Re: [gentoo-user] BASH Completion - Mixing directories and executables
On 3 October 2011 01:42, Hilco Wijbenga wrote: > foo It's possible that you would prefer zsh's completion style and configurability.
[gentoo-user] [OT]: Searching for information aboyt a VFD/flourescenz display
Hi, I am looking for informations about a VFD (Flourescenz Display) module. Its product number is DS M202-MD-07g-2, where DS stands for "Display Systems", which was part of the "Hegener & Glaser" company. The rest of the product number uses the same numbering scheme as FUTABA uses for its VFDs but I didnt found anything under their datasheets, which ca,e close to mine. There is a controller on this board, which is marked with MHS DS SPEEDY1 F1-80C51AXR (C) INTEL 80, 82 9505 .W31214K It has two (!) 16-pin connectors to control the board. Who can provide some informations how to drive this board or knows of a datasheet? Thank you very much in advance for any help! Best regards, mcc
Re: [gentoo-user] BASH Completion - Mixing directories and executables
On 3 October 2011 01:05, James Broadhead wrote: > On 3 October 2011 01:42, Hilco Wijbenga wrote: >> foo foo? :-) > It's possible that you would prefer zsh's completion style and > configurability. All right, thank you. I've seen ZSH mentioned several times before. I guess it's time to take a look.
[gentoo-user] Strange problem with urxvt
Hi First of all, sorry if this is not the right forum to post this question. I'm having a strange problem with urxvt on one of my computers. When I uses a command which type some text on the screen and then waits for input, the text does not appear in urxvt until I resize the window, or forces an redraw in other ways. It works fine with other terminal emulators, and urxvt works fine on my other computer (with the same settings). I'm using gentoo, xmonad and urxvt Does any of you have a clue on what might be my problem? Best regards Allan W. Nielsen
Re: [gentoo-user] Strange problem with urxvt
On Mon, Oct 3, 2011 at 1:54 PM, Allan Nielsen wrote: > Hi > > First of all, sorry if this is not the right forum to post this question. > > > I'm having a strange problem with urxvt on one of my computers. > > When I uses a command which type some text on the screen and then waits for > input, the text does not appear in urxvt until I resize the window, or > forces an redraw in other ways. > > It works fine with other terminal emulators, and urxvt works fine on my > other computer (with the same settings). > > I'm using gentoo, xmonad and urxvt > > > Does any of you have a clue on what might be my problem? My first guess would be a bug in xmonad, urxvt, or even your video driver. 1) If you use ssh X forwarding to run urxvt on the affected host (so, from a machine running xmonad, but not the machine with the problem, "ssh -x host_with_problem urxvt"), do you see the same artifact? 2) Do the versions of xmonad and urxvt on your various systems differ? Can you upgrade or downgrade xmonad and/or urxvt on the affected system to match the other systems? Does doing so resolve the problem? 3) Does the affected system use a different video driver than the unaffected systems? If it uses the same video driver, is it using a different version? (And if it uses the same video driver, but different versions, does switching to a video driver version matching an unaffected system solve the problem?) -- :wq
[gentoo-user] Hard drives not detected in repeatable order.
Just recently I've run in to problems because my hard drives are not detected in a predictable order, so my fstab that mount /dev/sdb1 and /dev/sdc1 sometimes result in directory trees in the wrong places (/dev/sda seems consistent, but I don't know why). What's the recommended way to fix this? -- Grant Edwards grant.b.edwardsYow! Boy, am I glad it's at only 1971... gmail.com
Re: [gentoo-user] Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 2:40 PM, Grant Edwards wrote: > Just recently I've run in to problems because my hard drives are not > detected in a predictable order, so my fstab that mount /dev/sdb1 and > /dev/sdc1 sometimes result in directory trees in the wrong places > (/dev/sda seems consistent, but I don't know why). > > What's the recommended way to fix this? Mount by volume name or UUID. For example, from fstab: UUID=bddee12c-0047-4b4b-b1d2-9e137a9a8915 / autonoatime 0 1 UUID=857d7723-9dbf-4222-ac28-e05b87b41997 none swapsw 0 0 UUID=8d3648cf-5260-4064-b6a5-50df42acb3d8 /mnt/prevhome autonoatime 0 2 UUID=d7c17623-255b-4313-b50b-99f0f79a0681 /home autonoatime 0 2 UUID=33cc682d-0dd4-4c2b-bf37-876e9f8d3ef4 /boot autonoatime 0 2 To find the UUID: shortcircuit:4@serenity/dev/disk/by-uuid Mon Oct 03 02:49 PM !509 #9 j0 ?0 $ ls -l total 0 lrwxrwxrwx 1 root root 10 Sep 24 00:11 33cc682d-0dd4-4c2b-bf37-876e9f8d3ef4 -> ../../sda1 lrwxrwxrwx 1 root root 10 Sep 24 00:11 857d7723-9dbf-4222-ac28-e05b87b41997 -> ../../sda2 lrwxrwxrwx 1 root root 10 Sep 24 00:11 8d3648cf-5260-4064-b6a5-50df42acb3d8 -> ../../sda3 lrwxrwxrwx 1 root root 10 Sep 24 00:11 bddee12c-0047-4b4b-b1d2-9e137a9a8915 -> ../../sda4 lrwxrwxrwx 1 root root 10 Sep 24 00:11 d7c17623-255b-4313-b50b-99f0f79a0681 -> ../../dm-0 -- :wq
[gentoo-user] Re: Hard drives not detected in repeatable order.
