Installation problem on HP DL360 G6 server
in [Debian]
|
Prev: openvz
Next: Suggestions for VoIP Phone with G.722
From: Michael Mohn on 13 Feb 2010 07:40 Am 13.02.2010 um 13:24:26 schrieb Emmanuel Chantry: > Le 12/02/2010 23:25, Michael Mohn a écrit : >> Am 12.02.2010 um 23:16:14 schrieb Emmanuel Chantry: >> >> >>> Le 12/02/2010 22:30, Stan Hoeppner a écrit : >>> >>>> Emmanuel Chantry put forth on 2/12/2010 12:32 PM: >>>> >>>> >>>>> Hi, >>>>> >>>>> I've tried to install a Debian Lenny on HP DL360 G6 server. >>>>> I have a RAID 5 smart array configured. >>>>> I use auto install with preseed. My partition are created through LVMs. >>>>> >>>>> The first time I install Debian, everything is ok. >>>>> But when I try to install again over the existing installation with the >>>>> same preseed configuration, I have an error message : "Volume group name >>>>> already in use" with a continue/go back. >>>>> If I choose continue, I have another error message "No root filesystem >>>>> found". >>>>> >>>>> Anyone else has had this error ? Is there a solution for that problem ? >>>>> >>>>> >>>> The first thing you should ask yourself, and should tell us, is why are you >>>> reinstalling over a fresh installation? That right there is odd behavior. >>>> >>>> Is there actually something you want to accomplish with this Proliant server, or >>>> do you just enjoy playing with the Debian installer? >>>> >>>> Another question: You already have the SmartArray controller presenting you >>>> with a single large block device. Why are you even using LVM? Is there >>>> something you are trying to accomplish that can't be accomplished by slicing >>>> that big block up with good 'ol primary and logical partitions? Do you *need* >>>> LVM or are you using it because it's *neat*? >>>> >>>> >>>> >>> I've reinstalling over a fresh installation because if one day I have to reinstall OS for some reason I will install over an existing one. >>> I must validate the HP server for my company, we have many software under Linux we're currently using. I try to establish a procedure as simple as possible with minimal handling for our technicians. >>> >>> LVM is used to allow me to extend a volume if needed for example if I must store a lot of data and my volume and I haven't enough space. >>> >>> We have many different HP servers with 3, 4, 6 disks, but sometimes we don't mount all the disk in the active RAID, there are spares for extending LVM volumes. >>> >> >> >> you should add a preinstall script, that deletes all partitions, i think. >> that way, there is a defined state to start the installation. >> >> bye, >> >> Michael. >> >> >> > I've tried to use a preinstall script with : > > d-i preseed/early_command string \ > wget http://indus.mydomain/remove_parts.sh -O /tmp/remove_parts.sh; sh /tmp/remove_parts.sh > > My script is like that : > > #!/bin/sh > > DISK='/dev/sda' > VG='debian' > > echo "Installing required software utilities" > anna-install parted-udeb > anna-install lvm2-udeb > > modprobe dm-mod > modprobe md-mod > > # Remove each partition > echo "Removing existing partitions on disk $DISK" > > vgremove -f VG > > for v_partition in $(parted -s $DISK print | grep "^ " | tr -s ' ' | cut -d ' ' -f2) > do > parted -s $DISK rm ${v_partition} > done > > echo "All partitions removed" > > This method works on a virtual machine without problems. But when I try this on my HP server it fails. My install is in amd64 not i386. > I've watched the steps of install in the debug console and it seems that in amd64, installation steps are not in the same order. > > In i386 my anna-install command for udeb packages is done but in amd64 is queued for later install, and my script fails. > > Any idea ? > google told me this: ### Partitioning # If the system has free space you can choose to only partition that space. #d-i partman-auto/init_automatically_partition select biggest_free # Alternatively, you can specify a disk to partition. The device name must # be given in traditional non-devfs format. # Note: A disk must be specified, unless the system has only one disk. # For example, to use the first SCSI/SATA hard disk: #d-i partman-auto/disk string /dev/sda # In addition, you'll need to specify the method to use. # The presently available methods are: "regular", "lvm" and "crypto" d-i partman-auto/method string lvm # If one of the disks that are going to be automatically partitioned # contains an old LVM configuration, the user will normally receive a # warning. This can be preseeded away... d-i partman-lvm/device_remove_lvm boolean true # The same applies to pre-existing software RAID array: d-i partman-md/device_remove_md boolean true # And the same goes for the confirmation to write the lvm partitions. d-i partman-lvm/confirm boolean true # You can choose one of the three predefined partitioning recipes: # - atomic: all files in one partition # - home: separate /home partition # - multi: separate /home, /usr, /var, and /tmp partitions d-i partman-auto/choose_recipe select atomic # Or provide a recipe of your own... # The recipe format is documented in the file devel/partman-auto-recipe.txt. # If you have a way to get a recipe file into the d-i environment, you can # just point at it. #d-i partman-auto/expert_recipe_file string /hd-media/recipe that could be worth a try ;) bye, Michael.
