Debian :: Using Apt-get For An Embedded Controller?
Jul 21, 2010
I've bought a small embedded controller running with a minimized Lenny for armel. Minimized in a way that there is even no apt-get or dpkg installed on it, due to keep the memory footprint as small as possible.Is there a way to keep a "shadow" system in a directory on my Laptop (e.g. ~/armel), maintained with dependencies by a package manager like apt-get? This would allow to install addition packages including all dependent with a simple apt-get call.Of course, later I have to put the whole tree from my Laptop towards my embedded controller. I've already tried the following:Setup a directory tree behind ~/armel for apt by a copy from my laptops Lenny
"sudo apt-get -o DIR=~/armel -o APT::Architecture=armel update"
to get the repositories, this works fine
"sudo apt-get -o DIR=~/armel -o APT::Architecture=armel -d -f install"
to download the full list of .deb packages, this works fine
I have been using lspci, dmidecode, and mpt-status to get hardware information on my Dell 1950 running Ubuntu 8.10. I'm pretty sure my server is using an embedded SCSI RAID controller from info I got from Dell's site:
PCI-Express is an actual...well, PCI card, right? But dmidecode shows that I have two x8 PCI Express slots that are both available. Sooo...I'm missing something. How am I running a PCI Express SCSI controller without using a PCI Express slot? In the event of not having the kind of info that I did (i.e. the service tag) how would I be able to tell at a glance whether a component like my RAID controller was embedded or not?
I have been trying for a few days to install CentOS 5.4 on an IBM x306 and I cannot get it to properly handle the Adaptec Embedded SATA HostRAID Controller. I have been working with Linux for a few years, but this is new territory for me. I typically use Debian-based distros, but I did some research on the IBM site and found out that RHEL is a supported OS for this machine. So, I decided to give Cent a try. I have some experience with Fedora, so it's not totally foreign to me.
Anyway, I'm a bit confused. Using the IBM RAID utility, I set up a mirrored pair of 1GB SATA HDDs. When I run the Cent installer, it sees the pair as a single array. I am able to partition the array and complete the install, but when I boot into the OS, it sees the drives as 2 separate devices, sda and sdb. I can pull either one of the drives and boot with a single disk, but it doesn't seem to behave as a mirrored pair. If I make changes on sda, they are not replicated to sdb. Also, I can't use the cli or GParted to format the existing space on the array. I get an error either way. I believe this is because Cent doesn't have a driver for the RAID controller, but I don't see why it would work in the installer, but not the installed OS.
My next approach was to start over and attempt to run "linux dd" at the start of the installation. I tried to find the driver for the controller on the IBM site so I could load it when prompted, but couldn't find a newer version than RHEL 4 Update 3 (I'm assuming this would coordinate with CentOS 4.3). I tried it anyway, but when I select the floppy during the setup, it tells me it's not for this version of CentOS. I read several times that there are .img files that might help me in the 'Images' directory of disk one, but I only see diskboot.img, minstg2.img, and stage2.img. I don't think any of these are what I'm looking for. I thought there was supposed to be a drvblock.img or driverdisk.img.
I forced my workplace to forgo windows and opt for linux for web and mail server. I'm setting up Centos 5.4 on it and I ran into a problem. The server machine is a HP Proliant DL120 G5 (quad core processor, 4GB Ram, two SATA drives, 150GB each attached to the hardware RAID Controller on board). RAID is enabled in the BIOS.I pop in the Centos disk and go through the installation process.
When I get to the stage where I partition my hard drive,it is showing one hard drive, not as traditional sda.but as mapper/ddf1_4035305a86a354a45.I looked around and figured that I need to give Centos the raid drivers. I downloaded it from:
[URL]
I follow the instructions and download the aarahci-1.4.17015-1.rhel5.i686.dd.gz file and unzipped it using gunzip. Then on another nix system, i do this:
dd if=aarahci-1.4.17015-1.rhel5.i686.dd of=/dev/sdb bs=1440k Note that I am using a usb floppy drive, hence the sdb. After that, during centos setup, i type: linux updates dd
It asks me where the driver is located. I tell it and the installation continues in the graphical mode. But I still get mapper/ddf1_4035305a86.a354a45 as my drive. I tried to continue to install centos on it. It was successfull but when i do a "df -h" it gives me /dev/mapper/ddf1_4035305a86......a354a45p1 as /boot
/dev/mapper/ddf1_4035305a86......a354a45p2 as / /dev/mapper/ddf1_4035305a86......a354a45p3 as /var /dev/mapper/ddf1_4035305a86......a354a45p4 as /external /dev/mapper/ddf1_4035305a86......a354a45p5 as /swap /dev/mapper/ddf1_4035305a86......a354a45p6 as /home
Well i know why it's giving these, because i set it up that way, but i was hoping it would somehow change to the normal /dev/sda, /dev/sdb. That means that the driver i provided did not work. I have another IBM server (5U) with raid scsi drive and it shows the usual /dev/sda. It also has hardware raid. So i know that there is something wrong with the /dev/mapper/ddf1_4035305a86......a354a45p1 format.
