M. Tibbits's questions

I'm trying to install all but one of the R packages which are currently in the Ubuntu repositories:

However, when I go to do so sudo apt-get install 'r-cran-*' I get an error:

Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 r-cran-mass : Conflicts: r-cran-vr

Ok, so I want to ignore this conflict. How? I see the flag --ignore-missing, but this is a conflict? I don't see any options to --ignore-conflicted??

computer:~> sudo apt-get install 'r-cran-*' --ignore-conflict
E: Command line option --ignore-conflict is not understood
computer:~> sudo apt-get install 'r-cran-*' --ignore-conflicted
E: Command line option --ignore-conflicted is not understood

Direct Question: Flash under firefox 4.0, npviewer.bin, in KUbuntu 10.04 was running at 100% and I couldn't kill it. I tried:

sudo kill -9 12996
sudo kill -15 12996

I logged in as root and ran both commands. The PPID was listed as 1. (Can't kill init without rebooting). I, as root, issued a reboot command and also a shutdown now. I went through the list of running processes and manually killed every single process most became <defunct>. The system has a RAID 5 without a battery backup -- so I don't like yanking the cord out of the UPS, but this is what I ended up having to do.

Is there something else I should've tried? Some other command to run? It was using 100% of the CPU, and I need to finish my dissertation, so I just forcibly turned off the computer, but if it happens again, I'd like to be prepared. Note: I also renice'd the process to 19 and it 'took' -- i.e. it responded that the process was now at nice 19 and was at nice 0. However, it was still using 100% of the cpu.

Background: A friend was using the computer, so I haven't yet cycled through the history to see if there was anything malicious, but it crashed when he started firefox and it returned to ESPN's fantasy baseball website -- not a place I would expect flash malware to grow... Any thoughts?

Direct Question: My evga classified motherboard has two ethernet ports. I want to use them simultaneously to improve the speed of a remote connection. (Yay for programming remotely). How do I do this? At present /etc/network/interfaces:

auto lo eth0 eth1

iface lo inet loopback

iface eth0 inet static
address 192.168.1.202
netmask 255.255.255.0
gateway 192.168.1.2

ifrace eth1 inet static
address 192.168.1.201
netmask 255.255.255.0

What should I change? Is there a package I need to install to bond the two channels.

Background: I do a lot of remote computing both from home to office and office to home. But (to take my ISP out of the equation), let's frame the question as if I'm really just interested in maximizing the file transfer rate between two computers on the same (local) network. Both boxes have dual ethernet motherboards. Using a simple Linksys WRT54GS as a 5 port switch (ignore wireless and isp) -- how do I maximize the transfer rates between the computers?

Please Note: I'm not interested in load balancing -- Correct me if I'm wrong, but I want to bond the two channels together so that the box appears as if it has one connection that's twice as fast.

Direct Question: I want to enable Perl to send emails. What package(s) should I install to setup a simplistic email server: no need for incoming, only outgoing. I can forward through gmail's SMTP if that's best?

Background: I am programmer with a nightly build script written in Perl. I would like to email myself the results of my nightly builds (especially if there's an error).

I've read about the perl package Mail::Sendmail briefly, but if something else is more appropriate, please tell me!!

I tried the simple aptget install sendmail, but that doesn't seem to work. I get the following errors:

Server said: 421 4.3.0 collect: Cannot write ./dfp1PFXl7W020719 (bfcommit, uid=0, gid=120): No such file or directory message transmission error (421 4.3.0 collect: Cannot write ./dfp1PFXl7W020719 (bfcommit, uid=0, gid=120): No such file or directory ) Server said: 421 4.3.0 collect: Cannot write ./dfp1PFXl7W020719 (bfcommit, uid=0, gid=120): No such file or directory

I've googled this problem a bit and tried a few things -- adding my username to /etc/mail/trusted-users and such, but to no avail.

In other words, I would be most grateful if you could provide simple instructions for setting up an outgoing mail server. I really don't understand the specifics, but as I understand, I need to forward the mail through an existing SMTP server -- so I can use my gmail account if need be (that's where I want to send the logs anyway). Any suggestions would be most greatly appreciated.

I want to attempt some simple programs using MPI on my desktop which runs KUbuntu 10.04.2. Which MPI package should I install?

I don't quite understand how the LTS releases work. Will gcc be updated to the newer 4.5 version or will it remain 4.4 and I'll have to wait for 11 before seeing 4.5?

Question: What does this dmesg mean? Is it bad? There is a two year old post about how this could imply degraded I/O performance, could this be/is this still true??

