  Both files have been copied to /boot on the test 
 server.
 
 When I boot to this kernel/image combination I use the following boot 
 parameters to get a psuedo netboot/pxeboot installation started:
 
 ks=nfs:IPADDRESS:PATH to kickstart ksdevice=link vnc vncpassword=PW
 
 The problem I'm running into is that the ksdevice parameter seems to be 
 getting ignored.  I've tried ksdevice=ethX with the appropriate network 
 interface name, ksdevice=link and ksdevice=MACADDR but the system 
 consistently stops at the screen asking me to choose which interface to use.
 
 I tried a similar method using a Fedora 15 vmlinuz/initrd.img and 
 ksdevice=link and it worked as expected.  Is there some magic I'm missing or 
 is that ksdevice parameter not functioning properly from the install DVD's 
 copies of vmlinuz/initrd.img?
 
 I'd really like to get this working to help with some remote system 
 installations.
 
 
 Forgot to mention this system has two NIC's, both enabled but only one has an 
 active link.  In SL6.x the active link shows up as eth1, during the test with 
 Fedora 15 that I mentioned the active link came up as eth0.

ksdevice=link seems to work for us when reinstalling SL5 systems with SL6. But 
maybe that's just because we have installed all systems with pci=bfsort since 
the SL4 times, making device numbering consistent across kernel versions (and 
we use only eth0 for PXE). Using pci=[no]bfsort, it should always be possible 
to make the interface in use primary one. But then I'm not sure what 
Networkmanager or the new parallel udev device discovery may be doing...

And since 6.1, some (recent Dell) servers have biosdevname enabled by default. 
Which makes the first onboard interface em1, for example :-) The advantage is 
that this will always match the physical socket labelled 1 on the server.

 I can get to this system to test pretty easily, the systems I want to do this 
 on if I can fix this issue I can't.  No iDrac type interface, I can remote in 
 and reboot them, but there's no easy way to select a network boot without 
 physically going to the machine and hooking up a keyboard/monitor.  And 
 that's what I'm trying to avoid if possible.

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: Informative Announcement for SL4.x End Of Life Today!

[Stephan Wiesand]

Goodbye SL4, r.i.p.

This was the first SL release supported over a full seven years life cycle, I 
believe.

Thanks so much to those who made it happen (including those no longer part of 
the project and unlikely to read this) and provided a decent, reliable service 
over such a long period.

- Stephan

On Feb 29, 2012, at 16:29 , Pat Riehecky wrote:

 In accordance with our Upstream Vendor's Errata Support Policy, the
 regular life-cycle of Scientific Linux 4 will end today, the last day
 of February 2012.
 
 After today's date, The Upstream Vendor will discontinue their regular
 update services.  We must follow them in this matter.  Therefore, new
 bug fix, enhancement, and security errata updates will no longer be
 available for Scientific Linux 4 after the End of Life date.  They will
 not be providing updates and so we cannot provide them.
 
 Anyone still running production workloads on Scientific Linux 4 should
 be aware that after today no updates of any kind will be published.
 Because of this, we hope everyone has completed their migration to
 Scientific Linux 5 or Scientific Linux 6 by now.
 
 The existing Scientific Linux 4 directories will be archived in April
 2012.  People wishing to obtain a personal archive of Scientific
 Linux 4 can do so during that window.
 
 Again, this is a reminder of the end of life for Scientific Linux 4.
 
 - Scientific Linux Development Team

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: server going slow

[Stephan Wiesand]

Hi Sunil,

On Feb 10, 2012, at 8:02 , Dr. Sunil M. Dogra wrote:

 One of my machine is going slow, I even canot do ls on terminal, when I see 
 the logs of I see the following lines, 
 Is it due to this 

you're having problems with your network connection to CERN. This will block 
any access to /afs/cern.ch at times. It should not affect the system in 
general, but having anything below /afs/cern.ch in environment variables like 
PATH or LD_LIBRARY_PATH will hang pretty much any command, for example.

Regards,
Stephan

 Feb 10 12:25:49 tier-3 kernel: afs: file server 137.138.144.23 in cell 
 cern.ch is back up (multi-homed address; other same-host interfaces may still 
 be down)
 Feb 10 12:28:29 tier-3 kernel: afs: Lost contact with file server 
 137.138.128.203 in cell cern.ch (all multi-homed ip addresses down for the 
 server)
 Feb 10 12:28:29 tier-3 kernel: afs: Lost contact with file server 
 137.138.128.203 in cell cern.ch (all multi-homed ip addresses down for the 
 server)
 Feb 10 12:28:56 tier-3 kernel: afs: file server 137.138.128.203 in cell 
 cern.ch is back up (multi-homed address; other same-host interfaces may still 
 be down)
 Feb 10 12:28:56 tier-3 kernel: afs: file server 137.138.128.203 in cell 
 cern.ch is back up (multi-homed address; other same-host interfaces may still 
 be down)
 Feb 10 12:29:26 tier-3 kernel: afs: Lost contact with file server 
 137.138.246.50 in cell cern.ch (all multi-homed ip addresses down for the 
 server)
 Feb 10 12:29:26 tier-3 kernel: afs: Lost contact with file server 
 137.138.246.50 in cell cern.ch (all multi-homed ip addresses down for the 
 server)
 Feb 10 12:30:04 tier-3 kernel: afs: file server 137.138.246.50 in cell 
 cern.ch is back up (multi-homed address; other same-host interfaces may still 
 be down)
 Feb 10 12:30:04 tier-3 kernel: afs: file server 137.138.246.50 in cell 
 cern.ch is back up (multi-homed address; other same-host interfaces may still 
 be down)
 Feb 10 12:30:24 tier-3 kernel: afs: Lost contact with file server 
 137.138.246.51 in cell cern.ch (all multi-homed ip addresses down for the 
 server)

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: machine hangs with kernel-2.6.32-220

[Stephan Wiesand]

On Feb 9, 2012, at 18:30 , Orion Poplawski wrote:

 Is anyone else seeing machines lockup with kernel-2.6.32-220?  On one machine 
 (Dell PowerEdge SC1435 Dual Opteron 2384) we've seen it lockup pretty 
 consistently.  Moving back to 2.6.32-131.21.1.el6.x86_64 appears to have 
 resolved it.

No hangs observed with the SL6.2 kernels. We never deployed -220 though, only 
-220.2.1 and -220.4.1. And we don't have any SL6 systems running on Opterons. 
But a wide range of systems, from Pentium D desktops to dual Westmere compute 
nodes.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: moved mysql - now broken - kind of urgent

[Stephan Wiesand]

Hi Philipp,

On Jan 28, 2012, at 21:31 , Philipp Schiffer wrote:

 Populating my mysql on SL6.1 with biological data I ran into my primary SSDs 
 limit and thus tried to move the location to another disk (HD). So I sudo cp 
 -R'ed everything from /var/lib/mysql to /data/mysqldata and changed the 
 permissions accordingly
 drwxr-xr-x. 10 mysql   mysql  4096 Jan 28 12:20 mysqldata
 and inside mysqldata
 drwx--. 2 mysql mysql  4096 Jan 28 12:20 b2g
 -rw-r-. 1 mysql mysql 178257920 Jan 28 20:50 ibdata1
 -rw-r-. 1 mysql mysql   5242880 Jan 28 20:50 ib_logfile0
 -rw-r-. 1 mysql mysql   5242880 Jan 28 20:50 ib_logfile1
 drwx--. 2 mysql mysql  4096 Jan 28 12:20 mysql
 drwx--. 2 mysql mysql  4096 Jan 28 12:20 orthotest
 Then went on to modify /etc/my.cnf to
 [mysqld]
 #datadir=/var/lib/mysql
 datadir=/data/mysqldata
 #socket=/var/lib/mysql/mysql.sock
 socket=/data/mysqldata/mysql.sock
 user=mysql
 # Disabling symbolic-links is recommended to prevent assorted security risks
 symbolic-links=0
 #basedir=/data
 
 [mysqld_safe]
 log-error=/var/log/mysqld.log
 pid-file=/var/run/mysqld/mysqld.pid.
 
 Having stopped the server before doing all this I now started again, but it 
 failed to work. less /var/log/mysqld.log says:
 120128 21:03:08 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid 
 ended
 120128 21:09:06 mysqld_safe Starting mysqld daemon with databases from 
 /data/mysqldata
 120128 21:09:06 [Warning] Can't create test file 
 /data/mysqldata/superrechenknecht.lower-test
 120128 21:09:06 [Warning] Can't create test file 
 /data/mysqldata/superrechenknecht.lower-test
 ^G/usr/libexec/mysqld: Can't change dir to '/data/mysqldata/' (Errcode: 13)
 120128 21:09:06 [ERROR] Aborting
 
 120128 21:09:06 [Note] /usr/libexec/mysqld: Shutdown complete
 
 120128 21:09:06 mysqld_safe mysqld from pid file /var/run/mysqld/mysqld.pid 
 ended
 
 Tried to set a symbolic link from /var/lib/mysql - /data/mysqldata but that 
 failed to work as well (actually why can't I change the ownership of the link 
 from root:root to mysql:mysql?). Finally reading up on SElinux I edited my 
 new mysql directory /data/mysqldata to look the same as the original mysql 
 folder:
 drwxr-xr-x. mysql   mysql   system_u:object_r:mysqld_db_t:s0 mysqldata
 drwxr-xr-x. mysql   mysql   system_u:object_r:mysqld_db_t:s0 mysql_old
 
 and within mysqldata:
 drwx--. mysql mysql system_u:object_r:mysqld_db_t:s0 b2g
 -rw-r-. mysql mysql unconfined_u:object_r:mysqld_db_t:s0 ibdata1
 -rw-r-. mysql mysql unconfined_u:object_r:mysqld_db_t:s0 ib_logfile0
 -rw-r-. mysql mysql unconfined_u:object_r:mysqld_db_t:s0 ib_logfile1
 drwx--. mysql mysql unconfined_u:object_r:mysqld_db_t:s0 mysql
 drwx--. mysql mysql unconfined_u:object_r:mysqld_db_t:s0 orthotest
 drwx--. mysql mysql system_u:object_r:mysqld_db_t:s0 sknstuff_insp
 
 However, even that does leads to the same error messages in the log file and 
 I am really at a loss. Any help with this would be highly appreciated 
 (especially as I need a working db quite urgently).

Try chcon -t var_t /data ?

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: SL6.1, openmpi and infiniband!

[Stephan Wiesand]

Hi Sven,

On Jan 27, 2012, at 14:20 , Sven Sternberger wrote:

 Hello!
 
 after migrating a infiniband cluster to SL6.1 we noticed
 that the MPI jobs can't use openib anymore. So we checked
 first with perftest but verbs is working. Even IPOIB
 works. 
 
 So we looked at the openmpi package, and observe that our 
 self compiled openmpi  package is still working as expected.
 
 So we would really prefer to go with the SL openmpi package,
 has anybody experienced the same problems? Is there any known SL6.1
 problem with openmpi??
 
 Kernel: 2.6.32-220.2.1.el6.x86_64

The latest update kernels come from SL 6.2 (NB you should probably be using 
-220.4.1 now ;-). Have you tried the openmpi packages from 6.2 beta?

Viele Grüße
Stephan

 openmpi-1.4.3-1.1.el6.x86_64
 
 
 best regards!
 
 sven

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: SL6 pvops guest on SL5 Xen host

[Stephan Wiesand]

On Jan 18, 2012, at 22:09 , Steven Timm wrote:

 Sorry for the fermi-ized url's below--but if anyone
 has been able to make SL6 on SL5 work with the corresponding
 vanilla SL, please let me know too.

Works fine here. But maybe that's just due to of a couple of compatibility 
symlinks we have:

EFI - os/EFI
GPL - os/GPL
Packages - os/Packages
SL - .
images - os/images
repodata - os/repodata

And as the location, we specify http://.../SL/6.1/x86_64 just like we would for 
SL5.

Regards,
Stephan

 
 (SL6 guest on SL6 host works fine with virt-install).
 
 Thanks
 
 Steve Timm
 
 
 On Wed, 18 Jan 2012, Steven Timm wrote:
 
 Has anyone yet successfully been able to install a SL6 guest
 on a SL5 Xen host?  according to TUV it is supposed to be supported
 but I am having difficulty figuring out what the location
 URL should be for SL6.
 