On 2011-10-03, Grant Edwards wrote: > Just recently I've run in to problems because my hard drives are not > detected in a predictable order, so my fstab that mount /dev/sdb1 and > /dev/sdc1 sometimes result in directory trees in the wrong places > (/dev/sda seems consistent, but I don't know why). I still don't know what changed to cause disks ordering to become non-deterministic. I recently upgraded from a single-core CPU to a dual-core CPU. Would that do it? > What's the recommended way to fix this? After a bit more googling, it looks like this is what disk labels are for. Never used them before, but it looks like it's time to give them a go. -- Grant Edwards grant.b.edwardsYow! My mind is making at ashtrays in Dayton ... gmail.com
Re: [gentoo-user] Hard drives not detected in repeatable order.
Am 03.10.2011 20:40, schrieb Grant Edwards: > Just recently I've run in to problems because my hard drives are not > detected in a predictable order, so my fstab that mount /dev/sdb1 and > /dev/sdc1 sometimes result in directory trees in the wrong places > (/dev/sda seems consistent, but I don't know why). > > What's the recommended way to fix this? > Mount by UUID or label. In /etc/fstab, specify UUID=foo or LABEL=bar instead of /dev/sdx1. You can current UUIDs and labels with `ls -l /dev/disk/by-uuid` and `ls -l /dev/disk/by-label`, respectively. Hope this helps, Florian Philipp signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 11:49 AM, Grant Edwards wrote: > On 2011-10-03, Grant Edwards wrote: >> Just recently I've run in to problems because my hard drives are not >> detected in a predictable order, so my fstab that mount /dev/sdb1 and >> /dev/sdc1 sometimes result in directory trees in the wrong places >> (/dev/sda seems consistent, but I don't know why). > > I still don't know what changed to cause disks ordering to become > non-deterministic. I recently upgraded from a single-core CPU to a > dual-core CPU. Would that do it? > >> What's the recommended way to fix this? > > After a bit more googling, it looks like this is what disk labels are > for. Never used them before, but it looks like it's time to give them > a go. They have the advantage over UUID's in that you can set them and therefore can be human readable. Also, if you use a desktop environment, they look nice in file managers. Good luck. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Strange problem with urxvt
On 10/03/2011 10:54 AM, Allan Nielsen wrote: > Hi > > First of all, sorry if this is not the right forum to post this question. > > > I'm having a strange problem with urxvt on one of my computers. > > When I uses a command which type some text on the screen and then waits > for input, the text does not appear in urxvt until I resize the window, > or forces an redraw in other ways. > > It works fine with other terminal emulators, and urxvt works fine on my > other computer (with the same settings). > > I'm using gentoo, xmonad and urxvt > > > Does any of you have a clue on what might be my problem? I have a similar problem with xterm in KDE using radeon. Is that your environment, too, Allan? (You should have provided your X version and video driver information in your question, BTW.) I see it across the board on radeon only, though, with X 1.10.4. I have "focus-follows-mouse" and I just mouse out and back to my xterm and it redraws. It seems any xterm-type terminal does the same thing. I don't think I have ever had the problem with konsole, though.