From: Emmanuel Chantry on 13 Feb 2010 07:40 Le 13/02/2010 13:36, Michael Mohn a �crit : > > Am 13.02.2010 um 13:24:26 schrieb Emmanuel Chantry: > >> Le 12/02/2010 23:25, Michael Mohn a �crit : >>> Am 12.02.2010 um 23:16:14 schrieb Emmanuel Chantry: >>> >>> >>>> Le 12/02/2010 22:30, Stan Hoeppner a �crit : >>>> >>>>> Emmanuel Chantry put forth on 2/12/2010 12:32 PM: >>>>> >>>>> >>>>>> Hi, >>>>>> >>>>>> I've tried to install a Debian Lenny on HP DL360 G6 server. >>>>>> I have a RAID 5 smart array configured. >>>>>> I use auto install with preseed. My partition are created through >>>>>> LVMs. >>>>>> >>>>>> The first time I install Debian, everything is ok. >>>>>> But when I try to install again over the existing installation >>>>>> with the >>>>>> same preseed configuration, I have an error message : "Volume >>>>>> group name >>>>>> already in use" with a continue/go back. >>>>>> If I choose continue, I have another error message "No root >>>>>> filesystem >>>>>> found". >>>>>> >>>>>> Anyone else has had this error ? Is there a solution for that >>>>>> problem ? >>>>>> >>>>>> >>>>> The first thing you should ask yourself, and should tell us, is >>>>> why are you >>>>> reinstalling over a fresh installation? That right there is odd >>>>> behavior. >>>>> >>>>> Is there actually something you want to accomplish with this >>>>> Proliant server, or >>>>> do you just enjoy playing with the Debian installer? >>>>> >>>>> Another question: You already have the SmartArray controller >>>>> presenting you >>>>> with a single large block device. Why are you even using LVM? Is >>>>> there >>>>> something you are trying to accomplish that can't be accomplished >>>>> by slicing >>>>> that big block up with good 'ol primary and logical partitions? >>>>> Do you *need* >>>>> LVM or are you using it because it's *neat*? >>>>> >>>>> >>>>> >>>> I've reinstalling over a fresh installation because if one day I >>>> have to reinstall OS for some reason I will install over an >>>> existing one. >>>> I must validate the HP server for my company, we have many software >>>> under Linux we're currently using. I try to establish a procedure >>>> as simple as possible with minimal handling for our technicians. >>>> >>>> LVM is used to allow me to extend a volume if needed for example if >>>> I must store a lot of data and my volume and I haven't enough space. >>>> >>>> We have many different HP servers with 3, 4, 6 disks, but sometimes >>>> we don't mount all the disk in the active RAID, there are spares >>>> for extending LVM volumes. >>>> >>> >>> >>> you should add a preinstall script, that deletes all partitions, i >>> think. >>> that way, there is a defined state to start the installation. >>> >>> bye, >>> >>> Michael. >>> >>> >>> >> I've tried to use a preinstall script with : >> >> d-i preseed/early_command string \ >> wget http://indus.mydomain/remove_parts.sh -O >> /tmp/remove_parts.sh; sh /tmp/remove_parts.sh >> >> My script is like that : >> >> #!/bin/sh >> >> DISK='/dev/sda' >> VG='debian' >> >> echo "Installing required software utilities" >> anna-install parted-udeb >> anna-install lvm2-udeb >> >> modprobe dm-mod >> modprobe md-mod >> >> # Remove each partition >> echo "Removing existing partitions on disk $DISK" >> >> vgremove -f VG >> >> for v_partition in $(parted -s $DISK print | grep "^ " | tr -s ' ' | >> cut -d ' ' -f2) >> do >> parted -s $DISK rm ${v_partition} >> done >> >> echo "All partitions removed" >> >> This method works on a virtual machine without problems. But when I >> try this on my HP server it fails. My install is in amd64 not i386. >> I've watched the steps of install in the debug console and it seems >> that in amd64, installation steps are not in the same order. >> >> In i386 my anna-install command for udeb packages is done but in >> amd64 is queued for later install, and my script fails. >> >> Any idea ? >> > > > google told me this: > > ### Partitioning > # If the system has free space you can choose to only partition that space. > #d-i partman-auto/init_automatically_partition select biggest_free > > # Alternatively, you