 In the SL5 on SL5 case I just use the same url that
 I would use for a normal http install, namely
 http://linux.fnal.gov/linux/slf57/x86_64/sites/Fermi/
 
 the SL6 boot image seems to indicate that the corresponding URL is
 
 http://linux1.fnal.gov/linux/fermi/slf6.1/x86_64/os/
 
 But when I try to do that with virt-install I get the error:
 
 
 ValueError:  Could not find an installable distribution at 
 http://linux1.fnal.gov/linux/fermi/slf6.1/x86_64/os/
 
 
 The code in question, DistroManager.py has code for RHEL4 and RHEL5
 as well as sl distro but no difference between version 5 and 6.
 Looks like it is trying to get ...images/xen/initrd.img and 
 images/xen/vmlinuz, which are there.
 
 Is there a newer version of virt-install out there that might
 be able to handle the 6-on-5 case?
 
 Steve Timm

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: Installing glibc.i686 on a 64 bit system?

[Stephan Wiesand]

On Jan 7, 2012, at 15:00 , Steven Haigh wrote:

 I'm hoping someone will either have a workaround or a suggestion for me.
 
 I'm trying to get arcconf and the firmware updater for an IBM SAS RAID 
 adapter. Sadly it seems the binaries that IBM supply are 32bit only and 
 require /lib/ld-linux.so.2
 
 So, I figure just to try 'yum install glibc.i686' - however that doesn't go 
 according to plan:
 
 # yum install glibc.i686
 Loaded plugins: fastestmirror
 Loading mirror speeds from cached hostfile
 * sl6x: mirror.aarnet.edu.au
 * sl6x-security: mirror.aarnet.edu.au
 Setting up Install Process
 Resolving Dependencies
 -- Running transaction check
 --- Package glibc.i686 0:2.12-1.47.el6 will be installed
 -- Processing Dependency: libfreebl3.so for package: glibc-2.12-1.47.el6.i686
 -- Processing Dependency: libfreebl3.so(NSSRAWHASH_3.12.3) for package: 
 glibc-2.12-1.47.el6.i686
 -- Running transaction check
 --- Package nss-softokn-freebl.i686 0:3.12.9-3.el6 will be installed
 -- Finished Dependency Resolution
 Error: Protected multilib versions: nss-softokn-freebl-3.12.9-3.el6.i686 != 
 nss-softokn-freebl-3.12.9-11.el6.x86_64
 You could try using --skip-broken to work around the problem
 You could try running: rpm -Va --nofiles --nodigest
 
 Am I attempting to do this the wrong way or is something else playing up here?

You're running 6.2 beta? Try using that repo instead of 6x, which points to 6.1.

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: Repo update error

[Stephan Wiesand]

On Dec 12, 2011, at 10:49 , jdow wrote:

 
 YUM - security
 
 Error: Package: icewm-1.3.7-1.el6.x86_64 (epel)
   Requires: bluecurve-icon-theme
 You could try using --skip-broken to work around the problem
 You could try running: rpm -Va --nofiles --nodigest
 
 I installed it to see what it was like. Then I never deleted it. It seems
 to need something not present.
 
 {^_^}

EPEL bug?

Re: SL6.1 installer requires internet connection?

[Stephan Wiesand]

On Nov 24, 2011, at 17:39 , menny_hambur...@dell.com wrote:

 Try adding the following dummy network to the Kickstart 
 
 Network --activate --device eth0 ip=10.0.0.1 --netmask=255.255.255.0
 
 This works for me when I have biosdevname=0 in the grub line (otherwise you 
 will get weird device names other than ethX).

Hmm, weren't those invented and pushed by a colleague of yours? ;-)

NB I like them. A lot. They're predictable, and guaranteed not to change across 
reboots. They can cause problems with broken software though - like certain 
license managers for really expensive software that know that a system can 
only have network devices called ethX or vmnetX.

- Stephan

 
 -Original Message-
 From: owner-scientific-linux-us...@listserv.fnal.gov 
 [mailto:owner-scientific-linux-us...@listserv.fnal.gov] On Behalf Of Connie 
 Sieh
 Sent: 23 November, 2011 01:46
 To: Konstantin Olchanski
 Cc: SCIENTIFIC-LINUX-USERS@LISTSERV.fnal.gov
 Subject: Re: SL6.1 installer requires internet connection?
 
 On Tue, 22 Nov 2011, Konstantin Olchanski wrote:
 
 On Tue, Nov 22, 2011 at 04:09:15PM -0600, Connie Sieh wrote:
 On Tue, 22 Nov 2011, Konstantin Olchanski wrote:
 
 Hi, SL users - I have constructed an SL6 installer USB disk and it works
 but after asking all the questions, right before starting to install 
 packages,
 it stops and requests that there be an active network connection. The 
 installation
 does not seem to proceed until the machine acquires an internet connection.
 
 Why is this?
 
 How did you make the USB image?
 
 
 Same as any bootable USB disk. Partition the USB disk, mkfs, rsync bootable
 filesystem contents, install boot loader.
 
 As follows:
 
 1) Partition table looks like this:
 
 Disk /dev/sdc: 7996 MB, 7996440576 bytes
 224 heads, 56 sectors/track, 1245 cylinders
 Units = cylinders of 12544 * 512 = 6422528 bytes
 Sector size (logical/physical): 512 bytes / 512 bytes
 I/O size (minimum/optimal): 512 bytes / 512 bytes
 Disk identifier: 0x
 
  Device Boot  Start End  Blocks   Id  System
 /dev/sdc1   *   11245 7808612   83  Linux
 
 NOTE: you need to repartition with -H224 -S56 (or whatever) because
 I see some USB disks come with random heads and sectors settings.
 
 NOTE2: note boot flag
 
 2) mke2fs -j /dev/sdc1 as normal, mount /dev/sdc1 /mnt/dst
 
 3) rsync -av /triumfcs/mirror/SL/6.1/x64_64/os /mnt/dst (same contents as 
 http://ftp.scientificlinux.org/linux/scientific/6.1/x86_64/os)
 
 4) directory Packages is probably not needed
 
 5) rsync -av .../SL-61-x86_64-2011-11-09-Everything-DVD1.iso 
 .../SL-61-x86_64-2011-07-27-Everything-DVD2.iso /mnt/dst
 
 
 So you should have a /images directory on your usb drive.  Copy the 
 /images/updates.img and /images/product.img from 
 .../SL-61-x86_64-2011-11-09-Everything-DVD1.iso to /images directory of 
 the usb drive.
 
 6) cd /mnt/dst, setup extlinux (since the original uses isolinux). I use 
 extlinux, mbr.bin, menu.c32 from extlinux-3.86.
 
 6a) cat mbr.bin  /dev/sdc, ./extlinux -i . (notice the dot)
 6b) extlinux.conf is a copy of isolinux.cfg with vesamenu.c32 replaced by 
 menu.c32 from extlinux-3.86
 
 7) umount /dev/sdc, try to boot it.
 
 I am making an SL6 installer for use on machines located where internet
 access is physically impossible, so is there a solution or workaround?
 
 
 There is a solution to this but I need to know the answer to the
 above question.
 
 
 Ideally, I would like the installer run from the Packages directory rather
 than the DVD iso images.
 
 
 
 
 -Connie Sieh

-- 
Stephan Wiesand
DESY - DV -
Platanenallee 6
15738 Zeuthen, Germany

Re: console slowness in sl6.1

[Stephan Wiesand]

On Nov 19, 2011, at 22:22 , Yi Ding wrote:

 I do have a i7-920 (Nehalem) processor in my machine, but I haven't
 seen the tsc unstable issue.

That shows up much less frequently than than the hangs. Trying the workaround 
is probably a good idea in your case, and extremely unlikely to break anything 
or make things worse. intel_idle.max_cstate=1 works fine for us, at the cost  
of some waste of electrical power. If this helps in your case, please consider 
adding data to the BZ.

Cheers,
Stephan

 
 Thanks,
 Yi
 
 On Fri, Nov 18, 2011 at 11:54 AM, Stephan Wiesand
 stephan.wies...@desy.de wrote:
 On Nov 18, 2011, at 18:44 , Fabrizio Giordano wrote:
 
 I get the same behaviour on my Scientific Linux 6.0 (kernel 2.6.32): my 
 console becomes terribly slow for about 5 minutes. This is what I read when 
 I call 'dmesg':
 
 Clocksource tsc unstable (delta = 26380341 ns)
 Switching to clocksource hpet
 
 It's not the first time while following this thread that I wonder whether 
 this is yet another manifestation of the Nehalem deep C states problem. 
 See https://bugzilla.redhat.com/show_bug.cgi?id=710265
 
 Cheers,
   Stephan
 
 
 That delta is suspiciously 5 minute long...
 
 -Original Message-
 From: owner-scientific-linux-us...@listserv.fnal.gov 
 [mailto:owner-scientific-linux-us...@listserv.fnal.gov] On Behalf Of 
 Turtaut Geoffroy
 Sent: Thursday, November 17, 2011 11:40 PM
 To: Gianluca Varenni; SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Subject: RE: console slowness in sl6.1
 
 Yes
 
 
 -Message d'origine-
 De : Gianluca Varenni [mailto:gianluca.vare...@riverbed.com]
 Envoyé : vendredi 18 novembre 2011 08:38 À : Turtaut Geoffroy; 
 SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Objet : RE: console slowness in sl6.1
 
 Does it happen if you run in purely text mode too (runlevel 3)? I've 
 noticed something similar (several seconds), but totally sporadically. Not 
 much runs on my machine, pretty much the a bare minimal text installation.
 
 GV
 
 -Original Message-
 From: Turtaut Geoffroy [mailto:geoffroy.turt...@dassault-aviation.com]
 Sent: Thursday, November 17, 2011 11:35 PM
 To: Gianluca Varenni; SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Subject: RE: console slowness in sl6.1
 
 When we hit a key, it takes 1 or 2 seconds (or more).
 Applications are slow, Ctrl+Alt+F2 can take 30 seconds, ps -ef 10 seconds, 
 df, ...
 
 The problems appears at random intervals, for 5 minutes.
 
 Just before and just after, no problem
 
 The problem appears on :
 Standalone system (no name services, dns, ldap, no network, ..) Network 
 workstation (dns, ldap, ...)
 
 Geoffroy
 
 
 -Message d'origine-
 De : Gianluca Varenni [mailto:gianluca.vare...@riverbed.com]
 Envoyé : vendredi 18 novembre 2011 08:26 À : Turtaut Geoffroy; 
 SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Objet : RE: console slowness in sl6.1
 
 What happens when it's extremely slow? Like you hit a key and it's not 
 echoed on the screen for a long time (seconds)?
 
 Have a nice day
 GV
 
 -Original Message-
 From: owner-scientific-linux-us...@listserv.fnal.gov 
 [mailto:owner-scientific-linux-us...@listserv.fnal.gov] On Behalf Of 
 Turtaut Geoffroy
 Sent: Thursday, November 17, 2011 11:18 PM
 To: SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Subject: Re: console slowness in sl6.1
 
 Hi,
 
 We have the same issue since we use RHEL 6.X/SL 6.X
 
 Topic : HP Z400 system very slow at ramdom times for 5 minutes (october 
 2011)
 
 We use HP workstations (XW4600, Z200, Z400, ...), VM, DELL laptops and the 
 problem is only present on Z400 (xeon) et an HP pavilion (corei7).
 
 The problem is not present if we use 2.6.32 kernel from kernel.org.
 
 We have a case on RHN ...
 
 Last RH comment was :
 We have analysed the results and see that there is nothing waiting on I/O 
 but there is a high CPU usage and large run queue.
 
 Geoffroy Turtaut
 
 --
 Stephan Wiesand
 DESY -DV-
 Platanenenallee 6
 15738 Zeuthen, Germany
 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: console slowness in sl6.1

[Stephan Wiesand]

On Nov 18, 2011, at 18:44 , Fabrizio Giordano wrote:

 I get the same behaviour on my Scientific Linux 6.0 (kernel 2.6.32): my 
 console becomes terribly slow for about 5 minutes. This is what I read when I 
 call 'dmesg':
 
 Clocksource tsc unstable (delta = 26380341 ns)
 Switching to clocksource hpet

It's not the first time while following this thread that I wonder whether this 
is yet another manifestation of the Nehalem deep C states problem. See 
https://bugzilla.redhat.com/show_bug.cgi?id=710265

Cheers,
Stephan

 
 That delta is suspiciously 5 minute long...
 