Re: [gentoo-user] Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 11:40 AM, Grant Edwards wrote: > Just recently I've run in to problems because my hard drives are not > detected in a predictable order, so my fstab that mount /dev/sdb1 and > /dev/sdc1 sometimes result in directory trees in the wrong places > (/dev/sda seems consistent, but I don't know why). > > What's the recommended way to fix this? You can set labels to all the partitions, and set /etc/fstab to use them: my fstab looks like: LABEL=Gentoo/ ext4noatime 0 1 LABEL=Swap noneswapsw 0 0 shm /dev/shmtmpfs nodev,nosuid,noexec 0 0 tmpfs /tmptmpfs defaults,nosuid 0 0 I believe this is the recommended way to use fstab in distros like Fedora and OpenSUSE, because of your use case exactly. You can set labels to ext[234] partitions with e2label, and for NTFS partitions you can use ntfslabel, and to swap partitions with mkswap. I suppose every filesystem in the world has a similar tool to set its label. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Strange problem with urxvt
On Mon, Oct 3, 2011 at 3:09 PM, Bill Longman wrote: > On 10/03/2011 10:54 AM, Allan Nielsen wrote: >> I'm using gentoo, xmonad and urxvt > > I have a similar problem with xterm in KDE using radeon. Is that your > environment, too, Allan? xmonad is a window manager, so he wouldn't be using kwin. I don't know how much KDE or xdg integration it does, though. -- :wq
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 2:03 PM, Canek Peláez Valdés wrote: > On Mon, Oct 3, 2011 at 11:49 AM, Grant Edwards >> After a bit more googling, it looks like this is what disk labels are >> for. Never used them before, but it looks like it's time to give them >> a go. > > They have the advantage over UUID's in that you can set them and > therefore can be human readable. Also, if you use a desktop > environment, they look nice in file managers. AFAIK that benefit of labels can also be a danger. If you have multiple systems and use the same label naming scheme on all of them (for example you call your partitions "root" "home" "swap" etc.) and someday you plug the HDD from one system into the other, it could cause confusion by potentially choosing the wrong one. But someone can correct me if I'm wrong. :)
Re: [gentoo-user] Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 2:01 PM, Florian Philipp wrote: > Am 03.10.2011 20:40, schrieb Grant Edwards: >> Just recently I've run in to problems because my hard drives are not >> detected in a predictable order, so my fstab that mount /dev/sdb1 and >> /dev/sdc1 sometimes result in directory trees in the wrong places >> (/dev/sda seems consistent, but I don't know why). >> >> What's the recommended way to fix this? >> > > Mount by UUID or label. In /etc/fstab, specify UUID=foo or LABEL=bar > instead of /dev/sdx1. You can current UUIDs and labels with `ls -l > /dev/disk/by-uuid` and `ls -l /dev/disk/by-label`, respectively. Or another way I think is easier: run /sbin/blkid which will tell you all of the info at once, such as: /dev/sda1: LABEL="boot" UUID="a3193af5-35e1-4908-bfbd-928e8841ead3" TYPE="ext2" /dev/sda2: LABEL="root" UUID="feaa6b06-5935-491d-9aef-fe1415c380b6" TYPE="ext4" /dev/sda3: LABEL="swap" UUID="da4437c5-6f19-409a-a71a-ee63be6ef2e5" TYPE="swap" /dev/sda4: LABEL="home" UUID="145fb951-6f01-4cff-b221-278b72c0604f" TYPE="ext4" It also tells you the RAID member uuid and sub_uuid for your RAID partitions, and more (man blkid).
Re: [gentoo-user] Hard drives not detected in repeatable order.
On Monday 03 Oct 2011 20:01:16 Florian Philipp wrote: > Am 03.10.2011 20:40, schrieb Grant Edwards: > > Just recently I've run in to problems because my hard drives are not > > detected in a predictable order, so my fstab that mount /dev/sdb1 and > > /dev/sdc1 sometimes result in directory trees in the wrong places > > (/dev/sda seems consistent, but I don't know why). Is it perhaps because /dev/sda is PATA on IDE and the rest are SATA? > > What's the recommended way to fix this? > > Mount by UUID or label. In /etc/fstab, specify UUID=foo or LABEL=bar > instead of /dev/sdx1. You can current UUIDs and labels with `ls -l > /dev/disk/by-uuid` and `ls -l /dev/disk/by-label`, respectively. UUID or labels will solve this problem for sure, but if you are still trying to find out what changed to cause this, have you looked at your BIOS settings and how it recognises the drives? Have you changed anything on the physical side (jumpers, cables, etc)? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Hard drives not detected in repeatable order.
