Re: How to migrate yum installed on RHEL 6 to update from SL6 repository?
On Wed, Apr 27, 2011 at 5:06 PM, Zoran Ovcin wrote: > Recently I installed RHEL6 from an installation DVD (not Beta). > > I was aware that my yum update will not get the RHN support. This will *BREAK* things, such as "redhat-release". Don't do it. Migrate the installed packages to SL6 first, especially the "*-release" packages. Also, rip out "yum-rhn-plugin". This will turn off the attempts to access the upstream RHN repositories. > Is it possible to use the SL6 rpm repository for updating my system? > > If yes, what do I have to change in order to be able to do yum update? It's a potentially nasty interaction: packages of the same name may have subtle discrepancies, and tools that look for /etc/issue.net contents will be confused at compilation time. > Thanks, Zoran Ovcin >
Re: How to convert a standar apps, to a Service - WORKING!!
Thanks a lot, for everyone. The Ultra VNC Repeater, release 17 include the scripts and sources to make the compiling, and the installing. But, the problem is that in Scientific Linux, have not the "start-stop-deamon" apps. Downloanding and compiling this, and adding the "start-stop-deamon" binary executable to the /sbin directory, all work so fine. A few configuration on the uvncrepeater.ini and All Rigths. If someone need any more details, to make this implementation, only tell me how I can help you. Best Regards, Pablo Cavero. System Engineer 2011/4/27 Florian Philipp > Am 27.04.2011 22:54, schrieb Alec T. Habig: > > Pablo Cavero writes: > >> The Only that I need is a set of bash script, like script used by the > >> "tomcat", for example, or need a special binary executable?? > >> http://www.laliluna.de/articles/tomcat-startup-script-linux.html > >> > >> I want to may will use chkconfig, like an standar service. > > > > The chkconfig script part is easy, just look in /etc/rc.d/init.d, steal > > an existing service's script, and modify it to run what you want. Note > > the commented line at the top starting with "chkconfig" - that > > information specifies the run levels chkconfig will set it up to be > > activated in, and the startup/shutdown priority. > > > > As for the code you run, that gets trickier. All the startup scripts do > > is start and stop things. Those "things" could be more scripts, > > executables, or combinations thereof. Areas to pay attention to with > > your code (to get it to behave nicely when run this way) include: > > getting it daemonized properly (dropping and/or redirecting IO streams), > > privilege seperation (can it run as a non privleged user? if so, do > > that, if not, be really careful and have only the bare minimum code run > > with elevated privileges), logging (have it log status information, > > ideally with customizeable log levels, to the appropriate place), and > > signal handling so it shuts down cleanly when told to, reloads > > configuration files if asked, etc. > > > > There is also libdaemon which is meant to facilitate converting your > program into a proper daemon (not tried it, though). BTW: It is a shame > that SL does not contain the start-stop-daemon like Gentoo or Ubuntu > ( > http://manpages.ubuntu.com/manpages/dapper/en/man8/start-stop-daemon.8.html > ). > It makes converting arbitrary executables into daemons a matter of minutes. > > Regards, > Florian Philipp > > -- Pablo Cavero System Engineer +569 8920 9509
submitting bugs OR an apparent bug in ifup
I've just run into an issue with 6.0 and I'm not sure how bugs are handled here. Should they just be submitted upstream to RH? Specifically, ifup isn't behaving as expected bringing up an aliases. /etc/sysconfig/network-scripts/ifcfg-lo:0 DEVICE=lo:0 IPADDR=10.0.0.1 NETMASK=255.255.255.255 NAME=lo:0 Running ifup lo:0 does not add the alias to lo as expected. However, if "ifup lo" is ran, it will add both the lo config and any aliases it finds. This config will result in the expected behavior after a reboot or a "service network restart". 5.x hosts (and earlier, at least in CentOS) behave as expected allowing the operator to add/remove aliases using ifup/ifdown. -- Kelsey Cummings - k...@corp.sonic.net sonic.net, inc. System Architect 2260 Apollo Way 707.522.1000 Santa Rosa, CA 95407
Re: SciPy in SL6 Success
Jean-Francois Caron wrote on 04/27/2011 06:04 AM:
Tired of waiting for a SciPy package to come out for SL6, I tried to
compile a version and was surprised to find success.
