Re: Suggestion for a hard drive CUPS-like data base
One point of clarification. CUPS was NOT developed by Apple, they bought it. I used to talk to the guy who actually created it many YEARS ago as free software. He then commercialized it some, then eventually sold it off. I guess Apple owns the rights to it now.
Re: xdiff for SL
xdiff can be found at the following link: http://reality.sgiweb.org/rudy/xdiff/ I had some problems getting it to work under the latest SL release (5.5) and the author, Rudy Wortel, sent me the source file. He didn't say I could give it to anyone else, so I can't offer it. I may be able to give out an executable, but I'd have to ask about that. Let him or me know if you have any problems with what is on his Web site above. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: Clean SL 5.4 Install and Anaconda Disk Partitioning Madness
Thanks for all the suggestions. Anaconda still did some rearranging, but I eventually figured out what I needed to do to get what I wanted. I had to use `Edit' to set the starting and ending cylinders as well as use the Primary Partition check box. I had to use `Edit' for all the partitions. I tried `Add' for the last partition and let it go to the end of the drive automatically, but it stopped short of the end, so I had to use `Edit' in order to use the full drive space. I assembled all my partitions into their respective software RAID's, set file system type, mount points, and clicked `Next'. New problem, it now says XFS is not valid for a boot partition. Hogwash! All my systems use XFS for boot partitions. I've got systems running SL 3.0.5, SL 4.5, and one running SL 5.3 all with ONLY XFS files systems including /boot. Any ideas on getting past this new hurdle? Thanks. Oh, in order to get XFS listed as a valid file system, I had to add `xfs' to the `linux' line when I started up the install. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Clean SL 5.4 Install and Anaconda Disk Partitioning Madness
I've been searching Google for answers and can't find any, so I decided to check here. I'm trying to do a clean install of SL 5.4. I'm booting from an SL 5.4 x86_64 DVD. When I get to the point of custom partitioning my drives, Anaconda makes a mess of things. I have 4 drives and I want 4 partions on each drive. The first partion will be `/boot', next `/', then `/data', and finally a swap partion. As I create each partition on each drive, Anaconda will suddenly rearrange the order of the partions. When I go onto another drive, the order may be different than the last drive I just partitioned. When I try to do the 4th and last partition, I get an `Extended' 4th partion (which is empty) and a real 5th partition, instead of a simple single partition. In the end, what I want is the first partition on each drive combined into a software RAID 1 and be `/boot'. The next 2 sets will be software RAID 0's and `/' `/data'. The final partitions will be 4 separate swap partitions that the OS will take care of. I've tried creating all the partitions on one drive and then moving onto the next one and the next one, but it scrambles things up. I've tried creating the first partition on each drive, then combining them into the RAID 1 md0 device, and specifying the files system type, and mount point `/boot'. Next I go onto the next partition, which I've tried as a software RAID 0 partition and swap at various times. This one usually works, but not always. When I get to the 3rd partition, then it will suddenly rearrange the partitions on that drive. I've even seen it suddenly create a swap partition on a different drive than what I'm actually working on. I've done this with earlier versions of SL, but I don't remember having this much trouble with Anaconda randomly rearranging things and creating an extra unneeded partion. Any insights would be greatly appreciated. Thanks. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: SL archives and indexed postings not accessible
I tried to do what Mark Stodola indicated he did and didn't get very far. I went to: http://listserv.fnal.gov/ I clicked on `View list archives', which took me to: http://listserv.fnal.gov/archives/ which gives me a blank page. I looked at the page source and the the page is incomplete, there is no ending `/BODY' or `/HTML'. I have no idea what else is missing, but it is enough for my browser to just give up on the whole page and just give me a blank screen. I tried a MUCH older browser, Mosaic, I received the same Web page, but it displayed everything that it downloaded despite it being incomplete. In the old browser, I clicked on the SCIENTIFIC-LINUX-USERS link and I get a page displayed. I go back to my usual browser and put that link into it and get a blank page again. This page has the ending `/BODY', but no ending `/HTML'. Again in the old browser, I tried clicking on `March 2010' and then `February 2010'. Instead of showing me a page, both times it wants to download a file. I went ahead with the download and the page has a bunch of lines with control-M's at the beginning of the file and after every line in the file. It is an HTML file, but again with out an ending `/BODY' or `/HTML'. Apparently, these links run an executable on a Windbloze system as the link ends with: `/scripts/wa.exe?A1=ind1002L=scientific-linux-users' -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: xcalc still available?