First, is there any way that I can put the aarahci-1.4.17015-1.rhel5.i686.dd (floppy image) on a CD?. I really need to set this up with raid. I know i could simply disable raid in bios and then i would get two normal hard drives sda and sdb. But it has to be a raid setup. Any way to slipstream the driver into the centos dvd? The hp link i provided above, under installation instructions, there are some instructions titled "Important". But I couldn't get it to work.
I searched the menu-bar of gedit, i searched the web, i searched the help of gedit.(try a search on "change background to black for gedit" and similar and you know what i have gone through).If i use the embedded terminal in gedit (the only reason i use gedit at all) its background color is set to white.I barely can read it and it is useless.This is not a gnome-installation. Perhaps i miss a package needed ?
I am to drop Linux onto an ARM9 NAS board to make it more useful than just a network attached storage device.
I know about U-Boot and I can install Debian Linux by booting the system via TFTP, and I can install Linux easily... if I have an internet connection.
My question is, how can I install Debian WITHOUT the use of an internet connection. I am looking into downloading the armel flavour of Debain and using that as a source, but I have no idea on how to install it onto an embedded system as it has no CD/DVD drive. It only has an ethernet port, 2 USB-2.0 ports and a serial UART port.
I've got a laptop running i386 Debian Lenny, and I was able to get Iceweasel to run flash videos by installing packages via debian-multimedia. However, now I need to get embedded streaming (real time) QuickTime videos to display (as part of a project at work). Is there something I should install from the repos for that? Or do I need a Firefox plug-in from the net? I saw a few libs in the repos that seem related, but I wasn't sure if that was what I needed.
I can not get the node or cloud controllers to startup using the init.d scripts. I have a fresh install of CentOS 5.4 with Eucalyptus 1.6.2 I have compiled Eucalyptus and all packages using the RPM supplied from Eucalyptus and utilizing yum installer. I do not currently have any processes or applications listening currently on the ports on the boxes as well. I think it may be a permissions issue or something because I get a "permission denied error", but I am not sure if it is Eucalyptus or CentOS. It looks as if it is not binding to the address on the interface of the NIC. It may be something else however. I have the Node controller, Cloud controller, and Cluster controller on seperate physical boxes. When I try to run either the cloud controller or the node controller I get this message:
Cloud Controller:
[root@cluster-cont ~]# /etc/init.d/eucalyptus-cc start Starting Eucalyptus cluster controller: (13)Permission denied: make_sock: could not bind to address [::]:8774 (13)Permission denied: make_sock: could not bind to address 0.0.0.0:8774
Basically i want it so that the analogue stick on my PS2 controller, which connects via a USB adaptor and is recognized, correctly controls in game characters. It seems to work fine enough in platformers, at least it does with Banjo Kazooie, but with games like perfect Dark it seem to be trying to move forward and look down at the same time, also backward and look up simultaneously and look sideways and move sideways simultaneously , rendering the game all but unplayable.
I have tried running jscal and, apart from the fact that it seems a bit beyond me, it reports "jscal: missing devicename" when i try to run "jscal -c". Do others out there have there analogue sticks working correctly with games like Perfect Dark, or the James Bond games, for example?
Iceweasel, Konqueror, makes no difference. Embedded flash videos make them crash every time. I had to install Flash Control on Iceweasel just to make it usable.
[pid 9417] --- SIGALRM {si_signo=SIGALRM, si_code=SI_KERNEL, si_value={int=1851877730, ptr=0x6e616962}} --- [pid 9520] <... close resumed> ) = ? <unavailable> [pid 9446] +++ killed by SIGALRM +++ [pid 9444] +++ killed by SIGALRM +++ [pid 9520] +++ killed by SIGALRM +++ [pid 9447] +++ killed by SIGALRM +++ +++ killed by SIGALRM +++
The kicker is I can't recall ever installing flash support in the first place.