Grant Edwards wrote: Just recently I've run in to problems because my hard drives are not detected in a predictable order, so my fstab that mount /dev/sdb1 and /dev/sdc1 sometimes result in directory trees in the wrong places (/dev/sda seems consistent, but I don't know why). What's the recommended way to fix this? I went the LABEL direction too. Here is a example fstab, mine actually. LABEL=boot/bootext2defaults1 2 LABEL=root/reiserfsdefaults0 1 LABEL=swapnoneswapsw0 0 LABEL=var/varext3defaults0 2 LABEL=portage/usr/portageext3defaults0 2 LABEL=home/homereiserfsdefaults0 2 LABEL=data/dataext4defaults0 2 You can set that when your put the file system on or use the file system tools to set it without formating the partition. It is usually -L and whatever other options you use. Since you already have a install on there, I strongly recommend the later, otherwise you lose your data. Hope that helps. Dale :-) :-)
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 3:28 PM, Paul Hartman wrote: > On Mon, Oct 3, 2011 at 2:03 PM, Canek Peláez Valdés wrote: >> They have the advantage over UUID's in that you can set them and >> therefore can be human readable. Also, if you use a desktop >> environment, they look nice in file managers. > > AFAIK that benefit of labels can also be a danger. If you have > multiple systems and use the same label naming scheme on all of them > (for example you call your partitions "root" "home" "swap" etc.) and > someday you plug the HDD from one system into the other, it could > cause confusion by potentially choosing the wrong one. But someone can > correct me if I'm wrong. :) I think we had that conversation one or two weeks ago, in the context of lvm volume names. -- :wq
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 12:28 PM, Paul Hartman wrote: > On Mon, Oct 3, 2011 at 2:03 PM, Canek Peláez Valdés wrote: >> On Mon, Oct 3, 2011 at 11:49 AM, Grant Edwards >>> After a bit more googling, it looks like this is what disk labels are >>> for. Never used them before, but it looks like it's time to give them >>> a go. >> >> They have the advantage over UUID's in that you can set them and >> therefore can be human readable. Also, if you use a desktop >> environment, they look nice in file managers. > > AFAIK that benefit of labels can also be a danger. If you have > multiple systems and use the same label naming scheme on all of them > (for example you call your partitions "root" "home" "swap" etc.) and > someday you plug the HDD from one system into the other, it could > cause confusion by potentially choosing the wrong one. But someone can > correct me if I'm wrong. :) You are right. But a) if you are swaping harddrives around, you better know what you are doing, and b) nothing "terrible" happens, I believe the first (or last) detected drive with a label in fstab will be mounted. The other one will still be available by UUID and /dev device. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Hard drives not detected in repeatable order.
On 10/03/2011 12:01 PM, Florian Philipp wrote: > Am 03.10.2011 20:40, schrieb Grant Edwards: >> Just recently I've run in to problems because my hard drives are not >> detected in a predictable order, so my fstab that mount /dev/sdb1 and >> /dev/sdc1 sometimes result in directory trees in the wrong places >> (/dev/sda seems consistent, but I don't know why). >> >> What's the recommended way to fix this? >> > > Mount by UUID or label. In /etc/fstab, specify UUID=foo or LABEL=bar > instead of /dev/sdx1. You can current UUIDs and labels with `ls -l > /dev/disk/by-uuid` and `ls -l /dev/disk/by-label`, respectively. Or, if they are ext, use tune2fs -l. One of its output lines will be: Filesystem UUID: 40ea622d-8265-4498-bc89-0c0f9020dffb