 -Original Message-
 From: owner-scientific-linux-us...@listserv.fnal.gov 
 [mailto:owner-scientific-linux-us...@listserv.fnal.gov] On Behalf Of Turtaut 
 Geoffroy
 Sent: Thursday, November 17, 2011 11:40 PM
 To: Gianluca Varenni; SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Subject: RE: console slowness in sl6.1
 
 Yes
 
 
 -Message d'origine-
 De : Gianluca Varenni [mailto:gianluca.vare...@riverbed.com]
 Envoyé : vendredi 18 novembre 2011 08:38 À : Turtaut Geoffroy; 
 SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Objet : RE: console slowness in sl6.1
 
 Does it happen if you run in purely text mode too (runlevel 3)? I've noticed 
 something similar (several seconds), but totally sporadically. Not much runs 
 on my machine, pretty much the a bare minimal text installation.
 
 GV
 
 -Original Message-
 From: Turtaut Geoffroy [mailto:geoffroy.turt...@dassault-aviation.com]
 Sent: Thursday, November 17, 2011 11:35 PM
 To: Gianluca Varenni; SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Subject: RE: console slowness in sl6.1
 
 When we hit a key, it takes 1 or 2 seconds (or more).
 Applications are slow, Ctrl+Alt+F2 can take 30 seconds, ps -ef 10 seconds, 
 df, ...
 
 The problems appears at random intervals, for 5 minutes.
 
 Just before and just after, no problem
 
 The problem appears on :
 Standalone system (no name services, dns, ldap, no network, ..) Network 
 workstation (dns, ldap, ...)
 
 Geoffroy
 
 
 -Message d'origine-
 De : Gianluca Varenni [mailto:gianluca.vare...@riverbed.com]
 Envoyé : vendredi 18 novembre 2011 08:26 À : Turtaut Geoffroy; 
 SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Objet : RE: console slowness in sl6.1
 
 What happens when it's extremely slow? Like you hit a key and it's not echoed 
 on the screen for a long time (seconds)?
 
 Have a nice day
 GV
 
 -Original Message-
 From: owner-scientific-linux-us...@listserv.fnal.gov 
 [mailto:owner-scientific-linux-us...@listserv.fnal.gov] On Behalf Of Turtaut 
 Geoffroy
 Sent: Thursday, November 17, 2011 11:18 PM
 To: SCIENTIFIC-LINUX-USERS@LISTSERV.FNAL.GOV
 Subject: Re: console slowness in sl6.1
 
 Hi,
 
 We have the same issue since we use RHEL 6.X/SL 6.X
 
 Topic : HP Z400 system very slow at ramdom times for 5 minutes (october 2011)
 
 We use HP workstations (XW4600, Z200, Z400, ...), VM, DELL laptops and the 
 problem is only present on Z400 (xeon) et an HP pavilion (corei7).
 
 The problem is not present if we use 2.6.32 kernel from kernel.org.
 
 We have a case on RHN ...
 
 Last RH comment was :
 We have analysed the results and see that there is nothing waiting on I/O but 
 there is a high CPU usage and large run queue. 
 
 Geoffroy Turtaut

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: console slowness in sl6.1

[Stephan Wiesand]

On Nov 17, 2011, at 20:07 , Christopher Tooley wrote:

 I can see from your original email that I missed the
 (and/or any other terminals I'm using, for example SSH from another machine)
 part. :\  Derp on my part!
 
 Have you modified your /etc/bash_profile at all?  I would assume that if this 
 is a clean install the software should be fine...

More precisely: what are the values of $PATH, $LD_LIBRARY_PATH and $LD_PRELOAD 
after a fresh login?

- Stephan

Re: Memory leak in Emacs 23.1 - SL.6.1

[Stephan Wiesand]

Hello Felip,

On Oct 24, 2011, at 21:48, Felip Moll wrote:

 Recently I installed a SL6.1 Cluster with 16 nodes, slurm resource manager, 
 etc.
 
 I use Emacs to edit my files as do some of the researchers of my 
 investigation center.
 
 One day I detected that some daemons hunged. I discovered that the kernel was 
 killing some processes because the system went out of memory. I couldn't 
 reproduce the error anymore, and the thing seemed to occurr on very random 
 times.
 
 Since this day, I limited with limits.conf the user stack to 15Gb. (my server 
 has 16gb, and consumes normally no more than 1gb).
 
 Today, when I was doing some tasks I could see what is causing the problem. 
 It's Emacs!. There seems to be a user that uses emacs and that does logout 
 from his session without exiting it. The access to the server is done by ssh.
 
 I will try to install the latest 23.3 version with .tar.gz package but I like 
 to use the Yum package if it's possible in order to keep the installation as 
 clean as possible.
 
 Should I report it to some place?

it seems someone else already has: 
https://bugzilla.redhat.com/show_bug.cgi?id=732157

There's a proposed patch attached to that BZ.

Regards,
Stephan


-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: Memory leak in Emacs 23.1 - SL.6.1

[Stephan Wiesand]

On Oct 25, 2011, at 15:36, Felip Moll wrote:

 Thanks to all!
 
 I didn't search very well into the bugs database. Sorry.
 
 Jean-Paul, I took a look to your repos but I didn't find any emacs package. I 
 can only see:
   • emacs-doxymacs - Doxygen add-on for Emacs/XEmacs
   • emacs-w3 - W3 package for Emacs
   • emacs-sdcc - Emacs extensions for SDCC
 
 But no emacs and emacs-common package.

I built packages with the proposed fix from BZ. It seems to work, and fix the 
issue. If you're interested even though they had no real testing, find them in 
http://www-zeuthen.desy.de/~wiesand/SL6/emacs/

Regards,
Stephan

 Best regards.
 Felip
 
 2011/10/25 Stephan Wiesand stephan.wies...@desy.de
 Hello Felip,
 
 On Oct 24, 2011, at 21:48, Felip Moll wrote:
 
  Recently I installed a SL6.1 Cluster with 16 nodes, slurm resource manager, 
  etc.
 
  I use Emacs to edit my files as do some of the researchers of my 
  investigation center.
 
  One day I detected that some daemons hunged. I discovered that the kernel 
  was killing some processes because the system went out of memory. I 
  couldn't reproduce the error anymore, and the thing seemed to occurr on 
  very random times.
 
  Since this day, I limited with limits.conf the user stack to 15Gb. (my 
  server has 16gb, and consumes normally no more than 1gb).
 
  Today, when I was doing some tasks I could see what is causing the problem. 
  It's Emacs!. There seems to be a user that uses emacs and that does logout 
  from his session without exiting it. The access to the server is done by 
  ssh.
 
  I will try to install the latest 23.3 version with .tar.gz package but I 
  like to use the Yum package if it's possible in order to keep the 
  installation as clean as possible.
 
  Should I report it to some place?
 
 it seems someone else already has: 
 https://bugzilla.redhat.com/show_bug.cgi?id=732157
 
 There's a proposed patch attached to that BZ.

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: hardware compatibility list

[Stephan Wiesand]

Hi,

On Oct 25, 2011, at 17:18 , Scott Williams wrote:

 I have been searching for a hardware compatibility list for scientific
 linux, specifically for motherboards (which I have had a problem with
 in the past). I have a hardware supplier that has limited stock, so I
 am not free to choose from an extensive list and would like to be able
 to check particular motherboards for compatibility. Initial searches
 have not proved successful. Does such a list exist?

have a look at https://hardware.redhat.com/ . Hardware certified for this 
distribution will very likely work with the corresponding SL release...

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany



smime.p7s
Description: S/MIME cryptographic signature

Re: UEFI

[Stephan Wiesand]

On Oct 20, 2011, at 17:47 , Yasha Karant wrote:

[more stuff with no bearing on SL whatsoever]

Could this person please be banned from the list.

Thanks,
Stephan

Re: SL 5.7 Intel Integrated HD Graphics 3000 SandyBridge

[Stephan Wiesand]

On Oct 17, 2011, at 18:20 , Yasha Karant wrote:
[...]
 Now I have a decision to make:  IA-32 SL 6.1 or X86-64 SL 6.1 .  The 
 processor will support X86-64, but the machine only has 4 Mbyte of RAM as 
 delivered (upgradeable to 8 Mbyte -- but this is a cost) -- which is only 0.5 
 Mword in X86-64 64 bit mode.  The hard drive is 500 Gbyte SATA at 5400 RPM -- 
 not a high performance unit.

I guess we're talking Gigabytes of RAM here, not Megabytes?

 There is no special need for 64 bit work on the machine that primarily is an 
 end-user linux workstation: web browser (including use of university services 
 only available through such an interface), IMAP email client, OpenOffice, 
 various LaTeX interfaces, some display of video, use of Linux VirtualBox to 
 use MS Win (for which the unit is licensed) to use a MS Win only application, 
 but no development or programming -- and the ability for a skilled end user 
 under Network Manager to connect to whatever 802.11 WAP that is available 
 without my intervention.
 
 Thus, my feeling is to stay with the IA-32 environment.  Any thoughts to the 
 contrary?

Yes: The 32-bit kernel will leave 25% of your 4 GB RAM unused, I believe. And 
all processes will be confined to 3 GB of address space (even if purely 
virtual). Increasingly, new features are only made available by TUV for the 
64-bit flavour (KVM, xfs, samba3x on SL5, pNFS). Since the Java and Flash 
plugins are now available as 64-bit builds, much of the hassle with running 
64-bit SL is now history. x86-64 has a future, ia32 IMHO hasn't (x32 seems 
interesting but will take a while to arrive and will use a 64-bit kernel). The 
extended register set and faster PC-relative addressing are not available to 
ia32 applications. A 500GB disk is plenty for installing the .i686 packages 
alongside the 64-bit ones.

That being said, staying with ia32 may still be slightly more convenient, and 
part of the 1 GB of real memory you gain with x86-64 will be consumed by 64-bit 
pointers/longs and alignment.

Choose your poison ;-)

HTH,
Stephan


-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: HP Z400 system very slow at ramdom times for 5 minutes

[Stephan Wiesand]

Hi,

On Oct 14, 2011, at 16:19, Turtaut Geoffroy wrote:

 We have a problem with RHEL 6.x and SL 6.X.
 The system is  very slow at ramdom times for 5 minutes and became normal.
 
 The problem arrives on HP Z400 and HP pavilion elite hpe i7.
 The problem does not arrives on HP XW4600 et HP Z200 (i5).
 
 We have a case opened with Redhat and HP.
 
 Does someone have this problem ?

probably. Does https://bugzilla.redhat.com/show_bug.cgi?id=710265 ring a bell?

Cheers,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: Does KVM have a Guest Additions?

[Stephan Wiesand]

On Sep 26, 2011, at 00:47 , William Scott wrote:

 On 26 September 2011 04:21, Stephan Wiesand stephan.wies...@desy.de wrote:
 
 
 NB for optimum performance (and possibly: reliability) of your XP VM, you'll 
 need the drivers from virtio-win - which is available from TUV, but closed 
 source, with a proprietary license, and can only be used with an active 
 subscription. If I were making a living from running Windows VMs on EL 
 hosts, I'd probably bite the bullet.
 
 What about http://alt.fedoraproject.org/pub/alt/virtio-win/latest/images/bin/ 
 ?
 
 Will that iso do?

I don't know, but one would hope that the interfaces for guest drivers are 
being kept stable...

Anyway, thanks for the pointers, I wasn't aware of those sources for the 
software at all.

TUV may still have some additional IP in what they provide to subscription 
customers. Otherwise I wouldn't understand the restrictive license and no 
source being provided at all.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Does KVM have a Guest Additions?

[Stephan Wiesand]

On Sep 26, 2011, at 05:41 , Todd And Margo Chester wrote:

 On 09/25/2011 11:20 AM, Stephan Wiesand wrote:
 while trying it (securely) with an SL6
 Thank you!
 
 You have mentioned Securely.  By this are you trying to
 operate Spice across two physical  machines?

I'm currently not using it at all. But back then, yes, I was running the spice 
client on a different system than the one hosting the VM, tunneling the spice 
connection through an ssh one. IIRC, the simple setup in my notes made the 
connection unencrypted and unauthenticated... and it took me while to even find 
the information, plus some trial and error, to get it up and running like that.

Maybe someone will now send a pointer to a Desktop Virtualization with KVM  
SPICE HOWTO I simply missed ?

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Does KVM have a Guest Additions?

[Stephan Wiesand]

Hi T,

On Sep 24, 2011, at 03:46 , Todd And Margo Chester wrote:

 On 09/23/2011 01:41 AM, Stephan Wiesand wrote:
 Hi,
 
 On Sep 23, 2011, at 00:30 , Todd And Margo Chester wrote:
 
 I am migrating from Virtual Box on my old server (CentOS 5.6 x32) to KVM on
 my new server (SL6.1 x64).
 
 Virtual box has a collection of drivers to assist the guest called
 Guest Additions.  These drives give you shared clipboard
 and shared mouse, among other things.
 