The Fedora 14 SRPM builds on SL6 after satisfying two dependencies from
the same source, and replacing ufsparse-devel with suitesparse-devel in
the BuildRequires. (A mystery to me why the FC14 spec does not already
have that.)
http://download.fedora.redhat.com/pub/fedora/linux/releases/14/Everything/source/SRPMS/
Build order:
tbb-2.2-2.20090809.fc14.src.rpm
suitesparse-3.4.0-2.fc12.src.rpm
scipy-0.7.2-2.fc14.src.rpm
It imports in python with no errors.
P.S. Possibly of interest - python-matplotlib-1.0.0-2.fc14.src.rpm also
builds with a minor kludge to the spec:
$ diff -u python-matplotlib.spec_orig python-matplotlib.spec
--- python-matplotlib.spec_orig 2010-07-22 22:38:51.0 -0400
+++ python-matplotlib.spec 2011-04-27 17:18:38.092666719 -0400
@@ -68,9 +68,9 @@
%doc license/LICENSE_PAINT license/LICENSE_PIL
%doc CHANGELOG CXX INSTALL INTERACTIVE KNOWN_BUGS
%doc PKG-INFO TODO examples
-%if 0%{?fedora} >= 9
+#%if 0%{?fedora} >= 9
%{python_sitearch}/*egg-info
-%endif
+#%endif
%{python_sitearch}/matplotlib/
%{python_sitearch}/mpl_toolkits/
%{python_sitearch}/pylab.py*
Perhaps EPEL could be persuaded to provide these.
Phil
Re: How to convert a standar apps, to a Service.
Am 27.04.2011 22:54, schrieb Alec T. Habig: > Pablo Cavero writes: >> The Only that I need is a set of bash script, like script used by the >> "tomcat", for example, or need a special binary executable?? >> http://www.laliluna.de/articles/tomcat-startup-script-linux.html >> >> I want to may will use chkconfig, like an standar service. > > The chkconfig script part is easy, just look in /etc/rc.d/init.d, steal > an existing service's script, and modify it to run what you want. Note > the commented line at the top starting with "chkconfig" - that > information specifies the run levels chkconfig will set it up to be > activated in, and the startup/shutdown priority. > > As for the code you run, that gets trickier. All the startup scripts do > is start and stop things. Those "things" could be more scripts, > executables, or combinations thereof. Areas to pay attention to with > your code (to get it to behave nicely when run this way) include: > getting it daemonized properly (dropping and/or redirecting IO streams), > privilege seperation (can it run as a non privleged user? if so, do > that, if not, be really careful and have only the bare minimum code run > with elevated privileges), logging (have it log status information, > ideally with customizeable log levels, to the appropriate place), and > signal handling so it shuts down cleanly when told to, reloads > configuration files if asked, etc. > There is also libdaemon which is meant to facilitate converting your program into a proper daemon (not tried it, though). BTW: It is a shame that SL does not contain the start-stop-daemon like Gentoo or Ubuntu (http://manpages.ubuntu.com/manpages/dapper/en/man8/start-stop-daemon.8.html). It makes converting arbitrary executables into daemons a matter of minutes. Regards, Florian Philipp signature.asc Description: OpenPGP digital signature
Re: SL6 site/spin building
2 - You can make the Install DVD so that it will force the installer to use your kickstart file. That way they will get the same set of packages that you selected for them. I was also thinking this, as far as I know, you can actually have scripts run from this can you not? I am not entirely familiar with kickstart but I have used it a number of occasions and it made multiple installations AMAZINGLY easy. -Chris
How to migrate yum installed on RHEL 6 to update from SL6 repository?