Re: [gentoo-user] Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 11:40 AM, Grant Edwards wrote: > Just recently I've run in to problems because my hard drives are not > detected in a predictable order, so my fstab that mount /dev/sdb1 and > /dev/sdc1 sometimes result in directory trees in the wrong places > (/dev/sda seems consistent, but I don't know why). > > What's the recommended way to fix this? > > -- > Grant Edwards grant.b.edwards Yow! Boy, am I glad it's > at only 1971... > gmail.com Did you possibly update your kernel? You might want to run smartctl and see if any drives are developing problems. Check out blkid for more device info and mount either by label or UUID for more reliability. - Mark c2stable ~ # /sbin/blkid -o list device fs_typelabel mount point UUID /dev/sdb1 ext2 (not mounted) b9dd1fb9-5c7b-459c-a7fc-ca8ebe032fcc /dev/sdb2 swap 2ea74bbb-8c1d-4a32-939d-512db4e40a54 /dev/sdb3 linux_raid_member c2stable:3 (in use) de47f991-86d9-8467-0637-635b9c6d0591 /dev/sdb5 linux_raid_member (in use) edb0ed65-6e87-b20e-dc0d-88ba780ef6a3 /dev/sdb6 linux_raid_member c2stable:6 (in use) 249c7331-a820-3540-c8f3-b020fb30a66b /dev/sdb7 linux_raid_member c2stable:7 (in use) ded5e0c6-1a5a-73ad-0194-9557bb4e015a /dev/sdd2 linux_raid_member c2stable:7 (in use) ded5e0c6-1a5a-73ad-0194-9557bb4e015a /dev/sdd3 linux_raid_member c2stable:3 (in use) de47f991-86d9-8467-0637-635b9c6d0591 /dev/sdd4 ext3 (not mounted) d218024b-c219-4e79-8d65-51f13ec43a45 /dev/sde2 linux_raid_member c2stable:7 (in use) ded5e0c6-1a5a-73ad-0194-9557bb4e015a /dev/sde3 linux_raid_member c2stable:3 (in use) de47f991-86d9-8467-0637-635b9c6d0591 /dev/sda1 ext2 boot(not mounted) 4efd04f2-839f-4e11-a32d-8fa25129b541 /dev/sda2 swap ed41217e-abf4-4c60-a985-c0420eb054f3 /dev/sda3 linux_raid_member c2stable:3 (in use) de47f991-86d9-8467-0637-635b9c6d0591 /dev/sda5 linux_raid_member (in use) edb0ed65-6e87-b20e-dc0d-88ba780ef6a3 /dev/sda6 linux_raid_member c2stable:6 (in use) 249c7331-a820-3540-c8f3-b020fb30a66b /dev/sda7 linux_raid_member c2stable:7 (in use) ded5e0c6-1a5a-73ad-0194-9557bb4e015a /dev/sdc1 ext2 (not mounted) 27f276fa-6f62-4769-b9d7-deb511dfc60a /dev/sdc2 swap 0c6a1bd0-291e-4157-9b6a-67f7801c7a5e /dev/sdc3 linux_raid_member c2stable:3 (in use) de47f991-86d9-8467-0637-635b9c6d0591 /dev/sdc5 linux_raid_member (in use) edb0ed65-6e87-b20e-dc0d-88ba780ef6a3 /dev/sdc6 linux_raid_member c2stable:6 (in use) 249c7331-a820-3540-c8f3-b020fb30a66b /dev/sdc7 linux_raid_member c2stable:7 (in use) ded5e0c6-1a5a-73ad-0194-9557bb4e015a /dev/md126 ext3 / d55d673f-76d8-41ec-8dda-908287a97e42 /dev/md3ext4 RAID6root (not mounted) 80c3ebf0-d92f-415f-a6d7-1cdb1ceab593 /dev/md7ext3 VirtualMachines /VirtualMachines eb5f3040-4114-442e-8d34-16859e720310 /dev/md6ext3 (not mounted) be24fdaa-dfb5-461e-8232-bb236fe6f1f3 c2stable ~ #
Re: [gentoo-user] Hard drives not detected in repeatable order.
Am Montag 03 Oktober 2011, 18:40:21 schrieb Grant Edwards: > Just recently I've run in to problems because my hard drives are not > detected in a predictable order, so my fstab that mount /dev/sdb1 and > /dev/sdc1 sometimes result in directory trees in the wrong places > (/dev/sda seems consistent, but I don't know why). > > What's the recommended way to fix this? use uuid to mount. -- #163933
Re: [gentoo-user] Strange problem with urxvt
On 10/03/2011 12:26 PM, Michael Mol wrote: > On Mon, Oct 3, 2011 at 3:09 PM, Bill Longman wrote: >> On 10/03/2011 10:54 AM, Allan Nielsen wrote: >>> I'm using gentoo, xmonad and urxvt >> >> I have a similar problem with xterm in KDE using radeon. Is that your >> environment, too, Allan? > > xmonad is a window manager, so he wouldn't be using kwin. I don't know > how much KDE or xdg integration it does, though. Yes, I was asking about his hardware environment, Michael. I do not see the same problem on my nvidia cards.