Looks like another useful tool TUV has decided not to include. Just like xman. The last time I checked, which has been a very long time ago, the sources had them, but TUV just decides not to allow them to be built. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: Logo Contest for SL6 extended
Personally, of all the logos on the page, I think #3 is the best. I say #12 is the worst, simply because of the complexity and it doesn't seem to scale well. However, #1 is a very close second, to the bottom. Although I did not say it in my last email comments, #1, like William Somsky indicated, also looks very familiar to me. I think I've seen it as the logo of some chemical or pharmaceutical company. I do not remember which one at the moment. That's my 25 cents, inflation. :-) -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: TESTING - yum update for SL5
Oh! I thought yumex was just a front end to yum and that updating yum would also update yumex. Sorry for the confusion, I didn't realize they were not fully connected. I'll try using just yum to see how that works.
Re: TESTING - yum update for SL5
Still no joy. I uninstalled: kernel-2.6.18-128.1.10.el5.x86_64 kernel-module-xfs-2.6.18-128.1.10.el5-0.4-2.sl5.x86_64 rebooted into old kernel. Did following command: # yum install kernel-2.6.18-128.1.10.el5.x86_64 kernel-devel-2.6.18-128.1.10.el5.x86_64 kernel-module-xfs-2.6.18-128.1.10.el5-0.4-2.sl5.x86_64 After the above finished, I double checked the size of /boot/initrd-2.6.18-128.1.10.el5.img and knew it was too small for things to have worked, but I rebooted anyways. It failed to fully boot as it couldn't mount the XFS file systems. Here is a list of all `yum' RPM's installed: yum-3.2.19-22.sl.noarch yum-aliases-1.1.16-14.el5.noarch yum-autoupdate-1-1.SL.noarch yum-changelog-1.1.16-14.el5.noarch yum-conf-53-1.SL.noarch yum-cron-0.6-3.el5.noarch yum-downloadonly-1.1.16-14.el5.noarch yumex-2.0.3-1.0.el5.noarch yum-filter-data-1.1.16-14.el5.noarch yum-keys-1.1.16-14.el5.noarch yum-kmod-1.1.16-14.el5.noarch yum-list-data-1.1.16-14.el5.noarch yum-merge-conf-1.1.16-14.el5.noarch yum-metadata-parser-1.1.2-2.el5.x86_64 yum-priorities-1.1.16-14.el5.noarch yum-protectbase-1.1.16-14.el5.noarch yum-protect-packages-1.1.16-14.el5.noarch yum-refresh-updatesd-1.1.16-14.el5.noarch yum-security-1.1.16-14.el5.noarch yum-tsflags-1.1.16-14.el5.noarch yum-updatesd-0.9-2.sl.noarch yum-upgrade-helper-1.1.16-14.el5.noarch yum-utils-1.1.16-14.el5.noarch yum-verify-1.1.16-14.el5.noarch If you have any suggestions or need more information, please let me know.
