Miscellaneous problems
lappi - Upgraded Linux Kernel from 2.6.13 r3 to 2.6.17 r8
"no interface module has been loaded" when running /etc/init.d/net.eth0 start on lappi
/etc/init.d/net.eth0: line 34: /lib/rcscripts/net.modules.d/helpers.d/functions
No such file or directory
* Starting
* no interface module has been loaded
found answer at http://www.m4tix.net/wordpress/?cat=6 which is to run etc-update, but this did not fix all problems, udev has some issues with a deprecated syntax in listing processes, and /etc/conf.d/net has changed syntax as well.
Installing gnome-system-monitor failed first. Fix was to:
# FEATURES="-sandbox" emerge sandbox
Now problem with gnome-vfs:
* Failed Running automake !
*
* Include in your bugreport the contents of:
*
* /var/tmp/portage/gnome-vfs-2.14.2-r1/temp/automake-29193.out
!!! ERROR: gnome-base/gnome-vfs-2.14.2-r1 failed.
Call stack:
ebuild.sh, line 1546: Called dyn_unpack
ebuild.sh, line 708: Called src_unpack
gnome-vfs-2.14.2-r1.ebuild, line 75: Called eautomake
autotools.eclass, line 176: Called eautoreconf
autotools.eclass, line 83: Called eautomake
autotools.eclass, line 185: Called autotools_run_tool 'automake' '--add-missing'
'--copy'
autotools.eclass, line 212: Called die
!!! Failed Running automake !
!!! If you need support, post the topmost build error, and the call stack if
relevant.
lappi ~ #
It appears that this is solved by upgrading to automake 1.9.6-r2.
Veritas NetBackup bpcd Problem
bpcd problem on 2.6.17 kernel. bpcd will work with some cajolling on 2.6.16 and prior, but apparently cannot be made to work on 2.6.17 and later kernels
udevd [973] nss_ldap reconnecting to LDAP server (sleeping 4 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 8 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 16 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 32 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 64 seconds) udevd [973] nss_ldap could not search LDAP server - server is unavailable udevd [973] nss_ldap reconnecting to LDAP server (sleeping 4 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 8 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 16 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 32 seconds) udevd [973] nss_ldap reconnecting to LDAP server (sleeping 64 seconds)
This is caused by a bug in post-239 versions of nss_ldap according to http://www.nabble.com/nss_ldap-and-udevd-t1200588.html. To downgrade to the working version, use:
emerge =sys-auth/nss_ldap-239-r1
NFS problem - doolin revealed a problem whereby copying a file to a NFS mounted drive would result in an error "cannot create regular file: Invalid argument" even though the file is created, but with zero length. Copying over the file will work. This appears to be related, perhaps caused, by the nfs_acl module (responding to the CONFIG_NFS_V3_ACL variable?).
Rebuilding NFS support in the Linux menuconfig, taking pains to uncheck the NFS v3 and v4 ACL support, then making sure that nfs_acl does not get loaded causes this problem to go away.
Problem While Installing Oracle JInitiator
Error Dialog
Oracle Jinitiator
An installation support file could not
be installed.
No mapping between account names and security IDs was done.
trying to install Banner. This problem simply went away, possibly following a reboot, overnight.
Problem emerging gentoo-sources on siraz
Ran mirrorselect -s4 first, but noone appears to have genpatches:
Resolving ftp.gtlib.cc.gatech.edu... 128.61.111.10, 128.61.111.11, 128.61.111.9
Connecting to ftp.gtlib.cc.gatech.edu|128.61.111.10|:21... connected.
Logging in as anonymous ... Logged in!
==> SYST ... done. ==> PWD ... done.
==> TYPE I ... done. ==> CWD /pub/gentoo/distfiles ... done.
==> PASV ... done. ==> RETR genpatches-2.6.15-8.extras.tar.bz2 ...
No such file `genpatches-2.6.15-8.extras.tar.bz2'.
>>> Downloading http://mirrors.tds.net/gentoo/distfiles/genpatches-2.6.15-8.extras.tar.bz2
--08:19:54-- http://mirrors.tds.net/gentoo/distfiles/genpatches-2.6.15-8.extras.tar.bz2
=> `/usr/portage/distfiles/genpatches-2.6.15-8.extras.tar.bz2'
Resolving mirrors.tds.net... 216.165.129.134
Connecting to mirrors.tds.net|216.165.129.134|:80... connected.