 Question: does KVM have a similar package of guest drivers?
 I am specifically looking for mouse sharing and shared clipboard.
 I am not sure if I need a special video driver as well.
 eventually, SPICE will be what you're looking for. See 
 http://www.spice-space.org/features.html .
 
 Some of the components come with SL, some can be downloaded elsewhere. But 
 right now, it's probably not quite finished and hard to get going.
 
 
 Hi Stephan,
 
 Is this the spice you are referring to?
 
 http://rpm.pbone.net/index.php3/stat/4/idpl/16005959/dir/scientific_linux_6/com/spice-xpi-2.4-4.el6.x86_64.rpm.html

I'm referring to the packages coming with SL6.1:

SL % ls 6.1/x86_64/Packages/spice* 
6.1/x86_64/Packages/spice-client-0.8.0-2.el6.x86_64.rpm
6.1/x86_64/Packages/spice-common-devel-0.4.2-7.el6.i686.rpm
6.1/x86_64/Packages/spice-common-devel-0.4.2-7.el6.x86_64.rpm
6.1/x86_64/Packages/spice-protocol-0.8.0-1.el6.noarch.rpm
6.1/x86_64/Packages/spice-server-0.8.0-1.el6.x86_64.rpm
6.1/x86_64/Packages/spice-server-devel-0.8.0-1.el6.x86_64.rpm
6.1/x86_64/Packages/spice-vdagent-0.6.3-8.el6.x86_64.rpm
6.1/x86_64/Packages/spice-xpi-2.4-4.el6.x86_64.rpm

The one you found on pbone is one of them, but not the most important one.

Here are the notes I jotted down while trying it (securely) with an SL6 VM on 
an SL6 host (probably using 6.0 beta):

--8--
o make sure xorg-x11-drv-qxl and spice-server are installed in the vm
o virsh edit vm
   graphics type='spice' port='5903' autoport='no' keymap='en-us'/
   video
 model type='qxl' heads='1'/
 alias name='video0'/
 address type='pci' domain='0x' bus='0x00' slot='0x02' function='0x0'/
   /video
o on the host, install spice-client
o ssh -L5903:localhost:5903 host
o spicec -h localhost -p 5903
TODO: SSL connection, USB, Audio, Windows Client, Windows VM, How to use 
spice-xpi?
--8--

I haven't gotten a round tuit for anything on the TODO list. For your Windows 
VM, you'll probably need the driver/agent/whatever from spice-space.org. I have 
no clue which version would work with what comes with 6.1.

NB for optimum performance (and possibly: reliability) of your XP VM, you'll 
need the drivers from virtio-win - which is available from TUV, but closed 
source, with a proprietary license, and can only be used with an active 
subscription. If I were making a living from running Windows VMs on EL hosts, 
I'd probably bite the bullet.

Cheers,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Does KVM have a Guest Additions?

[Stephan Wiesand]

Hi,

On Sep 23, 2011, at 00:30 , Todd And Margo Chester wrote:

 I am migrating from Virtual Box on my old server (CentOS 5.6 x32) to KVM on
 my new server (SL6.1 x64).
 
 Virtual box has a collection of drivers to assist the guest called
 Guest Additions.  These drives give you shared clipboard
 and shared mouse, among other things.
 
 Question: does KVM have a similar package of guest drivers?
 I am specifically looking for mouse sharing and shared clipboard.
 I am not sure if I need a special video driver as well.

eventually, SPICE will be what you're looking for. See 
http://www.spice-space.org/features.html .

Some of the components come with SL, some can be downloaded elsewhere. But 
right now, it's probably not quite finished and hard to get going.

 Many thanks,
 -T
 
 p.s. be really cool if all that stuff were built in

6.2 may have some improvements in that area.

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Sharing iscsi luns between sl5.x and sl6.x

[Stephan Wiesand]

On Sep 23, 2011, at 11:17, carlopmart wrote:

 Due to a certain app requeriments, I need to share an iscsi lun between two 
 sl5.x and five sl6.x servers. This lun is formatted with ext4 fileysytem (I 
 have used one of the sl5.x servers to do this).

You mean you want to mount the filesystem on both servers at the same time?

 Can I expect some type of problems ???

Oh yes... :-)


-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: Sharing iscsi luns between sl5.x and sl6.x

[Stephan Wiesand]

On Sep 23, 2011, at 13:10 , carlopmart wrote:

 On 09/23/2011 12:54 PM, Stephan Wiesand wrote:
 On Sep 23, 2011, at 11:17, carlopmart wrote:
 
 Due to a certain app requeriments, I need to share an iscsi lun between two 
 sl5.x and five sl6.x servers. This lun is formatted with ext4 fileysytem (I 
 have used one of the sl5.x servers to do this).
 
 You mean you want to mount the filesystem on both servers at the same time?
 
 No. Only one of these servers mounts this filesystem.

So by sharing you mean mounting it on any of the servers, but never on more 
than one at a time? I'd expect this to work (but I admittedly don't know for 
sure) as long as it's really never ever mounted more than once. I think ext4 
has no multiple mount protection. Any mistake may cause irreparable damage to 
the filesystem.

Sharing it using NFS is certainly the safer option.


-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Need KVM HD settings advice

[Stephan Wiesand]

On Sep 20, 2011, at 14:48 , Nico Kadel-Garcia wrote:

 On Tue, Sep 20, 2011 at 3:05 AM, Vladimir Mosgalin
 mosga...@vm10124.spb.edu wrote:
 Hi Todd And Margo Chester!
 
  On 2011.09.19 at 18:03:08 -0700, Todd And Margo Chester wrote next:
 
 What I need help with is getting the optimum performance
 settings while converting over my old hard (virtual) drive.
 
 This is what I have gathered from these parts as to the best
 settings:
 
 - controller: virtio
 - kvm option: cache=none
 - qcow2 disk format with metadata preallocation
 - create your disk image with:
  qemu-img create -f qcow2 -o \
  size=400,preallocation=metadata vdisk.img
 
 If you don't need snapshots and such, you might get better performance
 with LVM volumes for storing images, connected as raw images. At least
 that's only thing that I'm using in production and it works well.
 
 LVM has its uses. But the ability to re-allocate space without having
 to manipulate your partition tables is *vital* in a dynamic
 environmemnt, and it's a lot easier to do with image files.

I disagree. There's no need to manipulate partition tables in order to provide 
LVs as backing store to VMs. Extending a LV is as easy as enlarging a backing 
file. LV data alignment can be specified, at least with SL6.

I see no disadvantage in using LVs for VM backing store, other than having to 
learn about LVM. And it does avoid the significant overhead of going through a 
second filesystem layer. At least potentially, it also avoids double caching.

Since we introduced Linux-(Xen-)hosted VMs with SL5, we too have backed all our 
VMs - hundreds - with logical volumes. And we haven't changed that with SL6 and 
KVM.

 Tuning nthe partition the images reside on, now *that* is invaluable.
 Turn off noatime, use a fast and simple file system.

Well, just don't use a filesystem... NB noatime tunes the filesystem, not the 
block device (partition or LV) it resides on.

[...]


-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: SL6.1 memory usage

[Stephan Wiesand]

Hi Stijn,

On Sep 5, 2011, at 16:24, Stijn De Weirdt wrote:

 hi all,
 
 we are having an issue with some SL61 nodes. after a reboot, free reports 
 1.4GB of memory in use, of which 24+163=187MB buffers+cache.
 
 i'm unable to identify what is holding the memory, and i'd like to know if 
 others see this too and how i could proceed to find the culprit.

yes, we see this as well. On a 48 GB system without users or special processes:

# free -m
 total   used   free sharedbuffers cached
Mem: 48388   1374  47013  0 30186
-/+ buffers/cache:   1157  47231

In /proc/meminfo, I find that the difference to what I'd consider reasonable 
(and see on a 48GB SL5 system) is due to slabs.

A slabtop -s c reveals that it's a size-2097152 pool accounting for this. 
Do you see this as well?

Cheers,
Stephan

 
 (it is a 32core/64GB machine; kernel commandline has crashkernel=128M@16M 
 (but no difference then eg crashkernel=auto and kdump is off))
 
 many thanks,
 
 
 stijn
 
 free
 # free -m
 total   used   free sharedbuffers cached
 Mem: 64554   1604  62949  0 24166
 -/+ buffers/cache:   1413  63140
 Swap:16394  0  16394
 
 
 mem sorted top
 
 top - 16:13:52 up 13 min,  1 user,  load average: 0.00, 0.01, 0.01
 Tasks: 694 total,   1 running, 693 sleeping,   0 stopped,   0 zombie
 Cpu(s):  0.0%us,  0.0%sy,  0.0%ni,100.0%id,  0.0%wa,  0.0%hi,  0.0%si, 0.0%st
 Mem:  66103768k total,  1643336k used, 64460432k free,25164k buffers
 Swap: 16787916k total,0k used, 16787916k free,   170552k cached
 
  PID USER  PR  NI  VIRT  RES  SHR S %CPU %MEMTIME+  COMMAND 
 2788 root  20   0 37988  25m 2876 S  0.0  0.0   0:00.06 pbs_mom 
 2653 root  20   0  159m  12m 1472 S  0.0  0.0   0:00.19 ncm-cdispd 
 2643 root  20   0  138m 5604  840 S  0.0  0.0   0:00.00 cdp-listend 
 3276 root  20   0  120m 4156 3232 S  0.0  0.0   0:00.07 sshd 
 2620 root  20   0  745m 3788 1764 S  0.0  0.0   0:00.12 automount 
 3102 nslcd 20   0  427m 2936  488 S  0.0  0.0   0:00.00 nslcd 
 3301 root  20   0  103m 1688 1336 S  0.0  0.0   0:00.05 bash 
 3623 root  20   0 13528 1604  844 R  0.3  0.0   0:00.14 top 
1 root  20   0 21416 1544 1240 S  0.0  0.0   0:06.23 init 
 2482 root  20   0  194m 1484 1108 S  0.0  0.0   0:00.14 qlgc_dsc 
 2325 root  20   0  242m 1412  928 S  0.0  0.0   0:00.04 rsyslogd 
 2459 rpcuser   20   0 23112 1168  884 S  0.0  0.0   0:00.00 rpc.statd 
 2606 root  18  -2 10956 1144  412 S  0.0  0.0   0:00.03 udevd 
 3164 nscd  20   0  583m 1132  788 S  0.0  0.0   0:00.02 nscd 
 2697 root  20   0 62040 1064  464 S  0.0  0.0   0:00.00 sshd 
  943 root  16  -4 10960 1052  316 S  0.0  0.0   0:00.12 udevd 
 2607 root  18  -2 10956 1052  320 S  0.0  0.0   0:00.00 udevd 
 2723 root  20   0  112m 1012  380 S  0.0  0.0   0:00.00 crond 
 2707 root  20   0 22488  992  752 S  0.0  0.0   0:00.03 xinetd 
 2439 rpc   20   0 18940  908  672 S  0.0  0.0   0:00.04 rpcbind 
 2568 dbus  20   0 23448  876  604 S  0.0  0.0   0:00.01 dbus-daemon 
 2972 nagios20   0 37096  796  452 S  0.0  0.0   0:00.00 nrpe 

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: SL6.1 memory usage

[Stephan Wiesand]

Hi Stijn,

On Sep 5, 2011, at 17:23, Stijn De Weirdt wrote:

 hi stephan
 
 the first lines of slabtop show
 
 Active / Total Objects (% used): 238382 / 243482 (97.9%)
 Active / Total Slabs (% used)  : 4987 / 4987 (100.0%)
 Active / Total Caches (% used) : 64 / 80 (80.0%)
 Active / Total Size (% used)   : 68080.12K / 69386.02K (98.1%)
 Minimum / Average / Maximum Object : 0.01K / 0.28K / 8.00K
 
 is this similar?

not really.

% slabtop -s c --once |head -10
 Active / Total Objects (% used): 185625 / 205302 (90.4%)
 Active / Total Slabs (% used)  : 16756 / 16757 (100.0%)
 Active / Total Caches (% used) : 101 / 182 (55.5%)
 Active / Total Size (% used)   : 856865.31K / 859451.65K (99.7%)
 Minimum / Average / Maximum Object : 0.02K / 4.19K / 4096.00K

  OBJS ACTIVE  USE OBJ SIZE  SLABS OBJ/SLAB CACHE SIZE NAME   
   384384 100% 2048.00K3841786432K size-2097152   
 26276  26250  99%1.00K   65694 26276K ext4_inode_cache   
   181181 100%   32.12K1811 11584K kmem_cache 

 (btw size-2097152 sounds like one of the default name used by hugectl (or 
 hugeadm) from the hugetlbfs tools). is that mounted in your case? and are 
 there any hugepages reserved? )

Not that I'd know of. But wasn't there a new feature called transparent 
hugepage support in 6.1?