Recently I installed RHEL6 from an installation DVD (not Beta). I was aware that my yum update will not get the RHN support. Is it possible to use the SL6 rpm repository for updating my system? If yes, what do I have to change in order to be able to do yum update? Thanks, Zoran Ovcin
Re: SL6 site/spin building
On 04/27/2011 02:51 PM, Troy J Dawson wrote: On 04/27/2011 11:12 AM, James E. Davies wrote: ... If anyone has the time, please can they post some advice on whether SL6 site/spin building would work for me in the following context: I have a task that is on-going (like perpetual) meteorological data processing. I know what tools I need/don't need for this system. I'd like to build a system based on SL6 - which includes standard repository stuff plus a mix of my own code and non-repository tarball type installs (both binary and make from src) - then to build a distribution that somehow includes all this stuff. The aim is to have a distribution (as DVD-iso) to install on other machines (and that I can send to colleagues) without having to go through the hassle of finding and installing the "extras". Does this sound like what SL6 site- (now called "spin-") building is for? Am I on the right track - or is there a simpler/smarter way? I thought about gentoo and its system makefile type approach - but I don't think I want the bother of building absolutely everything - and I am quite sure my colleagues don't. To me, what you want to do sounds exactly what SL Spins (or Sites) are for. It sounds like you have already found the "How to build a site" webpage for SL6. http://www.scientificlinux.org/distributions/6x/build/sites I have gotten very little feedback on that page, so if you need something clarified, please let me know. OK, so my mind got to thinking of things. Just because something is on the Installation DVD, that doesn't mean it automatically get's installed. There are two ways to have the packages you added automatically get installed. 1 - Change the comps.xml file, either the main SL one, or make one for your smaller repository that just has your own stuff. Put the "extras" you want to be installed all the time in the "Base" or even the "Core" group. That way, no matter what they pick, they will get your "extras" 2 - You can make the Install DVD so that it will force the installer to use your kickstart file. That way they will get the same set of packages that you selected for them. Troy -- __ Troy Dawson daw...@fnal.gov (630)840-6468 Fermilab ComputingDivision/SCF/FEF/SLSMS Group __
Re: How to convert a standar apps, to a Service.
Pablo Cavero writes: > The Only that I need is a set of bash script, like script used by the > "tomcat", for example, or need a special binary executable?? > http://www.laliluna.de/articles/tomcat-startup-script-linux.html > > I want to may will use chkconfig, like an standar service. The chkconfig script part is easy, just look in /etc/rc.d/init.d, steal an existing service's script, and modify it to run what you want. Note the commented line at the top starting with "chkconfig" - that information specifies the run levels chkconfig will set it up to be activated in, and the startup/shutdown priority. As for the code you run, that gets trickier. All the startup scripts do is start and stop things. Those "things" could be more scripts, executables, or combinations thereof. Areas to pay attention to with your code (to get it to behave nicely when run this way) include: getting it daemonized properly (dropping and/or redirecting IO streams), privilege seperation (can it run as a non privleged user? if so, do that, if not, be really careful and have only the bare minimum code run with elevated privileges), logging (have it log status information, ideally with customizeable log levels, to the appropriate place), and signal handling so it shuts down cleanly when told to, reloads configuration files if asked, etc. -- Alec Habig, University of Minnesota Duluth Physics Dept. ha...@neutrino.d.umn.edu http://neutrino.d.umn.edu/~habig/
How to convert a standar apps, to a Service.