HTTP request sent, awaiting response... 404 Not Found
08:19:54 ERROR 404: Not Found.
>>> Downloading ftp://ftp.tu-clausthal.de/pub/linux/gentoo/distfiles/genpatches-2.6.15-8.extras.tar.bz2
--08:19:54-- ftp://ftp.tu-clausthal.de/pub/linux/gentoo/distfiles/genpatches-2.6.15-8.extras.tar.bz2
=> `/usr/portage/distfiles/genpatches-2.6.15-8.extras.tar.bz2'
Resolving ftp.tu-clausthal.de... 139.174.2.36
Connecting to ftp.tu-clausthal.de|139.174.2.36|:21... connected.
Logging in as anonymous ... Logged in!
==> SYST ... done. ==> PWD ... done.
==> TYPE I ... done. ==> CWD /pub/linux/gentoo/distfiles ... done.
==> PASV ... done. ==> RETR genpatches-2.6.15-8.extras.tar.bz2 ...
No such file `genpatches-2.6.15-8.extras.tar.bz2'.
!!! Couldn't download genpatches-2.6.15-8.extras.tar.bz2. Aborting.
livecd / #
emerge --sync got this fixed and should be done after untarring the stage and snapshot tarballs.
10/25/06 - There is an apparent limit to what Apache's directory listing will show. I uploaded two files to /downloads/yalchin on lcni, 14 GB and 20 GB. Neither shows in the browser directory listing. Other, comparatively tiny, files will show, and the index.html will display correctly also. I will try to split the files to get them to a size that will display in the listing.
chuck@lcni /var/www/localhost/htdocs/downloads/yalchin $ dd if=/dev/zero of=big.file
bs=1M count=2000
2000+0 records in
2000+0 records out
visible:
-rw-r--r-- 1 chuck staff 2097152000 Oct 25 09:34 big.file
not visible:
-rw-r--r-- 1 chuck staff 4194304000 Oct 25 09:37 big.file
-rw-r--r-- 1 chuck staff 3145728000 Oct 25 09:41 big.file
The limit appears to be 2 GB, which makes sense when considering a 31 bit factor.
Files can be divided into manageable pieces using the split command:
split -d -b 2000m some.file some.file.part.
Bizarre Linux behavior. Lappi will not give a command-prompt to an LDAP user, despite being authenticated:
SSH Secure Shell 3.2.9 (Build 283)
Copyright (c) 2000-2003 SSH Communications Security Corp - http://www.ssh.com/
This copy of SSH Secure Shell is licensed for educational, charity,
or personal recreational or hobby use.
Any commercial use requires a separate license.
Last login: Tue Oct 24 15:05:57 2006
==> Welcome to FSL ...
==> Welcome to FreeSurfer ...
Mricro is not installed
AFNI is not installed
mri3dX is not installed
CARet is not installed
************** ATTENTION X11 APPLICATION USERS ****************
Be sure your X server is running and you have run 'xhost +lappi'.
***************************************************************
I have not found reference to this problem on the Net yet.
1 November 2006
I am now getting an error on bleu when trying to run CARet:
bleu ~ # /usr/local/caret/bin/caret5
/usr/local/caret/bin/caret5: error while loading shared libraries: libGL.so.1:
cannot handle TLS data
bleu ~ # ldd /usr/local/caret/bin/caret5
/usr/local/caret/bin/caret5: error while loading shared libraries: libGL.so.1:
cannot handle TLS data
bleu ~ #
Note that this occurs even when trying ldd on the file. startx would produce the same error. Updated nvidia-drivers from 1.0.8762-r1 to 1.0.8776, which got X11 working, but CARet 5 still fails.
3 November 2006
Internet Explorer 7 can be started with a -noext option which disables "add-ons". Attempts to use add-ons, like the windowsupdate.microsoft.com site, will result in a message just above the web page display portion of the browser window saying that IE is running with add-ons disabled. Removing the -noext option to the command-line will fx this problem.