Cheers,
Stephan
 
 
 stijn
 
 On 09/05/2011 05:10 PM, Stephan Wiesand wrote:
 Hi Stijn,
 
 On Sep 5, 2011, at 16:24, Stijn De Weirdt wrote:
 
 hi all,
 
 we are having an issue with some SL61 nodes. after a reboot, free reports 
 1.4GB of memory in use, of which 24+163=187MB buffers+cache.
 
 i'm unable to identify what is holding the memory, and i'd like to know if 
 others see this too and how i could proceed to find the culprit.
 
 yes, we see this as well. On a 48 GB system without users or special 
 processes:
 
 # free -m
  total   used   free sharedbuffers cached
 Mem: 48388   1374  47013  0 30186
 -/+ buffers/cache:   1157  47231
 
 In /proc/meminfo, I find that the difference to what I'd consider reasonable 
 (and see on a 48GB SL5 system) is due to slabs.
 
 A slabtop -s c reveals that it's a size-2097152 pool accounting for 
 this. Do you see this as well?
 
 Cheers,
  Stephan
 
 
 (it is a 32core/64GB machine; kernel commandline has crashkernel=128M@16M 
 (but no difference then eg crashkernel=auto and kdump is off))
 
 many thanks,
 
 
 stijn
 
 free
 # free -m
 total   used   free sharedbuffers cached
 Mem: 64554   1604  62949  0 24166
 -/+ buffers/cache:   1413  63140
 Swap:16394  0  16394
 
 
 mem sorted top
 
 top - 16:13:52 up 13 min,  1 user,  load average: 0.00, 0.01, 0.01
 Tasks: 694 total,   1 running, 693 sleeping,   0 stopped,   0 zombie
 Cpu(s):  0.0%us,  0.0%sy,  0.0%ni,100.0%id,  0.0%wa,  0.0%hi,  0.0%si, 
 0.0%st
 Mem:  66103768k total,  1643336k used, 64460432k free,25164k buffers
 Swap: 16787916k total,0k used, 16787916k free,   170552k cached
 
  PID USER  PR  NI  VIRT  RES  SHR S %CPU %MEMTIME+  COMMAND
 2788 root  20   0 37988  25m 2876 S  0.0  0.0   0:00.06 pbs_mom
 2653 root  20   0  159m  12m 1472 S  0.0  0.0   0:00.19 ncm-cdispd
 2643 root  20   0  138m 5604  840 S  0.0  0.0   0:00.00 cdp-listend
 3276 root  20   0  120m 4156 3232 S  0.0  0.0   0:00.07 sshd
 2620 root  20   0  745m 3788 1764 S  0.0  0.0   0:00.12 automount
 3102 nslcd 20   0  427m 2936  488 S  0.0  0.0   0:00.00 nslcd
 3301 root  20   0  103m 1688 1336 S  0.0  0.0   0:00.05 bash
 3623 root  20   0 13528 1604  844 R  0.3  0.0   0:00.14 top
1 root  20   0 21416 1544 1240 S  0.0  0.0   0:06.23 init
 2482 root  20   0  194m 1484 1108 S  0.0  0.0   0:00.14 qlgc_dsc
 2325 root  20   0  242m 1412  928 S  0.0  0.0   0:00.04 rsyslogd
 2459 rpcuser   20   0 23112 1168  884 S  0.0  0.0   0:00.00 rpc.statd
 2606 root  18  -2 10956 1144  412 S  0.0  0.0   0:00.03 udevd
 3164 nscd  20   0  583m 1132  788 S  0.0  0.0   0:00.02 nscd
 2697 root  20   0 62040 1064  464 S  0.0  0.0   0:00.00 sshd
  943 root  16  -4 10960 1052  316 S  0.0  0.0   0:00.12 udevd
 2607 root  18  -2 10956 1052  320 S  0.0  0.0   0:00.00 udevd
 2723 root  20   0  112m 1012  380 S  0.0  0.0   0:00.00 crond
 2707 root  20   0 22488  992  752 S  0.0  0.0   0:00.03 xinetd
 2439 rpc   20   0 18940  908  672 S  0.0  0.0   0:00.04 rpcbind
 2568 dbus  20   0 23448  876  604 S  0.0  0.0   0:00.01 dbus-daemon
 2972 nagios20   0 37096  796  452 S  0.0  0.0   0:00.00 nrpe
 
 
 

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: hwmonitor or equivalent for SL 6 x86-64

[Stephan Wiesand]

On Jun 19, 2011, at 04:59 , Yasha Karant wrote:

 I have installed lshw.  lshw does seem to give an extensive listing, but 
 lshw-gui does not seem to give much.  As with lshw, does lshw-gui need to be 
 run by root?
 
 Also, I have a real 1.44 Mbyte floppy drive installed that goes to the floppy 
 drive controller on the mother board (this particular MSI motherboard has 
 SATA, EIDE, and floppy controllers and connectors on the motherboard).  It 
 worked fine under RHEL 5 (CentOS 5.6) on this motherboard.  Under RHEL 6 (SL 
 6), I find:
 
 ls -la /dev/fd/*
 ls: cannot access /dev/fd/255: No such file or directory
 ls: cannot access /dev/fd/3: No such file or directory
 lrwx--. 1 ykarant ykarant 64 Jun 18 19:49 /dev/fd/0 - /dev/pts/0
 lrwx--. 1 ykarant ykarant 64 Jun 18 19:49 /dev/fd/1 - /dev/pts/0
 lrwx--. 1 ykarant ykarant 64 Jun 18 19:49 /dev/fd/2 - /dev/pts/0s
 
 but I cannot seem to access these via a mount, even as root, to access a 
 MS-DOS floppy.  Obviously, I am doing something wrong, but what? Moreover, 
 the mtools (that provides MS-DOS compatibility) used to access the floppy 
 drive as A: but now does nothing.  Presumably, once I understand how to 
 access the floppy drive, things will work.  Would a ln -s /dev/floppy to 
 /dev/fd/0 as well as a ln -s /dev/fd0 to /dev/fd/0 work?

Probably not ;-) Try  echo 'I am not a floppy drive'  /dev/fd/1 for a hint 
what these actually are.

 Note that the output of lshw does not show the floppy drive, although the 
 hardware listing utility of RHEL 5 did show this. This is the same hardware 
 with no change to the motherboard BIOS -- the motherboard BIOS utility does 
 show the floppy.

Do floppy devices appear after modprobe floppy?

 I realize that this might have to be re-done upon the next upgrade (to SL 
 6.1), but otherwise should work until /dev is overwritten.


It's created at boot time.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: hwmonitor or equivalent for SL 6 x86-64

[Stephan Wiesand]

On Jun 18, 2011, at 09:36 , Phil Perry wrote:

 On 18/06/11 02:10, Yasha Karant wrote:
 
 
 2. The grub or whatever switch / configuration file so that the actual
 boot process and starting processes list (including any failures) is
 displayed to the console rather than simply some icon (spinning under
 noveau, progress bar under regular xorg including the Nvidia proprietary
 driver).
 
 
 Pressing F6 during boot shows the info for me. I've not found a way to get it 
 with a grub config yet.

Remove rhgb quiet from the kernel command line?

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: scilinux 6 install fail with log

[Stephan Wiesand]

At least this one is not from an SL6 install:

On Jun 15, 2011, at 22:25, Yasha Karant wrote:

 /mnt/sysimage/root/install.log:
 Installing setup-2.5.58-7.el5.noarch
 warning: setup-2.5.58-7.el5: Header V3 DSA signature: NOKEY, key ID e8562897
 Installing filesystem-2.4.0-3.el5.i386
 Installing desktop-backgrounds-basic-2.0-41.el5.centos.noarch
 Installing kernel-headers-2.6.18-194.el5.i386
 Installing centos-release-notes-5.5-0.i386

Is ist possible that you try to install without reformatting /, /var/ and /usr?

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: scilinux 6 install fail with log

[Stephan Wiesand]

On Jun 16, 2011, at 18:58 , Yasha Karant wrote:

 This one is from a SL6 install.

This particular file isn't. Look closer :-)
[snip]
 I did not reformat / , /var, /usr .  Must these be reformatted?  Will X86-64 
 SL6 allow me to keep these as ext2 (no journal)?

I think it's still possible to use ext2 on your partitions when installing SL6. 
I don't think it's reasonable to keep the old content of /, /var, and /usr, and 
I guess it's what makes your attempt to install SL6 fail.

Regards,
Stephan
 
 Thanks,
 
 Yasha Karant
 
 On 06/16/2011 01:43 AM, Stephan Wiesand wrote:
 At least this one is not from an SL6 install:
 
 On Jun 15, 2011, at 22:25, Yasha Karant wrote:
 
 /mnt/sysimage/root/install.log:
 Installing setup-2.5.58-7.el5.noarch
 warning: setup-2.5.58-7.el5: Header V3 DSA signature: NOKEY, key ID e8562897
 Installing filesystem-2.4.0-3.el5.i386
 Installing desktop-backgrounds-basic-2.0-41.el5.centos.noarch
 Installing kernel-headers-2.6.18-194.el5.i386
 Installing centos-release-notes-5.5-0.i386
 
 Is ist possible that you try to install without reformatting /, /var/ and 
 /usr?
 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: scilinux 6 install fail with log

[Stephan Wiesand]

On Jun 16, 2011, at 20:39 , Connie Sieh wrote:

 On Thu, 16 Jun 2011, Urs Beyerle wrote:
 
 On 06/16/2011 06:58 PM, Yasha Karant wrote:
 This one is from a SL6 install. SL and CentOS are both RHEL.  I keep the 
 system utilities stock (the same as TUV, RHEL in this case), except for the 
 use of
 the graphics card driver from the graphics card vendor, not generic X 
 (e.g., on this machine, the Nvidia driver for linux X).
 I am switching to SL over CentOS because (1) we do not have funding luxury 
 to license the binaries from RH and (2) CentOS 6 is not yet available 
 despite RHEL
 6.1 already having been released.  Other than re-branding, SL and CentOS 
 both claim to be RHEL clones -- I know that the RPMs that work on RHEL 
 release X work
 just as well on both CentOS and SL of the same release.
 
 (Why not SL over CentOS?  A matter of history, not a specific choice. With 
 the upcoming demise of Fermilab as a direct experimental facility, 
 hopefully the EU
 will continue to fund CERN and not be shortsighted as USA neoliberal 
 Republican Tea Partists force upon the USA, and thus maintain support for 
 SL.)
 
 I did not reformat / , /var, /usr .  Must these be reformatted?
 
 Yes, otherwise you will have a mixture of the old system (CentOS5?) and the 
 new SL6 system on /, /var, /usr. This will definitely not work.
 
 If you want to keep old data you have to do an update instead of an install. 
 But I don't know if CentOS5 can be update to SL60 with SL60 install DVD. My 
 guess
 is that this will not work.
 
 An update from 5 to 6 is NOT supported by either RedHat or SL.  RedHat 
 has code to specifically not allow it.
 Others have tried to force an upgrade and were not successful.

I figure it's possible to choose advanced partitioning and simply uncheck 
format for all existing partitions while choosing the old mountpoints. 
Technically, that's an unsupported upgrade, while you don't have to boot with 
an upgradeany kernel parameter or choose update existing linux installation 
in the anaconda GUI. And I figure it's what happened here.

- Stephan

 
 
 To be save make a backup before you format the partitions.
 
 
 Will X86-64 SL6 allow me to keep these as ext2 (no journal)?
 
 I think if you choose custom partitioning you can format your partitions 
 with ext2. Just curious, what's the reason to use old ext2?
 
 Cheers,
 
Urs
 
 
 
 
 -Connie Sieh

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: kernel-2.6.32-131.2.1.el6.x86_64 Xen DomUs.

[Stephan Wiesand]

Hi Steven,

On Jun 14, 2011, at 0:04, Steven Haigh wrote:

 In the latest batch of updates, I installed kernel-2.6.32-131.2.1.el6.x86_64 
 via a 'yum -y update'.
 
 It seemed all of my 5 Xen DomUs running SL6 failed to boot upon rebooting. It 
 looks like all of these systems failed at roughly where switchroot is called.
 
 Interestingly, I could recover from this by shutting down the DomU, mounting 
 then chrooting the DomUs filesystem and issuing:
   # rpm -ivh --force kernel-2.6.32-131.2.1.el6.x86_64.rpm
 
 Not sure how widespread this issue is, but I had a 100% failure rate on Xen 
 DomUs. The one bare metal system I updated did so with no errors.

no problems here on the two Xen domUs we're currently running. But then, we're 
still updating kernels with rpm -i rather than yum.