Hello, I have an apps, and I want to convert this to a service. I have the source in C, and the binary executable. Think for example, in the "ultra vnc repeater". http://sourceforge.net/projects/ultravnc/files/Repeater/win32%20and%20unix%20Repeaters/ The Only that I need is a set of bash script, like script used by the "tomcat", for example, or need a special binary executable?? http://www.laliluna.de/articles/tomcat-startup-script-linux.html I want to may will use chkconfig, like an standar service. Thank a lot, for your comments, -- Pablo Cavero System Engineer +569 8920 9509
Re: procedure for reporting kernel bugs? (sata_sil24+md+dm+jbd+ext4 hang)
On Wed, Apr 27, 2011 at 12:41 PM, Connie Sieh wrote: > On Wed, 27 Apr 2011, Akemi Yagi wrote: >> I would recommend you file a bugzilla at https://bugzilla.redhat.com >> even if you are not a customer. This is a bug report, not a support >> request. We contribute upstream by reporting bugs. :-) > > But do not reference Scientific Linux . Ah, actually that is an important piece of information (especially for those who came from CentOS). In CentOS, the devs encourage users to refer to CentOS when filing a bugzilla report upstream. TUV's dropdown menu for "external bugs" lists the 'CentOS bug tracker' to make the referencing easy. Akemi
Re: SL6 site/spin building
On 04/27/2011 11:12 AM, James E. Davies wrote: ... If anyone has the time, please can they post some advice on whether SL6 site/spin building would work for me in the following context: I have a task that is on-going (like perpetual) meteorological data processing. I know what tools I need/don't need for this system. I'd like to build a system based on SL6 - which includes standard repository stuff plus a mix of my own code and non-repository tarball type installs (both binary and make from src) - then to build a distribution that somehow includes all this stuff. The aim is to have a distribution (as DVD-iso) to install on other machines (and that I can send to colleagues) without having to go through the hassle of finding and installing the "extras". Does this sound like what SL6 site- (now called "spin-") building is for? Am I on the right track - or is there a simpler/smarter way? I thought about gentoo and its system makefile type approach - but I don't think I want the bother of building absolutely everything - and I am quite sure my colleagues don't. To me, what you want to do sounds exactly what SL Spins (or Sites) are for. It sounds like you have already found the "How to build a site" webpage for SL6. http://www.scientificlinux.org/distributions/6x/build/sites I have gotten very little feedback on that page, so if you need something clarified, please let me know. Troy -- __ Troy Dawson daw...@fnal.gov (630)840-6468 Fermilab ComputingDivision/SCF/FEF/SLSMS Group __
Re: procedure for reporting kernel bugs? (sata_sil24+md+dm+jbd+ext4 hang)
On Wed, 27 Apr 2011, Akemi Yagi wrote: On Wed, Apr 27, 2011 at 10:53 AM, Scott Mcdermott wrote= : SL users, we encountered a hung kernel with SL6, believe it is likely a real bug with dm code running snapshots. =A0 =A0sata_sil24 -> md raid5 -> dm snapshot -> ext4 jbd -> hung IO checked whole site sl.org and most documents there, but found no information on how to report bugs. =A0- since this is an "old" kernel relative to upstream, =A0 =A0doubtful that linux-kernel cares =A0- since this is not a "vendor" kernel i.e. =A0paid =A0 =A0rhel6, doubt redhat cares thus, not sure of the appropriate bug reporting method. I have tons of debug info... able to capture copious sysrq-induced dumps from serial console. Please advise on the appropriate procedure for reporting these kinds of kernel bugs. I would recommend you file a bugzilla at https://bugzilla.redhat.com even if you are not a customer. This is a bug report, not a support request. We contribute upstream by reporting bugs. :-) But do not reference Scientific Linux . -connie sieh Speaking of a sata_sil-related bug, there used to be one as seen in this CentOS bug tracker: http://bugs.centos.org/view.php?id=3D3165 The CentOSPlus kernel had a fix for it for years until it was eventually fixed upstream: https://bugzilla.redhat.com/show_bug.cgi?id=3D640586 Akemi
Re: OT: What kinds of Linux use business? (Was: Re: What kins of business use Linux?)
On 2011/04/27 04:08, Bluejay Adametz wrote:
What are the most popular distributions in business environment by list opinion?
Around here, a large manufacturing site, we use RedHat EL where we
need contract support (or we're running something that requires
Authentic RedHat in order for that vendor to support us), Scientific
Linux where we don't need that support.
SL has a slight lead over RedHat for our production systems. Then
there are a number of workstations and test machines that are running
SL.
- Bluejay Adametz
Um, somehow I don't think that question quite parses if you look at it
critically. And this is not the best place for a a question asking "what
kind of Linux do Linux user businesses use?" I'd think Distrowatch would
be a better place.
A smart business uses the tools that work best for them. I use Linux,
probably migrating to SL, for some infrastructure tools. I use Windows
because that is where I make my money. Trust me. It really IS nicer
than waiting tables or cleaning houses for a living. I write software.
The kind of software I write has little or no Linux market. So I follow
the money. Sensible businesses do the same. "I need to . The best
tool for is . And runs on OS on machines. So
I run machines with OS running for ing." I refuse
to get that order mixed up. If I did I'd find myself uncompetitive
and in some cases feeding spammers off my idle machine cycles.