8 November 2006
Silent failures of FSL are no surprise. Trying to find out what went wrong can be a matter of tracking down the command that failed, running it on the command line, and seeing what happened. Often, the command that threw up the error is not the one that caused it. The FSL processing sequence relies on the output of previous FSL commands which can fail without any visible notification. This will often leave an empty or corrupt data file to feed the next command in line. When the next command tries to ingest, it fails and produces the visible error. In one instance of running the feeds test suite, the cause of the error turned out to be a missing dc command. This is often installed on Gentoo as a dependency of some other package, but often is not. To explicitly install it, the bc package must be installed. This will install both bc and dc, but not ad or ac. :-)
16 November 2006
Using smbclient to connect to a Samba server from a Linux client fails:
-bash-3.1$ smbclient -U shfrey //freypan/
Password:
Domain=[FREYPAN] OS=[Unix] Server=[Samba 3.0.10]
tree connect failed: NT_STATUS_BAD_NETWORK_NAME
-bash-3.1$ smbclient -U shfrey //freypan/Storage/
Password:
Domain=[FREYPAN] OS=[Unix] Server=[Samba 3.0.10]
tree connect failed: NT_STATUS_BAD_NETWORK_NAME
-bash-3.1$ smbclient -U shfrey \\\\freypan\\Storage\\
Password:
Domain=[FREYPAN] OS=[Unix] Server=[Samba 3.0.10]
tree connect failed: NT_STATUS_BAD_NETWORK_NAME
-bash-3.1$ smbclient -U chuck //lauterbur/
Password:
Domain=[LCNI] OS=[Unix] Server=[Samba 3.0.21b]
tree connect failed: NT_STATUS_BAD_NETWORK_NAME
The key is to omit the trailing slash:
-bash-3.1$ smbclient -U shfrey //freypan/Storage
Password:
Domain=[FREYPAN] OS=[Unix] Server=[Samba 3.0.10]
smb: \>
November 27, 2006
Using a Belkin 1100VA UPS with Linux via USB. The software (Bulldog) suggests /dev/usb as the device to use to communicate with the UPS. This is incorrect on Gentoo systems. Finding the correct /dev thing apparently is not straightforward. The Guide (http://www.gentoo.org/doc/en/usb-guide.xml) does not mention how to discover which device to use.
A useful site is http://www.mscs.dal.ca/~selinger/ups/belkinunv.html
February 12, 2007
Here's a good one:
lauterbur{2}# find / -name truss.out
/root/110615-16/truss.out
find: cannot open /: Value too large for defined data type
lauterbur{3}#
WTH does this mean? Feels like a decaying system to me. This is from Solaris 8 on a V880. This machine has a failure in sendmail as well, ever since a reboot following an upgrade to account for the upcoming DST change. Patches applied: 108996-65, 109809-05, and 110615-16. Of course, I had to reinstall nss_ldap and pam_ldap from source to fix the LDAP authentication, but this should not have taken down sendmail - or should it have?
Sendmail fails with a segfault:
lauterbur{43}# /usr/lib/sendmail
Segmentation fault (core dumped)
lauterbur{44}#
I fixed this by installing sendmail (version 8.14.0) from source.
February 13, 2007
Got the following message when trying mirrorselect -s5 from the lcni web server:
"Netselect failed to return any mirrors. Try again using block mode"
There is exactly one link Google provides, and this is to a chat transcript that does not provide any analysis into our problem.
February 14, 2007
Adaptec 29160 limitation - Apparently, this card is unable to recognize drives of greater than 640 GB. We have a Maxtronic Arena 99EX2 RAID array (BIOS version 3.2Q) with six 250 GB Seagate drives installed and configured as RAID 5. The BIOS of the Adaptec 29160 SCSI adapter (BIOS version 3.10.0) sees only 640 GB. The Arena reports a 1.2 TB capacity. I still have not found a way to get around this problem, or even any mention of such a limtation on the 'net.