Just a data point.

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: Installing SL6 on a Dell PE1950

[Stephan Wiesand]

On May 27, 2011, at 15:48, Randall Martin wrote:

 I’m having trouble installing SL6 on a Dell PE1950 via PXE kickstart.

Works for me:

# dmidecode |grep Product
Product Name: PowerEdge 1950
# lspci|grep Ethernet
05:00.0 Ethernet controller: Broadcom Corporation NetXtreme II BCM5708 Gigabit 
Ethernet (rev 11)


  The node boots and loads the installer, but I get a message “Waiting for 
 NetworkManager to configure eth0” followed by the error “Network Error: There 
 was an error configuring your network interface”.  It looks like the bnx2 
 ethernet driver isn’t loaded.  What is the procedure to add this driver to 
 initrd in the install media?  Would creating a driver update image that is 
 referenced in the kickstart file be a more elegant solution? 

When we have trouble with PXE installs, it usually turns out that the switch 
port was forgotten too be set to portfast.

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: SL6, SELinux, openafs

[Stephan Wiesand]

Hi Orion,

On May 12, 2011, at 23:31 , Orion Poplawski wrote:

 On 05/12/2011 03:17 PM, Orion Poplawski wrote:
 On 05/12/2011 03:04 PM, Orion Poplawski wrote:
 I'm just trying out openafs on SL6 and ran into the following starting the 
 afs
 daemon:
 
 Starting AFS client.
 afsd: some file missing or bad in /usr/vice/etc
 
 Turns out this was caused by following the quick start guide and linking
 /usr/afs/etc/{CellServDB,ThisCell} to /usr/vice/etc. Just copying the files
 allows everything to work with SELinux enforcing.
 
 
 Spoke too soon.  Lots of denials on the server side of things.  Is there any 
 expectation that you can run openafs in enforcing mode?


the client will work if everything is labelled correctly, and we took care that 
this is true after installation of the packages, no matter whether under 
anaconda or later from yum. That's also why we changed the default cache 
location.

On the server side, this work hasn't been done yet. In addition, the policy is 
not quite complete and will prevent the fileserver from running successfully 
even if all labels are what restorecon will make them. There's at least a 
problem with the fssync.sock, but maybe more.

The fileserver *should* work with SELinux in enforcing mode, as long as the 
init script has type unconfined_exec_t.
But of course that's cheeting.

In any case, please keep in mind that the OpenAFS in SL6 is still a prerelease, 
though a relatively advanced and mature one. But if you want to run a 
production server, I recommend SL5 for the time being. More feedback on the SL6 
one is of course very welcome.

Best regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: RHEL/SL and iptables

[Stephan Wiesand]

On Apr 16, 2011, at 21:13 , Nicolas Kovacs wrote:

 Until recently, I've only been using the system-config-securitylevel-tui 
 utility, because it's easy to use while covering all my needs.
 
 Now I have to switch to a manual iptables configuration, because 1) the 
 system-config-securitylevel-tui utility has been dumbed down, and 2) some 
 of the things I want to do need a more fine-grained control.
 
 What's the most orthodox (e. g. clean) solution to configure iptables 
 manually (in a script, somewhere) with SL ?

Probably: creating an initial configuration with 
system-config-securitylevel[-tui] or at install time, then editing 
/etc/sysconfig/iptables. The format is that of iptables-save.


Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: evolution crashing after glibc update

[Stephan Wiesand]

On Apr 7, 2011, at 10:18, Matthias Schroeder wrote:

 On 04/06/2011 07:21 PM, Simon Butcher wrote:
 Hello
 
 After last night's yum security updates on our 5.3 and 5.5 machines,
 evolution is crashing with the dump below when trying to compose/send
 an email
 
 Does a reboot help?

No.

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: /sbin/dracut not found

[Stephan Wiesand]

On Mar 30, 2011, at 18:42 , Orion Poplawski wrote:

 On 03/29/2011 07:47 PM, Tim Potter wrote:
 Hi everyone.  I'm installing Scientific Linux 6.0 from a kickstart and
 see the following error in the /root/install.log when the installation
 completes:
 
 /sbin/new-kernel-pkg: line 296: /sbin/dracut: No such file or directory
 mkinitrd failed
 
 Works fine for me, and the deps appear to be there.  Are you adding any other 
 repos for the install?


We're also kickstarting SL6 all the time (99.9% x86_64), and haven't 
encountered this problem.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Virtual floppy as sda on Dell M605 blades

[Stephan Wiesand]

On Mar 17, 2011, at 19:20 , Joel Maslak wrote:

 I'm seeing some differences in SCSI drive numbering in SL 6 / RHEL 6, 
 compared to CentOS 5 / RHEL 5.  This means that on *some* of my servers, 
 /dev/sdb is actually the first hard disk, which makes kickstarting a bit more 
 annoying.  Does anyone know a way to reverse the order or similar easily?

We're using pci=bfsort for SL5 and pci=bfsort nousb for SL6 when 
kickstarting. The pci=bfsort is kept on the installed system, nousb isn't. 
Eventually, using /dev/disk/by-* will be the right solution, but for the time 
being these kernel parameters give us consistent device enumeration across 
SL5/6.

- Stephan

  A parameter to grub to either remove the virtual floppy or reorder these 
 would be handy, if someone knows of one.  Certainly I can disable the virtual 
 floppy in these machines, but I'd rather not do that (it's needed for BIOS 
 updates).
 
 In CentOS 5 / RHEL 5, I would see disk configurations such as:
 
 scsi 0:0:0:0: CD-ROM Virtual CDROM1.00 PQ: 0 ANSI: 0 
 CCS
 scsi 1:0:0:0: Direct-Access  Virtual Floppy   1.00 PQ: 0 ANSI: 0 
 CCS
 scsi 2:0:0:0: Direct-Access SEAGATE  ST973451SS   SM04 PQ: 0 ANSI: 5
 scsi 2:0:1:0: Direct-Access SEAGATE  ST973451SS   SM04 PQ: 0 ANSI: 5
 scsi 2:1:0:0: Direct-Access Dell VIRTUAL DISK 1028 PQ: 0 ANSI: 5
 s
 sd 0:1:0:0: Attached scsi disk sda
 sd 2:0:0:0: Attached scsi removable disk sdb
 
 Basically, sda was the internal RAID 1 disk array and sdb is the DRAC 
 virtual floppy.  This worked fine, and our scripts knew to expect sda as 
 the first disk.  Note that these are identical systems, purchased at the same 
 time and delivered together.
 
 On SL 6 / RHEL 6, it reverses the order of sda and sdb:
 
 scsi 0:0:0:0: CD-ROM Virtual CDROM1.00 PQ: 0 ANSI: 0 
 CCS
 scsi 1:0:0:0: Direct-Access  Virtual Floppy   1.00 PQ: 0 ANSI: 0 
 CCS
 scsi 2:0:0:0: Direct-Access SEAGATE  ST973451SS   SM04 PQ: 0 ANSI: 5
 scsi 2:0:1:0: Direct-Access SEAGATE  ST973451SS   SM04 PQ: 0 ANSI: 5
 scsi 2:1:0:0: Direct-Access Dell VIRTUAL DISK 1028 PQ: 0 ANSI: 5
 sd 2:1:0:0: [sdb] Attached SCSI disk
 sd 1:0:0:0: [sda] Attached SCSI removable disk
 
 (sdb becomes 1:0:0:0, sdb becomes 2:0:0:0)

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany



smime.p7s
Description: S/MIME cryptographic signature

Re: yum.conf ownership in SL6

[Stephan Wiesand]

On Mar 11, 2011, at 10:15, Tim Edwards wrote:

 On 11/03/11 10:05, Andreas Petzold wrote:
  Hi,
 
 in SL5 I replaced the stock yum repository files and /etc/yum.conf with 
 files 
 customized for our site (local repo mirrors, special excludes etc.) with an 
 RPM with the following properties:
 
 Requires:   redhat-release =  %{version}
 Provides:   yum-conf
 Provides:   epel-release-5
 Obsoletes:  yum-conf
 Obsoletes:  yum-conf-epel
 Obsoletes:  epel-release
 
 With this setup we were able to prevent updates from SL from clobbering our 
 repo setup. This worked fine, since the yum-conf package owns /etc/yum.conf 
 on 
 SL5.
 
 On SL6 however, yum.conf is owned by the yum package. It doesn't make sense 
 for us to rebuild that just to insert our own yum.conf. Of course, we will 
 have to change a few things to accomodate the new sl-release rpm, but that's 
 fine.
 
 Any ideas?

use a trigger:

%triggerin -- yum
install -m 644 /my/yum/conf /etc/yum.conf

 You're not the only one, I found this extremely irritating as we use
 puppet to manage yum config.

yum.conf is marked %config(noreplace). Once modified, updates won't overwrite 
it.

In my opinion, SL does the right thing here by not deviating from RHEL 
unnecessarily.

Regards,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: Missing xenblk.so in the last kernel for SL4/x86_64

[Stephan Wiesand]

Hi,

On Mar 8, 2011, at 19:56, Tomasz Wolak wrote:

 the last version of Xen kernel for SL4/64bit:
 kernel-xenU-2.6.9-89.35.1.EL.x86_64
 
 does not have xenblk.ko module in initrd file:
 /boot/initrd-2.6.9-89.35.1.ELxenU.img
 
 - so paravirtualized xen guests do not boot (they do not find Xen block 
 devices). After adding the module, lines loading the module to init script 
 and rebuilding initrd everything works fine.
 
 I do not know if this was intended, but as it seems weird for a Xen kernel I 
 am reporting it as a potential bug.

we did not encounter this problem when we updated our four remaining SL4 PVMs 
(2x 32-bit, 2x64, all 4.8 fully patched) to this kernel.

Cheers,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: SL6: Wrong baseurl for atrpms repo - renders yum unusable

[Stephan Wiesand]

On Mar 5, 2011, at 07:15 , Garrett Holmstrom wrote:

 On 3/4/2011 17:06, Jon Peatfield wrote:
 
 
 I've never understood why yum doesn't provide a $majorreleasever
 variable as well as $releasever - for those repos where you want to have
 a single tree for each major release (but not for the point releases)
 and don't expect people to edit the .repo files.
 
 I *think* it would be an easy change to yum's config.py (where
 yumvars['releasever'] gets set). Perhaps it would be better to have a
 syntax in the .conf files to allow variables/values to be defined which
 will be expanded in .repo files (you can already use YUM0-YUM9
 environment variables but those are horrid names).
 
 Yum 3.2.28 already has this type of feature.  If you drop a file in 
 /etc/yum/vars/ you can access the first line of its contents like a shell 
 variable.  For example, if you run ``echo 6  /etc/yum/vars/majorreleasever'' 
 then you can do exactly what you described.  So rather than implementing it 
 yourself it might be worth asking upstream to backport the code that already 
 exists.

No need to ask - SL6 already has this feature:

* Fri Apr 16 2010 James Antill james.ant...@redhat.com - 3.2.27-4
- A few minor bugfixes from upstream
- Add dynamic FS based yumvars

Variable names must be lower case.

 Yum has no concept of a major or minor release; it just grabs the version of 
 your sl-release package.  AFAIK, SL is the only distro among its siblings 
 that changes this version with every point release, so it is the only one on 
 which the usual $releasever scheme breaks.

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: TESTING - kernel update for SL5

[Stephan Wiesand]

Hi Jan,

On Feb 23, 2011, at 16:23 , Jan Kundrát wrote:

 On 02/23/11 15:51, Troy Dawson wrote:
 Just because the client kernel is running 2.6.18-238.1.1.el5xen does not
 mean it is trying to be a xen host.  It means that it is running
 paravirtualized.  If your xen machine was setup to be a paravirtualized
 client, then it *has* to continue to run the xen kernel.  You can't just
 switch from the one to the other (as far as I know).
 
 Hi Troy,
 the domU has always been a fully virtualized one, as requested by the
 builder=hvm stanza given in the configuration file.

I guess that explains why noone else is seeing this problem. Why would I want 
to run an SL5 Xen VM under an SL5 Xen  hypervisor as an HVM instead of a 
paravirt VM? I agree it should work, though, and I know that it did in the past.
 
 The disk image
 contains everything, from the bootloader and partitions to the kernel,
 and the in-the-image-installed Grub is invoked and asks me what kernel
 to boot.
 