Background: I have been experiencing degraded I/O but I attribute this to using the built in RAID on my motherboard as opposed to a dedicated card. I've been tracking down a few random kernel messages and now I've switched to my desktop -- more out of curiosity than anything. The box has an Intel Core i7 930 processor. The following messages appear with some regularity. I'm running openPBS, if that matters. But these messages appear six times within the past day and a half since the system was rebooted.

The output of uname -a is:

Linux ###### 2.6.32-26-generic #48-Ubuntu SMP Wed Nov 24 10:14:11 UTC 2010 x86_64 GNU/Linux

And the messages from dmesg are:

[  185.609169] CPU0 attaching NULL sched-domain.
[  185.609173] CPU1 attaching NULL sched-domain.
[  185.609175] CPU2 attaching NULL sched-domain.
[  185.609176] CPU3 attaching NULL sched-domain.
[  185.609178] CPU4 attaching NULL sched-domain.
[  185.609179] CPU5 attaching NULL sched-domain.
[  185.609180] CPU6 attaching NULL sched-domain.
[  185.609182] CPU7 attaching NULL sched-domain.
[  185.756764] CPU0 attaching sched-domain:
[  185.756766]  domain 0: span 0,7 level SIBLING
[  185.756768]   groups: 0 (cpu_power = 589) 7 (cpu_power = 589)
[  185.756772]   domain 1: span 0-7 level MC
[  185.756773]    groups: 0,7 (cpu_power = 1178) 1,4 (cpu_power = 1178) 2,5 (cpu_power = 1178) 3,6 (cpu_power = 1178)
[  185.756780] CPU1 attaching sched-domain:
[  185.756781]  domain 0: span 1,4 level SIBLING
[  185.756782]   groups: 1 (cpu_power = 589) 4 (cpu_power = 589)
[  185.756786]   domain 1: span 0-7 level MC
[  185.756787]    groups: 1,4 (cpu_power = 1178) 2,5 (cpu_power = 1178) 3,6 (cpu_power = 1178) 0,7 (cpu_power = 1178)
[  185.756793] CPU2 attaching sched-domain:
[  185.756794]  domain 0: span 2,5 level SIBLING
[  185.756795]   groups: 2 (cpu_power = 589) 5 (cpu_power = 589)
[  185.756799]   domain 1: span 0-7 level MC
[  185.756800]    groups: 2,5 (cpu_power = 1178) 3,6 (cpu_power = 1178) 0,7 (cpu_power = 1178) 1,4 (cpu_power = 1178)
[  185.756806] CPU3 attaching sched-domain:
[  185.756807]  domain 0: span 3,6 level SIBLING
[  185.756808]   groups: 3 (cpu_power = 589) 6 (cpu_power = 589)
[  185.756812]   domain 1: span 0-7 level MC
[  185.756813]    groups: 3,6 (cpu_power = 1178) 0,7 (cpu_power = 1178) 1,4 (cpu_power = 1178) 2,5 (cpu_power = 1178)
[  185.756819] CPU4 attaching sched-domain:
[  185.756820]  domain 0: span 1,4 level SIBLING
[  185.756821]   groups: 4 (cpu_power = 589) 1 (cpu_power = 589)
[  185.756825]   domain 1: span 0-7 level MC
[  185.756826]    groups: 1,4 (cpu_power = 1178) 2,5 (cpu_power = 1178) 3,6 (cpu_power = 1178) 0,7 (cpu_power = 1178)
[  185.756832] CPU5 attaching sched-domain:
[  185.756833]  domain 0: span 2,5 level SIBLING
[  185.756834]   groups: 5 (cpu_power = 589) 2 (cpu_power = 589)
[  185.756838]   domain 1: span 0-7 level MC
[  185.756839]    groups: 2,5 (cpu_power = 1178) 3,6 (cpu_power = 1178) 0,7 (cpu_power = 1178) 1,4 (cpu_power = 1178)
[  185.756845] CPU6 attaching sched-domain:
[  185.756846]  domain 0: span 3,6 level SIBLING
[  185.756847]   groups: 6 (cpu_power = 589) 3 (cpu_power = 589)
[  185.756850]   domain 1: span 0-7 level MC
[  185.756852]    groups: 3,6 (cpu_power = 1178) 0,7 (cpu_power = 1178) 1,4 (cpu_power = 1178) 2,5 (cpu_power = 1178)
[  185.756858] CPU7 attaching sched-domain:
[  185.756859]  domain 0: span 0,7 level SIBLING
[  185.756860]   groups: 7 (cpu_power = 589) 0 (cpu_power = 589)
[  185.756863]   domain 1: span 0-7 level MC
[  185.756865]    groups: 0,7 (cpu_power = 1178) 1,4 (cpu_power = 1178) 2,5 (cpu_power = 1178) 3,6 (cpu_power = 1178)

Question: I see [ 20.473125] composite sync not supported and several such entries when I run dmesg. What do they mean?