February 15, 2007
Last week, the web server started failing when trying to run CGI scripts. Apache2 error log recorded stuff like:
lcni apache2 # tail error_log
[Thu Feb 15 10:31:55 2007] [error] [client 128.223.78.215] File does not exist:
/var/www/localhost/htdocs/Library/images, referer: http://lcni.uoregon.edu/
[Thu Feb 15 10:31:57 2007] [error] [client 128.223.78.215] File does not exist:
/var/www/localhost/htdocs/cal/images, referer: http://lcni.uoregon.edu/cal/index.html
[Thu Feb 15 10:31:57 2007] [error] [client 128.223.78.215] File does not exist:
/var/www/localhost/htdocs/Library/images, referer: http://lcni.uoregon.edu/cal/index.html
[Thu Feb 15 10:31:57 2007] [error] [client 128.223.78.215] File does not exist:
/var/www/localhost/htdocs/cal/images, referer: http://lcni.uoregon.edu/cal/index.html
*** glibc detected *** corrupted double-linked list: 0x08394960 ***
*** glibc detected *** corrupted double-linked list: 0x08392990 ***
[Thu Feb 15 10:32:02 2007] [notice] child pid 23298 exit signal Aborted (6)
[Thu Feb 15 10:32:02 2007] [notice] child pid 23299 exit signal Aborted (6)
*** glibc detected *** double free or corruption (!prev): 0x08802570 ***
[Thu Feb 15 10:32:20 2007] [notice] child pid 23303 exit signal Aborted (6)
lcni apache2 # emerge -s libc
emerge glibc reports:
checking version of sed... 4.1.4, ok
checking for autoconf... autoconf
checking whether autoconf works... yes
configure: error:
*** These critical programs are missing or too old: gcc
*** Check the INSTALL file for required versions.
!!! ERROR: sys-libs/glibc-2.5 failed.
Call stack:
ebuild.sh, line 1614: Called dyn_compile
ebuild.sh, line 971: Called qa_call 'src_compile'
environment, line 4061: Called src_compile
glibc-2.5.ebuild, line 1121: Called toolchain-glibc_src_compile
glibc-2.5.ebuild, line 240: Called glibc_do_configure 'nptl'
glibc-2.5.ebuild, line 893: Called die
!!! failed to configure glibc
!!! If you need support, post the topmost build error, and the call stack if
relevant.
!!! A complete build log is located at '/var/tmp/portage/sys-libs/glibc-2.5/temp/build.log'.
lcni apache2 #
lcni apache2 #
This may indicate a gcc that does not support nptl. The "outdated" URL http://gentoo-wiki.com/NPTL gives a clue. Execute /lib/libc.so.6 (did not know you could execute shared libraries, now I will have to try it with every one).
lcni apache2 # /lib/libc.so.6 GNU C Library 20041102 release version 2.3.4, by Roland McGrath et al. Copyright (C) 2004 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Compiled by GNU CC version 3.3.5-20050130 (Gentoo 3.3.5.20050130-r1, ssp-3.3.5.20050130-1, pie-8.7.7.1). Compiled on a Linux 2.6.11 system on 2005-08-06. Available extensions: GNU libio by Per Bothner crypt add-on version 2.1 by Michael Glad and others linuxthreads-0.10 by Xavier Leroy The C stubs add-on version 2.1.2. GNU Libidn by Simon Josefsson BIND-8.2.3-T5B libthread_db work sponsored by Alpha Processor Inc NIS(YP)/NIS+ NSS modules 0.19 by Thorsten Kukuk Thread-local storage support included. For bug reporting instructions, please see: <http://www.gnu.org/software/libc/bugs.html>. lcni apache2 #
Now, following the upgrade, I am getting message:
/etc/host.conf: line 24: bad command `mdns off'
when logging in. the file was modified during the update:
(lcni:~) chuck% ls -l /etc/host.conf
-rw-r--r-- 1 root root 1302 Feb 15 15:03 /etc/host.conf
(lcni:~) chuck%
and apparently the keyword mdns is no longer supported, as it is not mentioned in the man page.
#
# CGI Modules
#
# These modules provide the ability to execute CGI Scripts.
#
##LoadModule cgi_module modules/mod_cgi.so
LoadModule cgid_module modules/mod_cgid.so
Note that /etc/apache2/apache2-builtin-mods specifies that mod_cgid.so is shared. Note also that the httpd.conf file loads mod_cgid.so rather than mod_cgi.so. and check this out:
lcni apache2 # ldd /usr/lib/apache2/modules/mod_cgid.so
linux-gate.so.1 => (0xffffe000)
libpthread.so.0 => /lib/libpthread.so.0 (0xb7fc3000)
libc.so.6 => /lib/libc.so.6 (0xb7e9b000)
/lib/ld-linux.so.2 (0x80000000)
lcni apache2 # ldd /usr/lib/apache2/modules/mod_cgi.so
linux-gate.so.1 => (0xffffe000)
libpthread.so.0 => /lib/libpthread.so.0 (0xb7edf000)
libc.so.6 => /lib/libc.so.6 (0xb7db7000)
/lib/ld-linux.so.2 (0x80000000)
lcni apache2 #
also, httpd.conf, mod_cgi.so, and mod_cgid.so were all touched on Feb 9, about when it all went to hell.