{o.o} Joanne Dow
Re: procedure for reporting kernel bugs? (sata_sil24+md+dm+jbd+ext4 hang)
On Wed, Apr 27, 2011 at 10:53 AM, Scott Mcdermott wrote: > SL users, > > we encountered a hung kernel with SL6, believe it is > likely a real bug with dm code running snapshots. > > sata_sil24 -> md raid5 -> dm snapshot -> ext4 jbd -> hung IO > > checked whole site sl.org and most documents there, but > found no information on how to report bugs. > > - since this is an "old" kernel relative to upstream, > doubtful that linux-kernel cares > > - since this is not a "vendor" kernel i.e. paid > rhel6, doubt redhat cares > > thus, not sure of the appropriate bug reporting method. > I have tons of debug info... able to capture copious > sysrq-induced dumps from serial console. > > Please advise on the appropriate procedure for > reporting these kinds of kernel bugs. I would recommend you file a bugzilla at https://bugzilla.redhat.com even if you are not a customer. This is a bug report, not a support request. We contribute upstream by reporting bugs. :-) Speaking of a sata_sil-related bug, there used to be one as seen in this CentOS bug tracker: http://bugs.centos.org/view.php?id=3165 The CentOSPlus kernel had a fix for it for years until it was eventually fixed upstream: https://bugzilla.redhat.com/show_bug.cgi?id=640586 Akemi
procedure for reporting kernel bugs? (sata_sil24+md+dm+jbd+ext4 hang)
SL users, we encountered a hung kernel with SL6, believe it is likely a real bug with dm code running snapshots. sata_sil24 -> md raid5 -> dm snapshot -> ext4 jbd -> hung IO checked whole site sl.org and most documents there, but found no information on how to report bugs. - since this is an "old" kernel relative to upstream, doubtful that linux-kernel cares - since this is not a "vendor" kernel i.e. paid rhel6, doubt redhat cares thus, not sure of the appropriate bug reporting method. I have tons of debug info... able to capture copious sysrq-induced dumps from serial console. Please advise on the appropriate procedure for reporting these kinds of kernel bugs. Thanks. -- Scott
Re: SL6 site/spin building
> One final question: Window managers can be a drag on machines intended > primarily as compute engines - so I am tempted leave them out - but sometimes > I get advice on problem solving that assumes GUI tools are available. Do > you think iceWM is a good solution to this dilemma? I have a lot of servers that I don't even install X windows on, much less a window manager. On those rare occasions when I need to run a GUI tool on them, I ssh -X into the server from my workstation (creating tunnel for X) and display on my workstation. You just need the xorg-x11-xauth package for this to work, and this does require some bandwidth... - Bluejay Adametz "It is practice of the right kind that makes perfect." - Anonymous NOTICE: This message, including any attachments, is only for the use of the intended recipient(s) and may contain confidential and privileged information, or information otherwise protected from disclosure by law. If the reader of this message is not the intended recipient, you are hereby notified that any use, disclosure, copying, dissemination or distribution of this message or any of its attachments is strictly prohibited. If you received this message in error, please contact the sender immediately by reply email and destroy this message, including all attachments, and any copies thereof.
SL6 site/spin building
First post - please tell me if this is not what list is intended for... I installed CentOS5.6 - but I have a minor problem using usb stick as root dir - so was going to wait for CentOS6 and re-install differently. In the interim I found SL, which I did not know about. If anyone has the time, please can they post some advice on whether SL6 site/spin building would work for me in the following context: I have a task that is on-going (like perpetual) meteorological data processing. I know what tools I need/don't need for this system. I'd like to build a system based on SL6 - which includes standard repository stuff plus a mix of my own code and non-repository tarball type installs (both binary and make from src) - then to build a distribution that somehow includes all this stuff. The aim is to have a distribution (as DVD-iso) to install on other machines (and that I can send to colleagues) without having to go through the hassle of finding and installing the "extras". Does this sound like what SL6 site- (now called "spin-") building is for? Am I on the right track - or is there a simpler/smarter way? I thought about gentoo and its system makefile type approach - but I don't think I want the bother of building absolutely everything - and I am quite sure my colleagues don't. One final question: Window managers can be a drag on machines intended primarily as compute engines - so I am tempted leave them out - but sometimes I get advice on problem solving that assumes GUI tools are available. Do you think iceWM is a good solution to this dilemma? Thanks,Jim.
