June 30th, 2009
It all started in the CentOS forums.
CentOS, being an Enterprise Class Linux distro, may not have support for new hardware, especially those found in desktops and laptops. CentOS forums get many postings from users who are looking for drivers that are needed for their network device, webcam, etc. In fact, googling for “driver” and “centos” yields much more hits from the forums than the mailing list as shown in my earlier blog.
Alan Bartlett who has been actively helping in the forum has strong background in the kernel and its modules and has been building, whenever possible, a driver module each time he sees a cry for help. In a fairly short period of time, quite an impressive number of such kernel modules were built this way and it was growing.
Because these drivers were designed to be independent of the kernel version (kABI-tracking, kABI = kernel application binary interface), they survive kernel updates. Also, they can be used in RHEL and all its rebuilds, not just CentOS. So, it was a good idea to make them available to wider communities. NedSlider proposed to establish a repository for that purpose. The ELRepo project was thus started by five members of the CentOS / Scientific Linux community.
A little more details of ELRepo can be found in the Ned’s recent blog.
Categories: centos |
No Comments
May 29th, 2009
One of the most commonly asked questions in the CentOS forums is concerned with 3rd party repositories. In addition to referring to the Repositories wiki article, we need to make sure people use the yum priorities plugin appropriately.
Often times, we want to know how they set up the priority scores but there is no easy way to display them. yum repolist all does not include this info. A forum helper jlehtone posted a one-liner that does just that:
cat /etc/yum.repos.d/*.repo | sed -n -e "/^\[/h; /priority *=/{ G; s/\n/ /; s/ity=/ity = /; p }" | sort -k3n
The output may look like this:
priority = 10 [base]
priority = 10 [updates]
priority = 30 [extras]
priority = 50 [centosplus]
priority = 60 [c5-testing]
priority = 70 [rpmforge]
priority = 75 [kbs-CentOS-Testing]
priority = 90 [kbs-CentOS-Misc-Testing]
priority = 94 [epel]
Almost immediately after the above post, Alan the perfectionist, a cat person himself, opposed the use of over-working cats and came up with his version of the one-liner:
sed -n -e "/^\[/h; /priority *=/{ G; s/\n/ /; s/ity=/ity = /; p }" /etc/yum.repos.d/*.repo | sort -k3n
Personally … I want to see the cats working for the community … 
Categories: centos |
No Comments
May 29th, 2009
A while ago, I was talking with my buddies and the subject was about installing / updating a package using yum. yum did not find the package that had just been added. No problem, either yum clean metadata or yum clean all will do. Err … it didn’t.
As it turned out, it didn’t work because the repository in question was disabled by default. The man page for yum clearly states:
Note that "all files" in the commands below means "all files in currently enabled repositories". If you want to also clean any (temporarily) disabled repositories you need to use --enablerepo=’*’ option.
Categories: centos |
No Comments
April 12th, 2009
For some reason, my desktop gnome menu had a “Suspend” option. It was right under the “Log out …” option in the System dropdown menu. One day, when I meant to click on the “Log out” option, my mouse went too far and I ended up selecting this “Suspend” option.
After struggling for the next few minutes, I gave up and rebooted the machine to come out of the suspended state.
The Suspend option should not have been there to begin with. Anyway, it was time to remove it. Fortunately, that was as easy as typing [1]:
gconftool-2 --type boolean --set /apps/gnome-power-manager/can_suspend false
Likewise, if you want to get rid of the Hybernate option,
gconftool-2 --type boolean --set /apps/gnome-power-manager/can_hibernate false
[1] http://magazine.redhat.com/2008/05/30/tips-and-tricks-how-do-i-remove-the-suspendhibernate-entries-from-the-system-menu-in-the-gnome-panel/
Categories: centos |
No Comments
March 19th, 2009
My first encounter with kernel modules was when my systems running Fedora Core 5 started crashing after the kernel update to 2.6.18. Long story short, it was due to a bug in the cifs kernel module which was fixed after more than 100 e-mail exchanges with samba developers. The problem was solved, but there was something I never understood; the patched cifs module I built was much larger than the distro’s. This is always true. A custom-built version can be more than 10-times larger than its counter part in the distro kernel.
I never pursued this issue until the other day when wolfy and Ned started a conversation on IRC about this size difference. It is not due to stripping because the distro version is “not stripped” and, in fact, if you strip it the module no longer loads.
Then later, Alan asked in his clever voice, “have you checked the options to strip? It might be that a total strip is verboten but a selective strip works?”
That was it! 
I tried a strip –strip-unneeded command and that reduced the size of my module file to something close to the distro’s AND the file remained “not stripped“. Soon after, Ned found the –strip-debug option does the trick, too. This is probably what is used when the kernel is built and modules compiled.
Mystery solved.
Categories: centos |
2 Comments
February 27th, 2009
There are several ways of getting help in CentOS as summarized on the Wiki Help page. The most commonly used CentOS-provided venues are the mailing lists and the forums. However, when people encounter problems or simply want to find out some information, they most likely just go to google.
I performed some searches on google for commonly used terms with the “site:centos.org” added and then counted the number of hits that pointed to the mailing lists (M/L) and did the same for the forums. The result is summarized below. Assuming most people will not go through many pages, I collected the results from the first 50 links (that’s 5 pages with the default of 10 per page).