Background: I'm trying to debug a problem where my laptop won't suspend. Since acpi seems happy and I can suspend easily from the command line, I've turned to tracking down all boot-up errors/warnings.

So I run dmesg | grep not and, amongst other shtuff, I get:

728:[   17.267120] composite sync not supported
733:[   18.009061] composite sync not supported
740:[   18.159289] registered panic notifier
749:[   18.162500] vga16fb: not registering due to another framebuffer present
757:[   18.598251] composite sync not supported
776:[   20.473125] composite sync not supported
777:[   20.932266] composite sync not supported
778:[   28.350231] composite sync not supported
779:[   28.924913] composite sync not supported
780:[   35.480658] composite sync not supported

And the full log for the few lines right around that first appearance (line 728) is listed at the bottom of my post (I'd happily include anything else). Any ideas what could be causing this? I've read several sites:

1. Ubuntuforums #1
2. IRC Chat #1

One post talks about ??Adobe flash?? causing this error?

Some others also suggest that it might be an nvidia related problem, but I've got a Dell Latitude D630 with an integrated Intel graphics -- so nvidia isn't the problem.

[   17.207142] phy0: Selected rate control algorithm 'minstrel'
[   17.207833] Registered led device: b43-phy0::tx
[   17.207849] Registered led device: b43-phy0::rx
[   17.207865] Registered led device: b43-phy0::radio
[   17.207927] Broadcom 43xx driver loaded [ Features: PL, Firmware-ID: FW13 ]
[   17.267120] composite sync not supported
[   17.415795] EXT4-fs (sda2): mounted filesystem with ordered data mode
[   17.602131] [drm] initialized overlay support
[   17.620201] input: DualPoint Stick as /devices/platform/i8042/serio1/input/input7
[   17.641192] input: AlpsPS/2 ALPS DualPoint TouchPad as /devices/platform/i8042/serio1/input/input8
[   18.009061] composite sync not supported
[   18.106042] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x100-0x3af: clean.
[   18.108115] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x3e0-0x4ff: clean.
[   18.108941] pcmcia_socket pcmcia_socket0: cs: IO port probe 0x820-0x8ff: clean.
[   18.109676] pcmcia_socket pcmcia_socket0: cs: IO port probe 0xc00-0xcf7: clean.
[   18.110356] pcmcia_socket pcmcia_socket0: cs: IO port probe 0xa00-0xaff: clean.
[   18.159286] fb0: inteldrmfb frame buffer device
[   18.159289] registered panic notifier
[   18.160218] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A03:00/LNXVIDEO:01/input/input9
[   18.160286] ACPI: Video Device [VID1] (multi-head: yes  rom: no  post: no)
[   18.160334] ACPI Warning for \_SB_.PCI0.VID2._DOD: Return Package has no elements (empty) (20090903/nspredef-433)
[   18.160432] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A03:00/LNXVIDEO:02/input/input10
[   18.160491] ACPI: Video Device [VID2] (multi-head: yes  rom: no  post: no)
[   18.160539] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
[   18.162494] vga16fb: initializing
[   18.162497] vga16fb: mapped to 0xc00a0000
[   18.162500] vga16fb: not registering due to another framebuffer present
[   18.176091] HDA Intel 0000:00:1b.0: PCI INT A -> GSI 21 (level, low) -> IRQ 21
[   18.176123] HDA Intel 0000:00:1b.0: setting latency timer to 64
[   18.285752] input: HDA Digital PCBeep as /devices/pci0000:00/0000:00:1b.0/input/input11
[   18.312497] input: HDA Intel Mic at Ext Left Jack as /devices/pci0000:00/0000:00:1b.0/sound/card0/input12
[   18.312586] input: HDA Intel HP Out at Ext Left Jack as /devices/pci0000:00/0000:00:1b.0/sound/card0/input13
[   18.328043] usbcore: registered new interface driver ndiswrapper
[   18.460909] Console: switching to colour frame buffer device 180x56
[   18.598251] composite sync not supported

Question: What package contains the subprogram responsible for the KMenu option "sleep"?