Re: TESTING - yum update for SL5
As I said in an earlier message, ALL file systems are XFS, /, /boot, and a data partition. Here is the output from the rpm commands: # rpm -q --scripts kernel-2.6.18-128.1.10.el5.x86_64 postinstall scriptlet (using /bin/sh): if [ `uname -i` == x86_64 -o `uname -i` == i386 ]; then if [ -f /etc/sysconfig/kernel ]; then /bin/sed -i -e 's/^DEFAULTKERNEL=kernel-smp$/DEFAULTKERNEL=kernel/' /etc/sysconfig/kernel || exit $? fi fi /sbin/new-kernel-pkg --package kernel --mkinitrd --depmod --install 2.6.18-128.1.10.el5 || exit $? if [ -x /sbin/weak-modules ] then /sbin/weak-modules --add-kernel 2.6.18-128.1.10.el5 || exit $? fi preuninstall scriptlet (using /bin/sh): /sbin/new-kernel-pkg --rminitrd --rmmoddep --remove 2.6.18-128.1.10.el5 || exit $? if [ -x /sbin/weak-modules ] then /sbin/weak-modules --remove-kernel 2.6.18-128.1.10.el5 || exit $? fi # rpm -q --scripts kernel-module-xfs-2.6.18-128.1.10.el5 postinstall scriptlet (using /bin/sh): depmod -a 2.6.18-128.1.10.el5 /dev/null 21 || : postuninstall scriptlet (using /bin/sh): depmod -a 2.6.18-128.1.10.el5 /dev/null 21 || : As I said before, do I need to add something someplace to make sure new kernels include XFS in the /boot/initrd* file?
Re: TESTING - yum update for SL5
I installed the lastest kernel: kernel-2.6.18-128.1.10.el5.x86_64 and XFS module: kernel-module-xfs-2.6.18-128.1.10.el5-0.4-2.sl5.x86_64 on a SL5.3 system. After rebooting, the kernel didn't know anything about XFS, when my previous one still did. I saw Troy's note, so I uninstalled both the kernel and XFS module. I enabled the `sl-testing' repo and updated yum. I then proceeded to install both of the above packages again. I rebooted and the new kernel still didn't know anything about XFS. Shouldn't the XFS module been automatically included in the new `initrd' file? Did I do something out of order or wrong? I want to make sure this doesn't happen again. I manually ran the `mkinitrd' command and now the system boots fine with the new kernel. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: TESTING - yum update for SL5
For the past week, I've been booting into the 2.6.18-128.1.1.el5 kernel. Now, today, after using `mkinitrd', I'm booting into the 2.6.18-128.1.10.el5 kernel. I've rebooted this system fine several times now with the new kernel.
Re: login to 5.3 livedvd
Just a data point, I've been using the 64 bit LiveDVD with out problems. livedvd64_SL53_2009-03-23.iso This is on a desktop system with an Intel mother board. I don't have any details handy at the moment. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: b.l.ba...@larc.nasa.govhttp://www.vigyan.com/~blbates/
Re: AFS on XFS or ext3?
Michael Mansour when are you going to stop the FUD?? Red Hat doesn't support it because it is developed and maintain by its competition. Dependability, reliability, robustness is NOT a factor, it is pure politics and nothing more. There have been plenty of people complaining that XFS isn't supported by RH SL as much as it should be. I do NOT want a `flame war', but if you are going to continue to lie about XFS, you leave me no alternative but to support it. If your posts had said positive things about ext3 instead of lying about XFS, I would have ignored you. However, YOU started the lying, so YOU prompted the XFS support replies. I do not mind someone supporting their favorite what ever, but when they start lying about something else just to support their position, then THAT is something entirely different. Stop the lies and I'll stop the replies. It is entierly up to you.
Re: AFS on XFS or ext3?
Michael Mansour, cut the CRAP/FUD out! I would NOT depend on ext3 if I CARED about what was stored on my disks. I ONLY use ext3 if the data stored is NOT of very high importance. I use XFS when I DO CARE, so I use it all the time. XFS is the most reliable, dependable, and robust file system out there and independent tests have consistently shown it to be much faster than ext3. It has far more YEARS and Pentabytes of service under it's belt than ext3, a LOT more! I've had XFS do a much better job of surviving system crashes and disk failures than ext3.
Problems Adding Things to SL 4.5 x86_64 Install DVD
I'm trying to add a few things to the install DVD for SL 4.5 x86_64. I
downloaded all 6 of the SL 4.5 CD ISO images. Used the notes on the SL site
to combine all the CD's into one DVD and that worked fine. I can boot,
install, rescue, etc. from this DVD. Now I'm trying to add some drivers,
modules, etc. to the DVD so I've got everything I need in one place and
don't have to have a floppy for drivers, etc..