Search term M/L Forums
============ === ======
install 0 21
installation 0 4
kernel 10 11
driver(s) 3 24
DNS 11 29
postfix 19 20
sendmail 22 16
selinux 5 10
apache 0 23
httpd 6 30
xen 38 7
kickstart 20 12
crash 16 11
panic 10 19
While I do not claim this is a representative result, I do notice that the forum posts tend to be found more often than the M/L posts. One notable exception (among this small set of searches) was “xen” — most of them were from the centos-virt mailing list (what else do you expect? ).
Also, I am not trying to compare the significance/importance of the two help channels, either. I only want to emphasize the fact that, in the age of google everything, many users come to read those posts and, therefore, we need to maintain the contents of the posts as accurate as we can.
I’ve heard about a plan to merge postings from both channels into a single database. That would become a valuable souce for all CentOS users.
Categories: centos |
1 Comment
January 14th, 2009
Last year, we saw many people who migrated to CentOS. There is no doubt we have many more CentOS “newcomers” this year. There are basically two types of CentOS newbies.
One is former/current Windows users who just want to see what Linux looks like or who are seriously considering saying goodbye to the blasted OS. These people need a lot of help but, as far as they are eager to learn, it’s not difficult to provide assistance.
The other type is those who have been using other distros for some time. They tend to demand the latest and the greatest version of everything. It is often necessary to explain all about an enterprise-class distro, stability and backporting. A recent post in the CentOS forum is from a person of such type:
“I guess I will go with a more updated and modern O/S.
Centos must be years behind other O/S”
Another concept that is often not so obvious among the CentOS newcomers is the package-based management. As soon as they find out their favorite application is not available or obsolete, they want to install it from source. Let me pull some example again from a recent post in the forums (with a slight adjustment):
“I was thinking removing the original version (e.g. yum remove glibc) and install the new version using /usr/local as the root path could be a better solution for the system stability.”
This poster is even talking about the stability ! However, this is not a laughing matter. We should all get busy and provide proper guidance to people who are serious about doing the migration.
Welcome to the CentOS community !
Categories: centos |
No Comments
December 28th, 2008
Jim Perrin put an impressive ASCII Christmas tree on Planet CentOS. This reminded me of my own very old ASCII snowman I used to place in motd of a Solaris workstation during holiday seasons.
-rw-r----- 1 yagi2 yagi 661 Dec 17 1998 holi
$cat holi
_XXX_ . . . . . . . . .
/ """ \ . . . . . . . . .
{ 0 0 3 . . . . . .
{ < 3 . . *** HAPPY HOLIDAYS *** . .
\ V / . . . . . . .. . . . .
&&&&&&& .. . . . . . . . .
/ & \ . .. . . . . . .
/ & \ . . . . . .. . . .
BB & ))) . . . . . . . .
\ /
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Categories: centos |
No Comments
December 28th, 2008
Like many other people, I use Linux as a backup server. The other day, I noticed that daily incremental backup of one of the Windows machines was well over 1 GB even on the day the user was mostly idle. The only thing the user was doing was … e-mailing. Aha! (heard the bell?) It must be that inflated Inbox.
Mozilla-based mail clients like Thunderbird and Seamonkey mail do not physically remove messages that user deletes. Instead they are only tagged “deleted”. This is true even after the Trash folder is emptied. The [supposedly] deleted mails get [really] deleted when Inbox (or any folder for that matter) is compacted.
I went to the blasted machine and did just that and the Inbox went from > 1 GB to a fraction of its original size.
Of course, this is not just Windows. Huge mail folders can potentially cause trouble and also degrade the performance of the client. The best strategy to prevent this is to set up an automatic clean up. In Thunderbird, go to Edit -> Preferences -> Advanced -> Network & Disk Space and then enable the “Compact folders when it will save over…KB” option.
Categories: centos |
No Comments
December 20th, 2008
One of the many things I use Linux for at work is to edit movies. No, not that kind of movies you are thinking about … this is strictly for science work. Anyway, the first step is to transfer video from a camcorder connected through a firewire port using kino. I am doing this on a machine running Fedora Core 6. I attempted to do the work on a CentOS-5 box. However, going to the capture mode on kino produces an error:
*raw1394 kernel module not loaded or failure to read/write .....
Long story short, this is a known issue, and the kino web site offers a couple of workarounds. I am going to summarize a CentOS version of these workarounds.
Method 1 — requires a kernel from the centosplus repository
(1) Edit /etc/modprobe.d/blacklist-firewire and comment out blacklist firewire-ohci
(2) Reboot
(or run modprobe firewire-ohci to achieve the same without reboot)
(3) Connect the camcorder
(4) Change the ownership of /dev/fw? (? = 0, 1, etc) to the logged in user
Method 2 — Use if Method 1 does not work
(1) Install ieee1394 from the ATrpms repository. Refer to the CentOS Wiki Repository article for the instruction.
(2) Download the kmod-ieee1394 package from Alan’s driver directory and install it using the rpm -Uvh command. There is no need to reinstall this kernel module upon kernel update.
(3) Download packages libraw1394_8 and libraw1394 from my collection and install them as in (2).
(4) Edit /etc/modprobe.d/blacklist and add these lines:
blacklist firewire_core
blacklist firewire_ohci
(5) Reboot
(6) Chenge the ownership of /dev/raw1394 to the logged in user.
Categories: centos |
2 Comments