Background: I've been running KUbuntu 10.04.1 and I am completely updated. Recently (within the past month), the "sleep" menu item has stopped working. It just sits there waiting like I clicked nothing. I've check all of the logs in /var/log and nothing is added when I click sleep.

I'm guessing that something I updated has bollux'ed things up, but I don't know which package contains the component that I need to roll back. In the meantime, I've installed uswsusp, but s2ram & s2both don't ask my password when the laptop resumes -- which really bothers me. So now that I've got a little time to track this down, I had to post -- any ideas??

Update: I've messed around with the versions of acpi for quite some time. It appears to be working. If I run sudo /etc/acpi/sleep.sh it works and the computer suspends. As above, it does not lock the screen first, but at least the damn thing suspends. Oddly, sudo /etc/acpi/sleepbtn.sh does not work.

I'm running KUbuntu 10.04.1 - I click on "System Settings" from the KMenu, click on the advanced tab, and start up "Service Manager". In the lower pane, I see "Free Space Notifier" is running.

Update: It might be a monitor for the free disk space in your home partition. Can anyone confirm this?

Is there a way to determine the speed of the PCI Express connection to a specific card? I have three cards plugged in:

two Nvidia GTX 480's (one at x16 & and one at x8) one Nvidia GTX 460 running at x8

Is there some way, either by a function call in C or an option to lspci that I can determine the bus speed of the graphics cards? When I only use one of the cards for my CUDA program, I'd like to use the one which is running at x16.

Thanks!

Note: lspci -vvv dumps out

For the two GTX 480s. I don't see any differences that pertain to bus speed.

03:00.0 VGA compatible controller: nVidia Corporation Device 06c0 (rev a3)
        Subsystem: eVga.com. Corp. Device 1480
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 16
        Region 0: Memory at d4000000 (32-bit, non-prefetchable) [size=32M]
        Region 1: Memory at b0000000 (64-bit, prefetchable) [size=128M]
        Region 3: Memory at bc000000 (64-bit, prefetchable) [size=64M]
        Region 5: I/O ports at df00 [disabled] [size=128]
        [virtual] Expansion ROM at b8000000 [disabled] [size=512K]
        Capabilities: <access denied>
        Kernel driver in use: nvidia
        Kernel modules: nvidia, nvidiafb, nouveau

03:00.1 Audio device: nVidia Corporation Device 0be5 (rev a1)
        Subsystem: eVga.com. Corp. Device 1480
        Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Interrupt: pin B routed to IRQ 5
        Region 0: [virtual] Memory at d7ffc000 (32-bit, non-prefetchable) [disabled] [size=16K]
        Capabilities: <access denied>

04:00.0 VGA compatible controller: nVidia Corporation Device 06c0 (rev a3)
        Subsystem: eVga.com. Corp. Device 1480
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 16
        Region 0: Memory at dc000000 (32-bit, non-prefetchable) [size=32M]
        Region 1: Memory at c0000000 (64-bit, prefetchable) [size=128M]
        Region 3: Memory at cc000000 (64-bit, prefetchable) [size=64M]
        Region 5: I/O ports at cf00 [size=128]
        [virtual] Expansion ROM at c8000000 [disabled] [size=512K]
        Capabilities: <access denied>
        Kernel driver in use: nvidia
        Kernel modules: nvidia, nvidiafb, nouveau

04:00.1 Audio device: nVidia Corporation Device 0be5 (rev a1)
        Subsystem: eVga.com. Corp. Device 1480
        Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0, Cache Line Size: 64 bytes
        Interrupt: pin B routed to IRQ 5
        Region 0: Memory at dfffc000 (32-bit, non-prefetchable) [size=16K]
        Capabilities: <access denied>

And the only differences I see relate specifically to the memory mapping:

myComputer:~> diff card1 card2
3c3
< Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
---
> Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
7,11c7,11
< Region 0: Memory at d4000000 (32-bit, non-prefetchable) [size=32M]
< Region 1: Memory at b0000000 (64-bit, prefetchable) [size=128M]
< Region 3: Memory at bc000000 (64-bit, prefetchable) [size=64M]
< Region 5: I/O ports at df00 [disabled] [size=128]
< [virtual] Expansion ROM at b8000000 [disabled] [size=512K]
---
> Region 0: Memory at dc000000 (32-bit, non-prefetchable) [size=32M]
> Region 1: Memory at c0000000 (64-bit, prefetchable) [size=128M]
> Region 3: Memory at cc000000 (64-bit, prefetchable) [size=64M]
> Region 5: I/O ports at cf00 [size=128]
> [virtual] Expansion ROM at c8000000 [disabled] [size=512K]
18c18
< Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
---
> Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
19a20
> Latency: 0, Cache Line Size: 64 bytes
21c22
< Region 0: [virtual] Memory at d7ffc000 (32-bit, non-prefetchable) [disabled] [size=16K]
---
> Region 0: Memory at dfffc000 (32-bit, non-prefetchable) [size=16K]

I have a RAID 5 system, ICH10R, on a eVGA classified x58 motherboard. The RAID failed because of an application failure in Windows 7 (dual boot) which forced a hard restart.