I had some problems with the build environment. I did some Google
searching and found out all the system scripts were expecting none of the
`LC_' language environmental variables to be set or at least not set to
anything but en_US.UTF-8. I corrected that problem. The build and DVD
burn all seem to go ok. Rescue mode seems fine and all my additions seem
to be there. The problem comes when I try to install SL from scratch.
Everything seems to work fine up to and including the point where one
sets the initial firewall parameters and SELinux. At this point the install
stops and gives me a popup window with the following errors:
Traceback (most recent call last):
File /var/tmp/anaconda-10.1.1.63.SL//usr/lib/anaconda/gui.py, line 792, in
nextClicked
self.setScreen ()
File /var/tmp/anaconda-10.1.1.63.SL//usr/lib/anaconda/gui.py, line 1135, in
setScreen
new_screen = apply(self.currentWindow.getScreen, args)
File
/var/tmp/anaconda-10.1.1.63.SL//usr/lib/anaconda/iw/language_support_gui.py,
line 181, in getScreen
self.defaultLang = self.langs.getDefault()
File /var/tmp/anaconda-10.1.1.63.SL//usr/lib/anaconda/language.py, line
250, in getDefault
name = self.getLangNameByNick(lang)
File /var/tmp/anaconda-10.1.1.63.SL//usr/lib/anaconda/language.py, line
235, in getLangNameByNick
return self.getLangNameByNick(en_US.UTF-8)
...
(last 2 lines repeated 993 times and then)
RuntimeError: maximum recursion depth exceeded
Local variables in innermost frame:
nick: en_US.UTF-8
self:
Google searching says this is the language environmental variable
setting problem I thought I had fixed. I've been beating on this for a
couple of weeks now and don't know where to go from here. Anyone here have
any ideas?
I have the anacdump.txt file, which includes the /tmp/syslog,
/tmp/anaconda.log, and /tmp/lvmout logs. The anaconda.log shows the last
thing Anaconda was working on was:
moving (1) to step languagesupport
While `buildinstall' is running, the `upd-instroot' script gives some
errors/warnings during the 'Creating bogl font..' section. It executes the
following command:
find /tmp/treedir/instimage/usr/share/locale -name '*.mo' -exec
msgunfmt '{}' ';'
and I see the following:
error -1 at position 2325299 (bytes: 16 Used chars: 1202 (2325299
processed)
find: msgunfmt terminated by signal 13
The last line repeated 209 times. If I try that type of command manually on
the system files, it seems to work with out errors. Next I get:
gzip:
/tmp/treedir/instimage/usr/lib/anaconda-runtime/loader/font.bgf.gz already
exists; do you wish to overwrite (y or n)?
I type `y' and things seem to proceed fine. Later I see 2 more warnings in
`upd-instroot' complaining about `/tmp/treedir/image-template/lib' and
`/tmp/treedir/instimage/lib' not existing, but it presses on. After that,
everything seems to run fine, although there are thousands of lines in the
output most of which I have no clue what is going on.
If any one has any suggestions, they would be greatly appreciated.
Thanks for any help.
--
Brent L. Bates (UNIX Sys. Admin.)