can specify a disk to partition. The device name must > # be given in traditional non-devfs format. > # Note: A disk must be specified, unless the system has only one disk. > # For example, to use the first SCSI/SATA hard disk: > #d-i partman-auto/disk string /dev/sda > # In addition, you'll need to specify the method to use. > # The presently available methods are: "regular", "lvm" and "crypto" > d-i partman-auto/method string lvm > > # If one of the disks that are going to be automatically partitioned > # contains an old LVM configuration, the user will normally receive a > # warning. This can be preseeded away... > d-i partman-lvm/device_remove_lvm boolean true > # The same applies to pre-existing software RAID array: > d-i partman-md/device_remove_md boolean true > # And the same goes for the confirmation to write the lvm partitions. > d-i partman-lvm/confirm boolean true > > # You can choose one of the three predefined partitioning recipes: > # - atomic: all files in one partition > # - home: separate /home partition > # - multi: separate /home, /usr, /var, and /tmp partitions > d-i partman-auto/choose_recipe select atomic > > # Or provide a recipe of your own... > # The recipe format is documented in the file devel/partman-auto-recipe.txt. > # If you have a way to get a recipe file into the d-i environment, you can > # just point at it. > #d-i partman-auto/expert_recipe_file string /hd-media/recipe > > > that could be worth a try ;) > > > bye, > > Michael. > I've already tried this one. That works on virtual machine in amd64 but not on the HP Server. It's the first thing I've done ;) Emmanuel
From: Stan Hoeppner on 13 Feb 2010 21:50 Emmanuel Chantry put forth on 2/13/2010 6:39 AM: > I've already tried this one. That works on virtual machine in amd64 but > not on the HP Server. It's the first thing I've done ;) Have you tried contacting HP support? IIRC, they support Debian. -- Stan -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/4B776322.7050104(a)hardwarefreak.com
From: Stan Hoeppner on 13 Feb 2010 21:50 Emmanuel Chantry put forth on 2/13/2010 6:24 AM: > This method works on a virtual machine without problems. But when I try I'm not at all familiar with preseed. What I can tell you is that VM guests typically deal with abstracted phantom hardware because the virtual machine itself is a phantom. That's the whole point. Guests never see the real hardware, or BIOS, or etc, except in the case of paravirtualization, but afaik, those guests still don't see the real hardware picture, only parts of it, specifically device drivers. Testing OS installs, partitioning, etc, is not something you ever want to be doing in a virtual machine environment as the results will always be different than on real bare metal hardware. The whole point of virtualization is abstracting the underlying hardware from the guest operating system. And yet, you are surprised by the results? -- Stan -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/4B7762B3.9080008(a)hardwarefreak.com
From: Alex Samad on 14 Feb 2010 01:30
On Sat, Feb 13, 2010 at 08:42:42PM -0600, Stan Hoeppner wrote: > Emmanuel Chantry put forth on 2/13/2010 6:39 AM: > > > I've already tried this one. That works on virtual machine in amd64 but > > not on the HP Server. It's the first thing I've done ;) smart arrays don't turn up as /dev/sda from memory /dev/cciss/c0d0 - controller zero and disk zero as for the other problem, lvm writes information onto the hard drives, which is why you are seeing the error message, I believe there is a way to tell the partition manager to clear all partitions on the disk to be installed onto. I would also stick to lvm on smartarrays, much easier to managing moving lv's etc > > Have you tried contacting HP support? IIRC, they support Debian. > -- "President Musharraf, he's still tight with us on the war against terror, and that's what I appreciate. He's a -- he understands that we've got to keep Al Qaeda on the run, and that by keeping him on the run, it's more likely we will bring him to justice." - George W. Bush 08/22/2002 Ruch, OR |