[gentoo-user] Re: Hard drives not detected in repeatable order.
On 2011-10-03, Paul Hartman wrote: > On Mon, Oct 3, 2011 at 2:01 PM, Florian Philipp wrote: >> Am 03.10.2011 20:40, schrieb Grant Edwards: >>> Just recently I've run in to problems because my hard drives are not >>> detected in a predictable order, so my fstab that mount /dev/sdb1 and >>> /dev/sdc1 sometimes result in directory trees in the wrong places >>> (/dev/sda seems consistent, but I don't know why). >>> >>> What's the recommended way to fix this? >>> >> >> Mount by UUID or label. In /etc/fstab, specify UUID=foo or LABEL=bar >> instead of /dev/sdx1. You can current UUIDs and labels with `ls -l >> /dev/disk/by-uuid` and `ls -l /dev/disk/by-label`, respectively. > > Or another way I think is easier: run /sbin/blkid which will tell you > all of the info at once, such as: > > /dev/sda1: LABEL="boot" UUID="a3193af5-35e1-4908-bfbd-928e8841ead3" > TYPE="ext2" > /dev/sda2: LABEL="root" UUID="feaa6b06-5935-491d-9aef-fe1415c380b6" > TYPE="ext4" > /dev/sda3: LABEL="swap" UUID="da4437c5-6f19-409a-a71a-ee63be6ef2e5" > TYPE="swap" > /dev/sda4: LABEL="home" UUID="145fb951-6f01-4cff-b221-278b72c0604f" > TYPE="ext4" > > It also tells you the RAID member uuid and sub_uuid for your RAID > partitions, and more (man blkid). That's one I need to remember! -- Grant Edwards grant.b.edwardsYow! I want a WESSON OIL at lease!! gmail.com
[gentoo-user] Re: Hard drives not detected in repeatable order.
On 2011-10-03, Mick wrote: > On Monday 03 Oct 2011 20:01:16 Florian Philipp wrote: >> Am 03.10.2011 20:40, schrieb Grant Edwards: >> > Just recently I've run in to problems because my hard drives are not >> > detected in a predictable order, so my fstab that mount /dev/sdb1 and >> > /dev/sdc1 sometimes result in directory trees in the wrong places >> > (/dev/sda seems consistent, but I don't know why). > > Is it perhaps because /dev/sda is PATA on IDE and the rest are SATA? Two are SATA, one is Firewire. They used to be detected consistently with the two SATA drives first and second and the firewire drive third. Within the past few weeks, that changed, and sometimes the Firewire drive shows up second > UUID or labels will solve this problem for sure, but if you are still > trying to find out what changed to cause this, have you looked at > your BIOS settings and how it recognises the drives? The only thing I can think of recently is the CPU upgrade, but that was a couple months ago. About six weeks ago I updated the kernel from 2.6.37 to 3.6.39, but the drive order randomness didn't start until the past week or two (I only reboot once or twice a month, so it's hard to pin down the date of the change). > Have you changed anything on the physical side (jumpers, cables, > etc)? Not that I know of. -- Grant Edwards grant.b.edwardsYow! How's it going in at those MODULAR LOVE UNITS?? gmail.com
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, 3 Oct 2011 14:28:05 -0500, Paul Hartman wrote: > > They have the advantage over UUID's in that you can set them and > > therefore can be human readable. You can set the UUIDs yourself too, but I think they have to stick to the standard format. > Also, if you use a desktop > > environment, they look nice in file managers. > > AFAIK that benefit of labels can also be a danger. If you have > multiple systems and use the same label naming scheme on all of them > (for example you call your partitions "root" "home" "swap" etc.) and > someday you plug the HDD from one system into the other, it could > cause confusion by potentially choosing the wrong one. But someone can > correct me if I'm wrong. :) If you have multiple systems, the sensible approach it to give each filesystem a unique label, such as hostname-mountpoint, which also has the benefit of making it clear which box a drive came from when you get them mixed up. -- Neil Bothwick This is as bad as it can get-but don't bet on it. signature.asc Description: PGP signature
Re: [gentoo-user] {OT} Development framework with access restriction?