I rebuilt the array using the Intel Matrox Storage Manager app in Windows 7 (all automatic), but now the random characters changed and my drives won't mount. Basically the random prefix changed and now Kubuntu doesn't know where to look. I have labeled the volume "STRIPE" and it had:

/dev/mapper/isw_bedfeijgaj_STRIPE#

where # is the number of my partition. Now the device shows up as:

/dev/mapper/isw_bedfhcecfj_STRIPE#

so I had to change all of the entries in /etc/fstab. The reason for my asking here is two fold:

1. Would there (by default) be any other places where the device would be listed explicitly & I should change it?

2. Is there approach I could've taken so that the characters wouldn't be random and wouldn't change after a rebuild? Specifically, I had to [go through hell] (https://bugs.launchpad.net/ubuntu/+source/parted/+bug/568050) just to get the system to boot. So I've already exhausted my googling skills, just curious if anyone knows of unpublished settings or things I may have missed?

Thanks!!!

On this page the final post mentions Xvfb as a way to run a fake Xserver.

Is it possible to tie this directly to a specific card as if a display was attached??

Specifically, I'd like to overclock my gpus, but short of that (since the Clock Frequencies tab won't make an appearance), I'd like to at least crank up the fan so that my CUDA kernels can run on a cooler gpu. To do so within the driver framework, a display must be attached to the card. But if I do that, then I can't set the card to "Compute Exclusive" mode.

Basically, does anyone know of a way to attach a fake (software based) display to a specific gpu in Ubuntu?

I would like to boost the Core frequency, shader clock, and memory clock on an eVGA GTX 480 I have in my development box running 10.04. Is it possible to overclock a recent graphics card from within Ubuntu? I tried the Coolbits approach, but I couldn't get this to work. I also tried nvclock, but as it has't been updated since January, 4th, 2009, it doesn't work with my card.

In windows, the driver itself ships with the ability to overclock the cards, but nvidia appears to have left this out of the linux drivers. Has anyone discovered a solution? Or would it be possible to stage the windows drivers within Ubuntu? (ick)

If there is a more appropriate forum to ask this question in, I'd be happy to do so -- but I'm hoping for a solution within Ubuntu. Thanks!

Update: It appears that I may need to have "Coolbit" "5" as discussed here. Hmm. Nope.

Here is the relevant section of my xorg.conf file:

Section "Device"
    Identifier     "Device0"
    Driver         "nvidia"
    VendorName     "NVIDIA Corporation"
    BoardName      "GeForce GTX 460"
    BusID          "PCI:2:0:0"
    Option         "Coolbits" "5"
    Option         "NoLogo" "True"
EndSection

Section "Device"
    Identifier     "Device1"
    Driver         "nvidia"
    VendorName     "NVIDIA Corporation"
    BoardName      "GeForce GTX 480"
    BusID          "PCI:3:0:0"
    Option         "Coolbits" "5"
    Option         "NoLogo" "True"
EndSection

When I restart with Coolbits set to 5, I can enable fan control on GTX 460. I can also set it to maximum performance mode. I cannot change the clock frequencies. I am about to try different levels. But here is what I see (with Coolbits = 5):

Update 2: I've tried driver version - 260.24(beta - nvdeveloper) & 260.19.12 (released today). I am not able to see the "Clock Frequencies" tab for any of the 4xx cards. I can however alter the fan speed for the card with a display attached. I'm going to ask a second question and wait for better drivers to be released.

I'd like to install the movabletype-opensource package for version 5.0 I currently have Ubuntu 10.04 installed. How do I install the experimental package? Do I need to add a site to my apt sources?

A fair while ago (~ 2004) I was told to always create a separate partition for the boot directory. And since then I always have. But after reading a few questions about partitioning questions, I'm curious. Do I still need a /boot partition?

Note: If helpful, I'm dual booting (Win XP or 7) & Ubuntu 10.04. I also have one partition (using ext2/3) which I use to share files between the two operating systems.