Found that php is not installed.
http://bugs.php.net/bug.php?id=36884
Trying to use mod_cgi.so rather than mod_cgid.so resulted in:
lcni apache2 # /etc/init.d/apache2 stop
* Apache2 has detected a syntax error in your configuration files:
Syntax error on line 276 of /etc/apache2/httpd.conf:
Cannot load /usr/lib/apache2/modules/mod_cgi.so into server: /usr/lib/apache2/modules/mod_cgi.so:
undefined symbol: apr_procattr_addrspace_set
lcni apache2 #
March 6, 2007 - I ended up flattening the machine and rebuilding. It's better now.
March 6, 2007 - A mysterious failure in Cygwin/X was corrected. The problem is not documented elsewhere, as far as I can tell, but the solution is here. The problem is when launching X from a cygwin shell, numerous "refused by server" messages are received. The problem was in the setting for HOME, which had been set to a Windows drive mounted to a network directory and was reading as HOME=/cygdrive/z
..
AUDIT: Mon Mar 5 08:28:50 2007: 3424 X: client 1 rejected from local host
Xlib: connection to ":0.0" refused by server
Xlib: No protocol specified
..
giving up.
xinit: unable to connect to X server
winDinitMultiWindowWM - Noting shutdown in progress
waiting for X server to shut down
xinit: Server error.
This may be the result of an incorrect HOMEDRIVE setting. Actually an
incorrect HOME setting. From within the cygwin shell:
export HOME="/cygdrive/c/Documents and Settings/chuck"
startx
March 13, 2007
Have pinned down the Dell Precision 340's problems with a Windows XP upgrade: the problem is with the video driver and the ATI Radeon 700 card. This is a dual-head card. The symptoms include a contiunal reset of the video card, causing the monitor to go cycle between almost-but-not-quite showing something intelligible to a powersave mode (move mouse, press any key, blah, blah). The updated drivers do not fit on a floppy <doh>. The most helpful site, aside from this one, is http://www.itc.virginia.edu/desktop/dci/pre360.html
The plan is to let the Windows install finish blind, then reboot into safe mode. This should allow seeing the screen to enable installation of the latest driver from AMD-ATI.
November 9, 2007
Have a problem trying to build fslview on Linux. The CMake appears to not be able to resolve all the libraries. I have opted to try to install from a binary distribution. This has its own set of problems:
malbec local # ldd fsl-4.0.0/bin/fslview_bin
linux-gate.so.1 => (0xffffe000)
libqt-mt.so.3 => /usr/qt/3/lib/libqt-mt.so.3 (0xb78f3000)
libSM.so.6 => /usr/lib/libSM.so.6 (0xb78ea000)
libICE.so.6 => /usr/lib/libICE.so.6 (0xb78d2000)
libX11.so.6 => /usr/lib/libX11.so.6 (0xb77e6000)
libXext.so.6 => /usr/lib/libXext.so.6 (0xb77d7000)
libdl.so.2 => /lib/libdl.so.2 (0xb77d3000)
libpthread.so.0 => /lib/libpthread.so.0 (0xb77bc000)
libqwt.so.4 => /usr/lib/libqwt.so.4 (0xb770f000)
libQVTK.so.5.0 => /usr/lib/libQVTK.so.5.0 (0xb76fb000)
libvtkHybrid.so.5.0 => /usr/lib/libvtkHybrid.so.5.0 (0xb7567000)
libvtkWidgets.so.5.0 => /usr/lib/libvtkWidgets.so.5.0 (0xb74f8000)
libvtkRendering.so.5.0 => /usr/lib/libvtkRendering.so.5.0 (0xb72b6000)
libvtkGraphics.so.5.0 => /usr/lib/libvtkGraphics.so.5.0 (0xb6d3d000)
libvtkImaging.so.5.0 => /usr/lib/libvtkImaging.so.5.0 (0xb6a11000)