Re: TESTING - glibc security update fix for SL5
On 04/26/2011 10:58 PM, Jon Peatfield wrote: On Tue, 12 Apr 2011, Troy Dawson wrote: Hello, There was a bug with the latest glibc update that went out for SL5. https://bugzilla.redhat.com/show_bug.cgi?id=693882 It causes evolution to fail, and the gnome panel to crash. ... We do not plan on pushing this out, we are waiting for a fix from TUV. ... I see a rumour that https://rhn.redhat.com/errata/RHBA-2011-0466.html glibc update may fix this. Not that I've had a chance to test it yet... -- Jon I have verified that it fixes the evolution crash. I never had a machine where the gnome panel crashed. So I am going to put this into testing, so that people can let me know. If it fixes the gnome panel crash as well as the evolution crash, we will push it out in the security errata updates. Troy -- __ Troy Dawson daw...@fnal.gov (630)840-6468 Fermilab ComputingDivision/SCF/FEF/SLSMS Group __
Re: OT: What kinds of Linux use business? (Was: Re: What kins of business use Linux?)
> What are the most popular distributions in business environment by list > opinion? Around here, a large manufacturing site, we use RedHat EL where we need contract support (or we're running something that requires Authentic RedHat in order for that vendor to support us), Scientific Linux where we don't need that support. SL has a slight lead over RedHat for our production systems. Then there are a number of workstations and test machines that are running SL. - Bluejay Adametz "It's bad luck to be superstitious." - Andrew W. Mathis NOTICE: This message, including any attachments, is only for the use of the intended recipient(s) and may contain confidential and privileged information, or information otherwise protected from disclosure by law. If the reader of this message is not the intended recipient, you are hereby notified that any use, disclosure, copying, dissemination or distribution of this message or any of its attachments is strictly prohibited. If you received this message in error, please contact the sender immediately by reply email and destroy this message, including all attachments, and any copies thereof.
SciPy in SL6 Success
Tired of waiting for a SciPy package to come out for SL6, I tried to compile a version and was surprised to find success. I have all standard SL6 packages for the dependencies, only the SciPy source itself has to be downloaded. This means that to follow my steps you'll need to install python, numpy, gcc, gfortran, whatever BLAS/ATLAS you want. I basically went through the Add/Remove packages GUI and clicked anything that looked like it might be a SciPy dependency. I highly recommend the IPython shell too, but I guess it's not necessary. Since TUV shipped an older version of NumPy, you can't use the latest SciPy. The latest one that I found to work was http://sourceforge.net/projects/scipy/files/scipy/0.7.1/ , though I had to dig a bit to find version depency information, so I might have missed a slightly higher version that also would work. My instructions will assume not much familiarity with compiling code. Download the tar archive to a suitable location, like ~/Software. This location should be where scipy will "live" in the future, so don't put it on your desktop. Issue the command "tar -xzvf scipy-0.7.1.tar.gz" which will extract everything to ~/Software/scipy-0.7.1. I like to move the original .tar.gz file into this directory as well. >From your scipy-0.7.1 directory, issue "python setup.py build". This will perform the actual compilation. It took several minutes on my modern quad-core machine. >From the same directory, but now as root, issue "python setup.py install". If you don't have root access on your system, you will need to use an option like --prefix=/home/user/wherescipywillive, but then you'll also need to tell your python interpreter to look there for scipy. Now in your python shell, you can "import scipy" and try to run "scipy.test()". Mine returned a small number of errors, but the error reports are not formatted for humans, so I'm willing to call it a success. When a packaged SciPy eventually comes out for SL6, I do not know how to properly undo the steps outlined above. The setup.py script doesn't come with an "uninstall" function, so some manual cleaning up will probably be required. Jean-François