I'm bitterly regretting the upgrade to 8.2. It's just one damn thing after another.
Recently we bought a server machine in purpose of hosting game servers.Debian is the optimal choice for us but we have a huge problem.We have 2 x HDD Fujitsu MBA3300RC, 300GB, 15000rpm, 16MB [MBA3300RC] , but NO SAS-controller , so standard Debian DVD1 didn't recognized our HDD's during installation.Is there any way to install Debian on our machine without haveing to buy a SAS-controller? answer in "step-by-step" method, i am not very experimentated in linux.
Does Debian have any particular tools or nuances for installing new hardware? I saw some stuff on Ubuntu which is related... but you know.
The dmesg file shows that it's being recognized, but I don't think it's actually being used (i.e. there is no driver installed). This is what's in the dmesg file:
Unless someone has already compiled a driver for Debian Lenny for this hardware, I'm going to have to compile my own I guess. The driver package seems to come with something called mptlinux-4.00.43.00-src.tar.gz which I'm guessing can be compiled for any Linux, but looking at the instructions, it's pretty much beyond me how to get started. It talks about using kernel source to build a module and such and such.
At the school i work in i have a server2k3 server that provides a domain to all the windows clients, aswell as a fedora server that acts as an imaging machine and webserver.
Im rather concious of the fact that if for any reason the Server2k3 server was to die there is no backup of active directory, or anything that can take its place whilst a replacement is found.
So is it possible to use a fedora machine with samba as a secondary domain controller? so it can be used as a login server, and has a copy of AD.
Can Debian act as a Windows Domain Controller? I'm just curious because my boss recently tossed out some old servers and I grabbed them. They're good machines but I can't afford to pay $1000 for Server 2003 R2 just to setup a domain at my place and run Endpoint Protection and such. I've never used a domain setup in Linux before so I thought I'd ask and possibly try to, if Debian/Linux is capable of such a thing.
However, I have the problem I can't see any devices on my IDE port in my Asrock P43 mainboard, because the VIA VT6415 IDE controller isn't recognized. I typed an 'lspci -k', and I can clearly see there wasn't loaded any module to handle this VIA PATA controller.
I have been having trouble installing a working Debian 6.0.1a (Squeeze) desktop; I cannot seem to get Xwindoes working (under KDE). Is it possible that my video monitor is simply incompatible with Squeeze? I am using an old but not ancient Samsung SyncMaster 997DF with PS/2 keyboard and mouse. Output of lspci -nn | grep VGA is: 01:00.0 VGA compatible controller [0300]: Silicon Integrated Systems [SiS] 661/741/760 PCI/AGP or 662/761Gx PCIE VGA Display Adapter [1039:6330]
When I compared the contents of /usr/lib/xorg/modules and output of kdm.log between my working (but no longer trusted) Lenny and my nonworking Squeeze, I found that the driver sis_drv.so seems to be loaded by my Lenny but not by my Squeeze, and someone said that Squeeze uses a new version of Xorg which is incompatible with SIS. If so, does that mean that I need to get a new monitor? A new motherboard with video controller? I never had a video card--- do I need one to use Squeeze with KDE?....
I have a computer and I need to know whether its Sata-controller supports hotplugging. Therefore I (think I) need to know which Sata-controller is used in my computer. Can anyone tell me how I can find that piece of info?
ps. Some info on the environment: It is running an AMD Geode processor which only support IDE and therefore the board has a Pata to Sata converter build in. When I do 'lspci' it only shows the IDE connection: root@voyage:/etc# lspci
00:01.0 Host bridge: Advanced Micro Devices [AMD] CS5536 [Geode companion] Host Bridge (rev 33) 00:01.2 Entertainment encryption device: Advanced Micro Devices [AMD] Geode LX AES Security Block 00:06.0 Ethernet controller: VIA Technologies, Inc. VT6105M [Rhine-III] (rev 96) 00:07.0 Ethernet controller: VIA Technologies, Inc. VT6105M [Rhine-III] (rev 96) 00:08.0 Ethernet controller: VIA Technologies, Inc. VT6105M [Rhine-III] (rev 96) 00:09.0 Ethernet controller: VIA Technologies, Inc. VT6105M [Rhine-III] (rev 96)
I've been trying for probably 10 hours now to get the bcm4306 driver working on Debian. I finally got it installed using ndiswrapper... However, now ndiswrapper says that the hardware is not present! When I run lspci, though, it lists BCM4306 as the network controller. What is going on, and how do I fix it? It's on a Dell latitude D600.