libvtkftgl.so.5.0 => /usr/lib/libvtkftgl.so.5.0 (0xb6a06000)
libvtkfreetype.so.5.0 => /usr/lib/libvtkfreetype.so.5.0 (0xb69a1000)
libGL.so.1 => /usr/lib/opengl/nvidia/lib/libGL.so.1 (0xb690b000)
libXt.so.6 => /usr/lib/libXt.so.6 (0xb68bc000)
libvtkIO.so.5.0 => /usr/lib/libvtkIO.so.5.0 (0xb65dd000)
libvtkFiltering.so.5.0 => /usr/lib/libvtkFiltering.so.5.0 (0xb63a8000)
libvtkCommon.so.5.0 => /usr/lib/libvtkCommon.so.5.0 (0xb6224000)
libvtksys.so.5.0 => /usr/lib/libvtksys.so.5.0 (0xb61fc000)
libvtkDICOMParser.so.5.0 => /usr/lib/libvtkDICOMParser.so.5.0 (0xb61de000)
libvtkpng.so.5.0 => not found
libvtktiff.so.5.0 => not found
libvtkzlib.so.5.0 => not found
libvtkjpeg.so.5.0 => not found
libvtkexpat.so.5.0 => not found
libvtkexoIIc.so.5.0 => /usr/lib/libvtkexoIIc.so.5.0 (0xb61b1000)
libvtkNetCDF.so.5.0 => /usr/lib/libvtkNetCDF.so.5.0 (0xb6183000)
libstdc++.so.6 => /usr/lib/gcc/i686-pc-linux-gnu/4.1.2/libstdc++.so.6 (0xb609f000)
libm.so.6 => /lib/libm.so.6 (0xb607a000)
libgcc_s.so.1 => /usr/lib/gcc/i686-pc-linux-gnu/4.1.2/libgcc_s.so.1 (0xb606e000)
libc.so.6 => /lib/libc.so.6 (0xb5f46000)
libmng.so.1 => /usr/lib/libmng.so.1 (0xb5ee8000)
libjpeg.so.62 => /usr/lib/libjpeg.so.62 (0xb5ec9000)
libpng12.so.0 => /usr/lib/libpng12.so.0 (0xb5ea6000)
libz.so.1 => /lib/libz.so.1 (0xb5e94000)
libXi.so.6 => /usr/lib/libXi.so.6 (0xb5e8b000)
libXrender.so.1 => /usr/lib/libXrender.so.1 (0xb5e83000)
libXrandr.so.2 => /usr/lib/libXrandr.so.2 (0xb5e7f000)
libXcursor.so.1 => /usr/lib/libXcursor.so.1 (0xb5e76000)
libXft.so.2 => /usr/lib/libXft.so.2 (0xb5e64000)
libfreetype.so.6 => /usr/lib/libfreetype.so.6 (0xb5df8000)
libfontconfig.so.1 => /usr/lib/libfontconfig.so.1 (0xb5dcc000)
libXau.so.6 => /usr/lib/libXau.so.6 (0xb5dc9000)
libXdmcp.so.6 => /usr/lib/libXdmcp.so.6 (0xb5dc4000)
/lib/ld-linux.so.2 (0xb7fe5000)
libQtGui.so.4 => /usr/lib/qt4/libQtGui.so.4 (0xb561d000)
libQtCore.so.4 => /usr/lib/qt4/libQtCore.so.4 (0xb5484000)
libtiff.so.3 => /usr/lib/libtiff.so.3 (0xb542f000)
libexpat.so.0 => /usr/lib/libexpat.so.0 (0xb5410000)
libGLcore.so.1 => /usr/lib/opengl/nvidia/lib/libGLcore.so.1 (0xb4a78000)
libnvidia-tls.so.1 => /usr/lib/opengl/nvidia/lib/libnvidia-tls.so.1 (0xb4a76000)
libXfixes.so.3 => /usr/lib/libXfixes.so.3 (0xb4a70000)
libxml2.so.2 => /usr/lib/libxml2.so.2 (0xb4957000)
librt.so.1 => /lib/librt.so.1 (0xb494e000)
malbec local #
I tried rebuilding VTK with USE="tiff png zlib jpeg expat" emerge vtk", but the missing libraries were still not created. As is usual for items on this page, I found no useful advice on the 'net.
I ended up installing the binary for CentOS5.
October 2007
Following a series of updates to the Linux boxes, I found very slow response upon logging into zamorano. The problem was manifested in log entries about lockd errors. The problem originated in an update to nfs-utils and is solved by starting rpc.statd on each NFS client machine.
November 2007
Annoying Firefox behavior, mailto links open gmail in a browser window rather than Thunderbird. To fix, go to about:config and change network.protocol-handler.external.mailto to true. This will allow the default email handler, set in Windows Set Program Access and Defaults, to be used for mailto: links.