M.S. 912 Phone:(757) 865-1400, x204
NASA Langley Research CenterFAX:(757) 865-8177
Hampton, Virginia 23681-0001
Email: [EMAIL PROTECTED] http://www.vigyan.com/~blbates/
Re: Verbose Bootup of SL 4.5 - I want it
I knew I forgot to mention something. I had noticed the `quiet' on the kernel lines and deleted it. I didn't know what the `rhgb' was for so I didn't touch it. I'll try also deleting the `rhgb' and see how that works on the next reboot. Is this documented any place? I just tried a google search on this and couldn't find anything relevant. Thanks. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: [EMAIL PROTECTED] http://www.vigyan.com/~blbates/
Adding Things to Rescue Mode
I'd like to add a few things to rescue mode. I'm tired of having to use a floppy with hardware drivers when ever I need to go into rescue mode and there are a few other utilities that I'd like to add as well. How would I go about doing this? Any pointers to FAQ's or How to's? I've done some searching, but haven't found anything recent enough to be useful. Thanks for any help. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: [EMAIL PROTECTED] http://www.vigyan.com/~blbates/
Re: XFS file system
I guess I forgot to add that I downloaded the x86_64 set of SL 4.5 CD's. At least under SL 3.0.2/5, XFS will work with 4k stacks, as long as one doesn't pile a lot of layers on top of the file system. Layers like logical volumes and NFS and what ever else one can thing of. We have been running fine with software RAID 0 and 1 (2 separate partitions), XFS, and NFS. So, with the x86_64 set of six SL 4.5 CD's combined into one DVD, if we just do: linux xfs do we have a chance? I was under the impression that the installer doesn't give XFS as a default option and that if `xfs' is used on the command line it would be available. Someone gave me some diff's for anaconda source and that seemed to be the only change. XFS isn't `officially' support by RedHat's installer, but if one added the command line option it will do it, but with a warning message. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: [EMAIL PROTECTED] http://www.vigyan.com/~blbates/
Re: XFS file system
The installer will not look in the contrib directories? Ok. Could I as part of my combining the CD's into a single DVD process, move the XFS RPM's out of the contrib area and into the main stream directories? Or perhaps, with the DVD I've already burned, do some kind of shell escape out of the install GUI and install them from there? People want XFS and we really don't care about the top level vendors prejudices. I'm willing to work with the SL people to get a more reasonable solution to this on going problem.
Re: Problems with Tape Backups - Continued
Thanks for the comments. I checked the system logs and we are not using an Adaptec SCSI controller. It is a Fusion MPT (integral part of the Intel mother board), so I guess that rules out the aic79xx driver as the problem. Doesn't it? The other comment was that our kernel, 2.4.21-37.EL.XFSsmp, is too old for SCSI tape drives and I need at least 2.4.24. I do not see any SCSI errors in any of the logs. Should I see errors in any log? I only have problems when the number of files gets over some unknown limit. We've got 3 systems here, all running the same version of SL and have the same tape drive (the 3rd one has an Adaptec controller). I haven't seen any problems on the 3rd system, but I haven't tried pushing the number of files on the system to test it. The problem with trying to upgrade to a newer kernel is that SL doesn't have an XFS capable kernel newer than what I have installed, at least not for the SL 3.0.x series. Any ideas? Thanks for the help.
Problems with Tap Backups
We've been using a Sony SDX-700V AIT-3 SCSI tape drive to do full backups of our system for about a year now with out too much problem. That is until recently. Every once in a while the backup would take forever to complete. What would usually take about 1.5 hours, started taking about 6 hours, then 11 hours. The disk space to backup has been fairly constant at about 25GB. After one of these long backups, things would return to normal. Well for almost 2 weeks now I've not had a `normal' backup. In fact, one backup stopped about 40% of the way through and said the tape was full. These are 100/260GB tapes, so they can take at least 100GB. 25GB of disk space shouldn't be able to fill a tape. I'm trying a different tape today. It is still taking forever, but after almost 9 hours, it is at least 54% done. All our file systems are XFS and I'm using xfsdump for the backup. We are running SL 3.0.5 with kernel 2.4.21-37.EL.XFSsmp. We have 4 160GB SATA drives on 2 controls and 1 partition on all 4 drives is striped together to make the root partition of about 600GB. We have an almost identically configured system that hasn't been in service quite as long, but almost, that hasn't shown any problems. Only about 17GB of disk space is used and takes about 50 minutes for the full backup to complete. I've done a lot of Google searching but haven't found anything relevant. Any one have any ideas on what could be going wrong here? Any and all suggestions welcome. Thanks in advance. -- Brent L. Bates (UNIX Sys. Admin.) M.S. 912 Phone:(757) 865-1400, x204 NASA Langley Research CenterFAX:(757) 865-8177 Hampton, Virginia 23681-0001 Email: [EMAIL PROTECTED] http://www.vigyan.com/~blbates/