 When I was speaking about kernel changes, I meant that I have installed
 various versions of the kernel RPM inside the domU, one of them being
 kernel-2.6.18-194.32.1.el5, other kernel-2.6.18-238.1.1.el5 and yet
 another being kernel-xen-2.6.18-238.1.1.el5.
 
 Now, no matter what kernel and Xen versions I choose to run in the dom0,
 physical host, I haven't managed to boot the domU using
 kernel-2.6.18-238.1.1.el5. I'm always using full virtualization, this
 has remained fixed during all tests.
 
 If I pick any of kernel-2.6.18-194.32.1.el5 or
 kernel-xen-2.6.18-238.1.1.el5 at the Grub's prompt displayed inside the
 vncviewer which I use to access the domU's console, it boots fine. Note
 that the kernel-xen package actually boots using the following lines:
 
 title Scientific Linux SL (2.6.18-238.1.1.el5xen)
root (hd0,0)
kernel /boot/xen.gz-2.6.18-238.1.1.el5
module /boot/vmlinuz-2.6.18-238.1.1.el5xen ro root=LABEL=/
module /boot/initrd-2.6.18-238.1.1.el5xen.img
 
 so that kernel is actually running on top of Xen which itself runs in
 the fully virtualized machine, which runs inside Xen on a physical machine.

Interesting, I thought it was impossible to run Xen under Xen and that this 
kind of recursive virtualization is an exclusive feature of KVM.

 When you are running it on Gentoo, you probably set it up to not be
 paravirtualized, so it happily ran the regular kernel.
 
 In fact, the virt-manager run it via kvm, so without any traces of Xen
 at all. That particular physical machine has never had Xen on it.

KVM is readily available on SL5, so this may be a way to solve your actual 
problem.

Best regards,
Stephan

 If you are wondering, I did test the scenario you have.  I currently
 have a xen host running 2.6.18-238.1.1.el5.  Some of it's clients
 are/were running the older kernel, some 2.6.18-238.1.1.el5.  All of them
 are working fine.
 
 I've just updated kernel on another domU instance to the -238, again a
 fully virtualized one, and the symptoms are the same, ie. it won't boot
 and gets stuck on the serial thing. I suspect that both domUs have been
 installed via the same (or at least very similar) kickstart file via PXE.
 
 I guess I can clean up the image and provide it for testing, if you
 think it could help debugging this issue. The same applies for the
 kickstart file.
 
 Thank you for your help so far, I'm really lost at what I'm doing wrong
 here.
 
 With kind regards,
 Jan
 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: TESTING - kernel update for SL5

[Stephan Wiesand]

Hi Troy,

On Feb 4, 2011, at 15:33, Troy Dawson wrote:

 We have had alot of good testing, and thus far there haven't been any show 
 stoppers.
 
 Unless something comes up, we will release this errata to all of SL5 on 
 Wednesday February 8, 2011

we were made aware of this issue: 
http://code.google.com/p/google-perftools/issues/detail?id=305

I wouldn't consider it a showstopper, but it seems this is used by at least one 
LHC experiment (which should have a workaround in place now ;-).

- Stephan

 
 Thanks
 Troy
 
 On 01/20/2011 11:30 AM, Troy J Dawson wrote:
 Hello,
 We have had our first kernel security update following the release of SL
 5.6.  We have tested it on a SL5.0 machine.  It installs, runs and
 openafs works on it.  I would feel much better if others ran it to make
 sure it works for them.
 
 Can others test this kernel out on their machines to make sure it
 doesn't break something we didn't expect.
 
 I have also put the new kvm into the x86_64 testing area with the kernel.
 
 To test or update
 
 SL5
 ---
 
   yum --enablerepo=sl-testing update kernel\*
 
 or you can download rpm's by hand at
 
 http://ftp.scientificlinux.org/linux/scientific/5rolling/testing/i386/kernel/
 http://ftp.scientificlinux.org/linux/scientific/5rolling/testing/x86_64/kernel/
 
 kernel-2.6.18-238.1.1.el5
 
 Thanks
 Troy Dawson

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: SL6 Beta1 ypbind selinux policy.

[Stephan Wiesand]

On Feb 2, 2011, at 00:34 , Don Krause wrote:

 Is selinux on a default install of SL6 Beta 1 supposed to prevent ypbind from 
 working?

Probably:

# getsebool -a |grep yp
allow_ypbind -- off

Does setsebool -P allow_ypbind on make it work?

- Stephan


 I'm getting this error in the audit.log
 
 type=USER_AVC msg=audit(1296601650.114:34350): user pid=2262 uid=81 
 auid=4294967295 ses=4294967295 
 subj=system_u:system_r:system_dbusd_t:s0-s0:c0.c1023 msg='avc:  denied  { 
 send_msg } for msgtype=method_call interface=org.freedesktop.NetworkManager 
 member=state dest=org.freedesktop.NetworkManager spid=4805 tpid=3995 
 scontext=unconfined_u:system_r:ypbind_t:s0 
 tcontext=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 tclass=dbus : 
 exe=/bin/dbus-daemon sauid=81 hostname=? addr=? terminal=?'
 
 When run through audit2allow, umm... damn, not found.. Hmm... Yeah, 
 policycoreutils is installed.. wtf?
 
 begin rant
 audit2allow was moved from policycoreutils to policycoreutils-python. Has it 
 become a game at TUV to see how many separate packages can be built from one 
 src.rpm?
 end rant
 
 Sorry, distracted for a moment..
 
 Anyway, after installing pcu-python for audit2allow, I get:
 
 module ypbind 1.0;
 
 require {
type unconfined_t;
type ypbind_t;
class dbus send_msg;
 }
 
 #= ypbind_t ==
 allow ypbind_t unconfined_t:dbus send_msg;
 
 
 which looks reasonable, but I'm not an selinux guru. 

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: SL6: OpenAFS + KRB; klist does not show AFS token...

[Stephan Wiesand]

 05:22:57
 
 Has anybody an idea how to track down this issue?
 
 Thanks a lot
 
 Andreas

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany






smime.p7s
Description: S/MIME cryptographic signature

Re: TESTING - kernel update for SL5

[Stephan Wiesand]

Hi Troy,

On Jan 20, 2011, at 18:30 , Troy Dawson wrote:

 Hello,
 We have had our first kernel security update following the release of SL 5.6. 
  We have tested it on a SL5.0 machine.  It installs, runs and openafs works 
 on it.  I would feel much better if others ran it to make sure it works for 
 them.

we deployed this kernel on a dozen SL5.5 systems, including a number of Xen 
DOM0s and DOMUs, on friday. No obvious problems have shown up (at least none we 
haven't experienced with earlier kernels as well;-)

- Stephan

 Can others test this kernel out on their machines to make sure it doesn't 
 break something we didn't expect.
 
 I have also put the new kvm into the x86_64 testing area with the kernel.
 
 To test or update
 
 SL5
 ---
 
 yum --enablerepo=sl-testing update kernel\*
 
 or you can download rpm's by hand at
 
 http://ftp.scientificlinux.org/linux/scientific/5rolling/testing/i386/kernel/
 http://ftp.scientificlinux.org/linux/scientific/5rolling/testing/x86_64/kernel/
 
 kernel-2.6.18-238.1.1.el5
 
 Thanks
 Troy Dawson
 --
 __
 Troy Dawson  daw...@fnal.gov  (630)840-6468
 Fermilab  ComputingDivision/LCSI/CSI DSS Group
 __

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany



smime.p7s
Description: S/MIME cryptographic signature

Re: default I/O Scheduler for SL 5.5 tuning question

[Stephan Wiesand]

Hi,

On Jan 5, 2011, at 13:30, g1vrg wrote:

 Hi, I have noticed that in default SL5.5 disk operations tend to hugely
 tie up my pc. For example when writing large files (of the order of 10
 GB) I may have to wait 30 seconds before I can get a response from
 another application on the desktop. Is there another i/o scheduler other
 than the default that I can specify presumably as a kernel command in
 the grub boot loader config file? I recently swapped over from debian
 lenny and the i/o scheduler there

which one was it?

You can change I/O schedulers per block device on the fly:

# cat /sys/block/sda/queue/scheduler
noop anticipatory deadline [cfq] 
# echo noop  /sys/block/sda/queue/scheduler
# cat /sys/block/sda/queue/scheduler
[noop] anticipatory deadline cfq 

I guess you get the idea. Let us know which one works best for you (and what 
hardware you have).

Regards
Stephan

 was much more equitable to the point
 where the desktop user didn't notice - it meant disk writes were slower
 but I can put up with that. Sorry if this has already been asked
 recently
 
 Thanks.
 
 Richard

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany

Re: Replacing Hard Disks with a logical disk names.

[Stephan Wiesand]

On Oct 15, 2010, at 17:02, martin@stfc.ac.uk wrote:

 Don't you fix the initial problem with a 'rescue' image?  Seem to recall
 doing this several times before on a variety on version of
 RedHat/Fedora/Scientific Linux.  Or am I misunderstanding?

Not unless I'm as well. Of course you don't even need a rescue disk, nor the 
root password, if you just know the grub password. ;-)

Stephan

 
   Martin.
 -- 
 Martin Bly
 RAL Tier1 Fabric Manager
 
 
 -Original Message-
 From: owner-scientific-linux-us...@listserv.fnal.gov
 [mailto:owner-scientific-
 linux-us...@listserv.fnal.gov] On Behalf Of Larry Linder
 Sent: Friday, October 15, 2010 1:51 PM
 To: scientific-linux-us...@fnal.gov
 Subject: Replacing Hard Disks with a logical disk names.
 
 A simple problem that I had done for years, turned  out to be
 difficult due to
 a mistake I made and what I believe is an error in the Linux OS.
 How you set it up is to forget to remove the logical drive from
 /etc/fstab
 in the past it was never a problem.
 But in SL 5.5 it is a serious problem because during boot it can't
 find the
 drive name.  It drops you to a maintenance level and all you used to
 do in
 put in the root pass word, edit the files etc.
 What happen now:
 put in your password
 bash  /usr/bin/id:   no such file or dir
 bash  [: =: unary operator expected
 bash  /usr/bin/id:
 bash [: =: unary operator expected
 bash /usr/bin/kpg-config: no such file or dir
 repair file system1
 As a result you can do nothing because your passwd has been rejected.
 
 You are back to using your install disks. It recognizes un initialized
 disks
 and initializes them - do a new install and set up disks and disk
 names and
 do not format anything, except new disk,  setup root / passwd, set up
 internet, do not install any thing.   and it knows there is an active
 OS
 present and the install aborts.
 
 The system reboots and runs normally everthing is preserved all
 because some
 security nit modified the code and never checked the end result.
 Sometime you can be so secure that the system becomes worthless.
 
 What used to be a simple thing of replacing disks has now been
 difficult at
 best.
 What I fixed is to get rid of the logical names in in the fstab and
 went back
 to the /dev/sda1 etc.   This was done because I didn't have a good way
 to
 look at disks and their names but knew the hardware.
 
 For back up on paper you need to do df and pipe it to lpr, keep in
 you file
 folder as a true back up.
 
 You can easily create this problem by simply unplugging a disk and
 trying a
 reboot.
 
 I have three backups but I never had a disk that was good but the
 electronics
 became intermittent as a function of temperature.  suspect a bad
 solder joint
 or circuit trace crack somewhere.   The symptom was a nice running
 drive that
 was sluggish.   A reboot solved the problem but the failures began to
 increase.   Users don't seem to understand a system being down.
 Some of these boxes are shut down ever six months for cleaning.
 
 Disks being cheep it time to install a new one and toss the old one.
 
 Larry Linder
 --
 Scanned by iCritical.

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany






smime.p7s
Description: S/MIME cryptographic signature

Re: Scientific Linux 3 - End of Life - Now

[Stephan Wiesand]

Hi Troy,

On Oct 10, 2010, at 17:10 , Troy Dawson wrote:

 That's right. as of right now Scientific Linux 3 is no longer supported.
 
 There will be no more security errata, enhancements, fastbugs, or bug fixes 
 for Scientific Linux 3.
 
 If you have not already moved off Scientific Linux 3, you need to migrate off 
 as soon as possible.
 
 The first release of Scientific Linux 3 was on May 10, 2004.  The official 
 day for it's end of support is October 10, 2010.

huge thanks for keeping SL3 going for those (almost) 6.5 years, and smoothly. 
Fine work, really. It was a major move for us to adopt SL back in 2004 - and I 
haven't regretted it.