Im trying to install Debian on my server. Some hardware descriptions:
- Inel Xeon processor - 03:01.0 RAID bus controller: Adaptec (formerly DPT) SmartRAID V Controller (rev 01)
There is configured by hardware one RAID 5 on four scsi disks of 73GB. I tryed with Lenny and Squeeze versions and both presented the same error when the install try to install grub:
main-menu[1250]: INFO: Falling back to the package description for auto-install main-menu[1250]: INFO: Falling back to the package description for ai-choosers main-menu[1250]: INFO: Menu item 'grub-installer' selected
I have a very weird problem with a HP DL180 G6 server that uses this controller.
The server has 4 SSDs configured as separate logical devices (no hardware RAID). I can install Wheezy just fine (creating a software RAID1 from sda and sdb). I installed zfs on it and installed a Windows 2008 VM on kvm with the disk image stored on zfs. The entire server froze when Windows was installing updates.
I rebooted and everything worked again, this time I launched iostat -x 2 on a separate ssh session, here's whet I saw after the server froze again:
Thinking that zfs might be causing this, I destroyed the zpool, configured software raid on sdc and sdd, created a volume group and stored the image on a logical volume. The problem happened again.
I opened a ticket with the support of the company that provides the server and they updates the firmware of the controller. Now the server no longer froze while installing Windows updates, with lvm or zfs, so I went back to zfs.
Wanting to be sure, I launched iometer on the windows VM and set it to "4KB, 75% read 100% random" access pattern over 5GB of the Windows drive. The server froze after 7 hours.
I told the company that, and they replaced the SSDs with ones of a different model. I ran the test again and the server froze again, this time after 9 hours. I asked the company to replace the whole server. They did it (but kept the SSDs), I ran the test again and the server froze again, this time after 20 hours.
Now, this is starting to look like some kind of incompatibility between the HP controller and Debian rather than a problem with the controller. The settings are all default. It looks like the entire controller freezes because the system drives (sda, sdb) also show 100% busy despite not being under test from iometer.
I recently assembled a Dell T5400 workstation. I'm getting some strange issues with this machine concerning fan controller.
-Only 2/4 motherboard fans are detected. All fans have RPM signal -The only sensors that are reporting temperatures are for the memory modules,graphic card and disk but not from my Intel CPU's. This chip reports memory module temperatures as they tend to get pretty hot (DDR2-FBDIMM). -Every time a fan speed gets reported with sensors command the bus freezes for a moment. Videos playback and audio chops for a moment and continue normally.
After removing module dell_smm_hwmon the sensors command executed without interrupting multimedia flow but no fan speeds were reported also.
As a temp solution I blacklisted the dell_smm_hwmon module and I solved the high fan noise problems and the bus freeze.
As it turns out i8k module for fan control is very buggy for some DELL models. Can I manually configure it somehow to work?
So to sum up the problems my problems are:
No CPU temperatures reporting. 2 Fans Speed and PWM controller info missing, and freezing the bus when probing to get speeds from those identified (dell_smm_hwmon module issues).
The object is to get fancontrol working in my setup.
I just bought a new server with one of the most recent v3 Intel processors (socket LGA1150). The board (Supermicro X10SLM-F) uses the new Intel NICs (i217LM + i210AT), not supported by the current Debian stable release. Hence, the ethernet adapter is not recognized during the installation process.
I would like to be able to install something as close to stable as possible in a production server such as this one.
I have DL120 Proliant server that has a P212 raid card, if I install Lenny it works fine however I need squeeze. If I upgrade or install a new version of squeeze the raid controller is no longer visible. I have done some snoopping and it seems as though the ciss drivers have been replaced by the hpsa drivers but I still cant seem to get the raid card recognised any body got any tips ?
i am working with an old system that uses a BIOS meant for embedded systems. According to my coworkers this thing boot some version of debian about two years ago. currently I have used there old image and a new one I made of the latest Debian stable build. both images fail to get passed grub.
to be clear the BIOS simply replies "loading grub" takes ten minutes and then crashes.
has anyone ever had trouble with grub crashing systems? this problem seems odd since is did boot with this two years ago and i still have that image.