>> I think separate repositories would only be necessary when using >> distributed version control (git) as opposed to centralized >> (subversion). I think subversion's path-based authorization should >> eliminate the need for separate repositories? >> > > Separate repos aren't strictly necessary, but it's much harder to verify > your path permissions than it is to verify that your repositories are > separate. > > The first involves config files and cascading information; the second > involves being able to count to two =) Would multiple repos work in a scenario where different developers have access to different stuff and some stuff should be accessible to multiple devs? I don't think you want the same stuff in more than one repo. It seems like managing multiple repos would get out of hand in that sort of situation and I might be better off with config files and a single repo. - Grant
Re: [gentoo-user] {OT} Development framework with access restriction?
On Mon, Oct 3, 2011 at 5:54 PM, Grant wrote: >>> I think separate repositories would only be necessary when using >>> distributed version control (git) as opposed to centralized >>> (subversion). I think subversion's path-based authorization should >>> eliminate the need for separate repositories? >>> >> >> Separate repos aren't strictly necessary, but it's much harder to verify >> your path permissions than it is to verify that your repositories are >> separate. >> >> The first involves config files and cascading information; the second >> involves being able to count to two =) > > Would multiple repos work in a scenario where different developers > have access to different stuff and some stuff should be accessible to > multiple devs? I don't think you want the same stuff in more than one > repo. It seems like managing multiple repos would get out of hand in > that sort of situation and I might be better off with config files and > a single repo. With SVN, you do have 'externals' available: http://svnbook.red-bean.com/en/1.0/ch07s03.html Where I work, externals are used for holding common code like large libraries which are needed or useful to multiple projects. -- :wq
Re: [gentoo-user] Blank screen in X, no errors in Xorg.log
FYI, this is fixed. I ran 'strace startx > startx.strace.out 2>&1' and this showed; /usr/bin/X: symbol lookup error: /usr/lib64/xorg/modules/drivers/fglrx_drv.so: undefined symbol: GlxInitVisuals2D Which means that i forget to eselect opengl set 1, though it wasn't logged to Xorg.0.log (so it was logged to STDOUT, but i couldn't see that as the screen was blank). Then after fixing that; which: no keychain in (/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/opt/bin:/usr/x86_64-pc-linux-gnu/gcc-bin/4.5.3) /etc/X11/xinit/xinitrc: line 63: exec: xterm: not found /etc/X11/xinit/xinitrc: line 61: xterm: command not found xinit: connection to X server lost So i emerged keychain and xterm and it now works. Of course i didn't need xterm as i'm running gnome, so i could have just kicked that off instead.
Re: [gentoo-user] {OT} Development framework with access restriction?
>> >>> I'm not sure if you are overcomplicating this by trying to use Unix >> >>> permission. Have you instead considered webdav? You can restrict this >> >>> to particular (apache) users/groups, directories, files. It also uses >> >>> lockfiles so with two users editing a file simultaneously will cause a >> >>> warning when you try to save it. >> >> >> >> How does webdav relate to something like subversion? Do they >> >> compliment each other or are they substitutes? >> >> >> >> - Grant >> > >> > WebDAV has no version control. It is just an extension to HTTP for >> > distributed authoring. It supports locking files and methods which make >> > it more filesystem-like. AFAIK SVN uses WebDAV for its HTTP transfer >> > protocol. >> >> Got it. It sounds like I could use WebDAV in conjunction with >> subversion or git. I'll look into that. > > The reason I mentioned webdav(s) earlier was because you can have granular > control of what each user can access. You need specify only what > directory/file you want them to be able to access and they shouldn't have > access to anything else. I believe the same is accomplished through the use of subversion's path-based authorization. > Also, you don't need to give them shell access - which I find > (psychologically) more reassuring. ;-) I like that a lot. The way I understand it, in order to edit a file with subversion, you would check out the file from the repository to your local machine, edit it on your local machine, and commit your changes to the repository. How does the workflow change with webdav? - Grant
[gentoo-user] {OT} Computers and mental/emotional health
Over the years I've found that the time I spend on the computer has a negative affect on my mental/emotional health. It seems to suck the life out of life and impair my ability to function in the real world. I've tried various things to counteract the problem, but the only thing that really works is not getting on the computer, and that works really well. I think part of the problem is caused by my processing information, but I think part is due to radiation/glare from the laptop screen. Has anyone dealt with this successfully? I'd love to know how you did it. - Grant