Best regards
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany



smime.p7s
Description: S/MIME cryptographic signature

Re: openssh gssapi with proxycommand bug

[Stephan Wiesand]

Hi,

On Aug 24, 2010, at 15:45, Troy Dawson wrote:

 Hi,
 We've talked about this and feel it isn't something we have the resources to 
 maintain.  Please put this bug in with The Upstream Vendor (RedHat).

the best way to proceed would probably be to create a clone of #531849 against 
5.6? Jörgen, would you do it?

- Stephan

 
 I'm sorry.
 Troy
 
 Joergen Samson wrote:
 Hi,  there is a bug in SL5's openssh client which is introduced by RetHat's 
 openssh-4.3p2-gssapi-canohost.patch, if you use Kerberos5 authentication in
 conjunction with the ProxyCommand option.
 To verify the bug run
 ssh -v -o ProxyCommand nc %h %p -o PasswordAuthentication no -o
 PubkeyAuthentication no -o GSSAPIAuthentication yes  $HOST echo work
 s
 on a host which allows login with a Kerberos5 ticket.
 On SL5 openssh fails with [...]
 debug1: Next authentication method: gssapi-with-mic
 debug1: An invalid name was supplied
 Hostname cannot be canonicalized
 [...]
 With a vanilla build of openssh this command succeeds.
 The fedora project already uses a fixed version of th
 openssh-4.3p2-gssapi-canohost.path
 http://cvs.fedoraproject.org/viewvc/rpms/openssh/devel/openssh-4.3p2-gssa
 pi-canohost.patch?sortdir=downview=log
 Could you backport the fixed patch to the SL5 openssh packages?
 Cheers,
  Jörgen Samson
 
 
 -- 
 __
 Troy Dawson  daw...@fnal.gov  (630)840-6468
 Fermilab  ComputingDivision/LSCS/CSI/USS Group
 __

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany






smime.p7s
Description: S/MIME cryptographic signature

Re: ROOT on SL 4 and SL 5

[Stephan Wiesand]

Connie,

On Aug 24, 2010, at 17:06 , Connie Sieh wrote:

 Are there prebuilt rpms and src.rpm of ROOT (http://root.cern.ch) for both SL 
 4 and SL 5 for version 5.26 available?  I have looked for rpms at Dag, EPEL , 
 http://root.cern.ch without success.  There have been a few references to a 
 src rpm but the person who made the spec file is not supporting it any more.


a minute after your mail, I received the announcement of Fedora 14 alpha - and 
it's supposed to come with ROOT. No clue who packaged it, but that src.rpm is 
probably a reasonable starting point.

I have my doubts regarding a generally useful ROOT build, though: This software 
has tons of build time options, often depending on external software. A one 
size fits all build would probably depend on a dozen or more additional 
packages most users won't care for. On the other hand, building ROOT for a 
certain use case is fairly trivial. This may explain why there are few prebuilt 
packages around.

Another issue is that users tend to create code that works with a certain ROOT 
version only, and any update - even within a major version - is likely to cause 
grief for some of those depending on the previous minor release. At the same 
time, there will be other users requesting the very latest release as soon as 
it's available. Hence, unless you're catering for a small and well defined user 
community, you inevitably end up providing multiple releases at the same time, 
and some mechanism to switch between them. I wonder whether the Fedora package 
provides this...

The bottom line is that providing tailored builds for those actually requesting 
them is probably the best way to go. Teaching them to create those builds 
themselves may make even more sense.

The ROOT release schedule has slowed down recently, so things may be changing. 
But I don't think that a prebuilt ROOT rpm is a general solution yet. 

Hope this helps,
Stephan

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: TESTING - openafs update for SL5

[Stephan Wiesand]

On Jul 8, 2010, at 19:11 , Dag Wieers wrote:

 On Thu, 8 Jul 2010, Troy Dawson wrote:
 
 Dag Wieers wrote:
 On Thu, 8 Jul 2010, Troy Dawson wrote:
   With many minor releases, we update the version of openafs for that   
  minor release.  This new version then get's pushed out to the rest of   
  the releases.
   With SL 5.5 we updated openafs to 1.4.12, and we are about to push that 
version out to the rest of the SL5 releases.  It currently is in   
  testing, and it has passed every updating test I could think to throw at 
it and it updated without any problems.
We plan on pushing this out on Monday - 12 July 2010
To test or update
SL5
   ---
  yum --enablerepo=sl-testing update kernel-module-openafs\*
or you can download rpm's by hand at
   http:  
   //ftp.scientificlinux.org/linux/scientific/5rolling/testing/i386/openafs/
  http:  
  //ftp.scientificlinux.org/linux/scientific/5rolling/testing/x86_64/openafs/
 
 Would there be any interest if we provided kmod-openafs modules that are
 kernel-agnostic (or kABI-tracking as we say) from ELRepo ?
 
 The advantage is that the modules keep on working through kernel-updates,
 which makes update-cycles (and maintenance) to be less work.
 
 I am tempted to create those packages, but without an interested party
 that can provide sufficient testing the effort is kinda moot.
 
 Let me know,
 
 I thought that the openafs kernel modules didn't work well with kABI, but I 
 would love to find that incorrect.  If you think it is possible, please 
 build it, and I'm certain we'll have plenty of testers.
 
 If that is true we might have a discussion with Red Hat to see whether we can 
 have those symbols as part of the kABI whitelist. Let's find out :-)

There are symbols missing from the whitelist, so there was no way to use 
kABI-tracking modules cleanly. That being said, it probably would have 
worked. If someone has the time, it would be really interesting to force the 
module built for the SL5.0 GA kernel into -194.8.1 and see whether that works.

The guy in charge at Red Hat (Jon Masters) seems very openminded, so talking to 
them is certainly worth the effort. I have my doubts though whether there's any 
chance to have the whitelist extended while it still matters.

 For SL5, I'd like to stick with what we have with the supported release, but 
 I'm very sure that we would have plenty of users wiling to test and use the 
 kmod-openafs module.  If everything goes well, we could offer it as an 
 alternative.
 
 For SL6, if this works we could use that and save us from having to create 
 kernel modules with each kernel update.
 
 Sure, I don't want to force anyone anyway. A clean upgrade path will be very 
 hard due to the fact that these kernel-module packages have the 
 kernel-version in the name. So your position makes a lot of sense.

That's my point of view as well. SL5 should not drop the kernel-module packages 
(at least not anytime soon), but having kmods for testing would be very useful. 
With SL6, AFAIK, the whitelist problem is going to vanish, and IMO we should 
use kABI for the next major release if at all possible.

We should also make an effort to (re-)unite the SL/Elrepo/... packaging with 
the one from openafs.org. And Christof Hanke, who's crafting the OpenSuSE RPMs, 
also expressed interest in a unified spec during the European AFS Workshop in 
Rome last autumn. This is probably the time to actually try getting there. It 
may turn out that it's not feasible, but let's try.

Opinions?

 
 -- 
 --   dag wieers,  d...@wieers.com,  http://dag.wieers.com/   --
 [Any errors in spelling, tact or fact are transmission errors]

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Note - Firefox 3.6 comming today for SL5

[Stephan Wiesand]

On Jun 25, 2010, at 17:48 , Troy Dawson wrote:

 Simon Butcher wrote:
 Thanks Troy and Connie for the heads-up, and all the testing.
 Is this a new policy by T.U.V. to perform major version upgrades to fix a 
 major security problem instead of backporting security fixes into the old 
 version? It only seems to be recent behaviour (openoffice, firefox).
 simon
 
 This is a good question, and it depends on what you mean by new policy.  
 This is a policy that they started a couple of years ago (2 or 3 years), but 
 it isn't for all packages, only for a small set.

But the policy is (or was) to provide such possibly not quite backward 
compatible enhancements with minor releases only. Which in particular means, 
after a beta phase. Not with a critical security update that ought to be rolled 
out a.s.a.p.

IMO this firefox update is either a breach of policy, or indeed establishing a 
new one. It will hit our users' systems monday morning. I'm very curious what 
will happen then.

- Stephan

 When they originally said it, they listed firefox, openoffice, and I 
 *believe* evolution.  They said they were going to update them to the latest 
 release about once a year.  It turns out that it's actually taking them two 
 years.
 
 Here is how often they have done major updates for each of these
 
 SL4
 Firefox 1.0 - 1.5 : Aug. 2006
 Firefox 1.5 - 3.0 : Sep. 2008
 Firefox 3.0 - 3.6 : Jun. 2010
 
 SL5
 Firefox 1.5 - 3.0 : Jul. 2008
 Firefox 3.0 - 3.6 : Jun. 2010
 openoffice 2.0 - 2.3 : Jun. 2008
 openoffice 2.3 - 3.1 : Jun. 2010

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany

Re: Mem leak with fopen/fseek/fclose ?

[Stephan Wiesand]

On Jun 21, 2010, at 3:12, Khoa To wrote:

 I observed a memory leak (about 20 bytes per iteration) when I ran this 
 
 simple program on my Scientific Linux 5.4 box (kernel version 2.6.18-
 128.1.1.e15).

Pretty old kernel (the one from the 5.3 initial release). What's the AFS client 
version? (output of rxdebug localhost 7001 -v)

 What I did was noting the output of free -m before and after I ran the 
 
 program (which loops for 30). I observed a 6MB increase in memory use
 d.  
 No error was printed during the run.
 
 If I comment out the 2 fseek statements, then I don't see any memory 
 increase.

Check the amount of memory used for afs_inode_cache before and after the run 
(using slabtop). Does the difference account for the leak?
 
 The file size is about 700KB.
 
 I was wondering if someone can tell me if this is a known issue or if I'm
 
 doing something wrong.

Well, the code below is not what you're actually running since amp is 
undefined. And the whole test for the while() loop doesn't make too much sense 
to me (although the first statement could probably simply be omitted).

Regards,
Stephan

 
 Thanks,
 Khoa.
 
 ==Source file, compiled with gcc=
 
 #include stdio.h
 #include errno.h
 
 int main()
 {
   FILE * ptr = NULL;
   int count = 0;
 
   ptr = fopen(myBinaryFile,r);
   while (ptr != NULL amp; count++  30)
   {
usleep(1000);
 
if (fseek(ptr, 0, SEEK_END))
printf(Error 1: %d\n, errno);
if (fseek(ptr, 0, SEEK_SET))
printf(Error 2: %d\n, errno);
 
if (fclose(ptr))
break;
 
usleep(1);
 
ptr = fopen(myBinaryFile,r);
   }
 
fclose(ptr);
 
if (count = 30)
   print(Success\n);
 
return 0;
 }

-- 
Stephan Wiesand
DESY -DV-
Platanenallee 6
15738 Zeuthen, Germany






smime.p7s
Description: S/MIME cryptographic signature

Re: Memory footprint on 64bit SL vs. 32bit

[Stephan Wiesand]

On Apr 27, 2010, at 00:15 , Brett Viren wrote:

 We recently started running our C++ analysis code on 64bit SL5.3 and
 have been surprised to find the memory usage is about 2x what we are
 used when running it on 32 bits.  Comparing a few basic applications
 like sleep(1) show similar memory usage.  Others, like sshd, show only a
 30% size increase (maybe that is subject to configuration differences
 between the two hosts).
 
 I understand that pointers must double in size but the bulk of our
 objects are made of ints and floats and these are 32/64 bit-invariant.
 I found[1] that poorly defined structs containing pointers can bloat
 even on non-pointer data members due the padding needed to keep
 everything properly aligned.  It would kind of surprise me if this is
 what is behind what we see.
 
 Does anyone have experience in understanding or maybe even combating
 this increase in a program's memory footprint when going to 64 bits?

Is it real or virtual memory usage that's increasing beyond expectations?

Example: glibc's locale handling code will behave quite differently in the 
64-bit case. In 32-bit mode, even virtual address space is a scarce resource, 
while in 64-bit mode it isn't. So in the latter case, they simply mmap the 
whole file providing the info for the locale in use, while in the former they 
use a small address window they slide to the appropriate position. The 64-bit 
case is simpler and thus probably less code, more robust and easier to 
maintain. And it's probably faster. The 32-bit case uses less *virtual* memory 
- but *real* memory usage is about the same, since only those pages actually 
read will ever be paged in. This has a dramatic effect on the VSZ of hello 
world in python. It does not on anything that really matters - in particular, 
checking the memory footprints of sleep  co. is not very useful because 
they're really small compared to typical HEP analysis apps anyway.

What are your actual figures?

 Thanks,
 -Brett.
 
 [1] http://www.codeproject.com/KB/winsdk/Optimization_64_bit.aspx#IDAJLKNC

-- 
Stephan Wiesand
DESY -DV-
Platanenenallee 6
15738 Zeuthen, Germany



smime.p7s
Description: S/MIME cryptographic signature