[gentoo-user] Strange partition on USB stick
I can't recall if I asked this before, but I am looking at a partition on a USB stick which seems to have a FAT16 fs on it and in parted says: == Model: Crucial Gizmo! overdrive (scsi) Disk /dev/sdb: 1023MB Sector size (logical/physical): 512B/512B Partition Table: loop Number Start End SizeFile system Flags 1 0.00B 1023MB 1023MB fat16 == What does "Partition Table: loop" mean? fdisk -l is more confusing: == Disk /dev/sdb: 1022 MB, 1022623744 bytes 32 heads, 61 sectors/track, 1023 cylinders, total 1997312 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk identifier: 0x69737369 This doesn't look like a partition table Probably you selected the wrong device. Device Boot Start End Blocks Id System /dev/sdb1 ? 1869771365 203846088684344761 69 Unknown /dev/sdb2 ? 1701519481 3571400945 934940732+ 73 Unknown /dev/sdb3 ?25732573 0 74 Unknown /dev/sdb4 0 3435113471 17175567360 Empty Partition table entries are not in disk order == What are the partitions shown as sdb1-4? Also Id 73 and 74 are I think reserved Ids? Id 69 is I think Novell Netware 5+ (according to: http://www.win.tue.nl/~aeb/partitions/partition_types-1.html) not FAT16. I am not sure I understand what it is showing me. BTW, cfdisk spews it out right from the start: == FATAL ERROR: Bad primary partition 1: Partition begins after end-of-disk Press any key to exit cfdisk == So, does it have a partition table? How can you explain the fdisk output? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] {OT} Computers and mental/emotional health
On Mon, Oct 3, 2011 at 6:15 PM, Grant wrote: > Over the years I've found that the time I spend on the computer has a > negative affect on my mental/emotional health. It seems to suck the > life out of life and impair my ability to function in the real world. > I've tried various things to counteract the problem, but the only > thing that really works is not getting on the computer, and that works > really well. I think part of the problem is caused by my processing > information, but I think part is due to radiation/glare from the > laptop screen. Sitting inside at a computer all the time generally means you're not getting sunlight. In effect, you get winter depression year-round. Look into getting more sunlight. Failing that, try a full-spectrum lamp. > Has anyone dealt with this successfully? I'd love to > know how you did it. Made most of my friends online. Met my fiancee through a Facebook connection. Also, go outside regularly. -- :wq
Re: [gentoo-user] {OT} Development framework with access restriction?
On Mon, Oct 3, 2011 at 6:01 PM, Grant wrote: >>> >>> I'm not sure if you are overcomplicating this by trying to use Unix >>> >>> permission. Have you instead considered webdav? You can restrict this >>> >>> to particular (apache) users/groups, directories, files. It also uses >>> >>> lockfiles so with two users editing a file simultaneously will cause a >>> >>> warning when you try to save it. >>> >> >>> >> How does webdav relate to something like subversion? Do they >>> >> compliment each other or are they substitutes? >>> >> >>> >> - Grant >>> > >>> > WebDAV has no version control. It is just an extension to HTTP for >>> > distributed authoring. It supports locking files and methods which make >>> > it more filesystem-like. AFAIK SVN uses WebDAV for its HTTP transfer >>> > protocol. >>> >>> Got it. It sounds like I could use WebDAV in conjunction with >>> subversion or git. I'll look into that. >> >> The reason I mentioned webdav(s) earlier was because you can have granular >> control of what each user can access. You need specify only what >> directory/file you want them to be able to access and they shouldn't have >> access to anything else. > > I believe the same is accomplished through the use of subversion's > path-based authorization. > >> Also, you don't need to give them shell access - which I find >> (psychologically) more reassuring. ;-) > > I like that a lot. The way I understand it, in order to edit a file > with subversion, you would check out the file from the repository to > your local machine, edit it on your local machine, and commit your > changes to the repository. How does the workflow change with webdav? It doesn't need to. From the user's perspective, the URI to the repo changes. That's pretty much it. (I.e. we went from svn://servername/path to https://servername/svn/path ) -- :wq
Re: [gentoo-user] {OT} Computers and mental/emotional health
> >> Has anyone dealt with this successfully? I'd love to >> know how you did it. > You're right to worry about thisand I suspect it's also aging related. The older I get, the more sensitive I am to how many hours/day is healthy. I don't think there is a perfect solution, especially as more and more important things in life require an internet device of some kind. It's not uncommon for me to spend 2-3hrs researching something, up to 8hrs working, and then have 2 hrs of other emails/social/community stuff all in one day that involve computers. 12 hours/day in a roughly fixed position indoors is never ever going to be healthy. Especially if it must be kept up for years and years as one gets older. So, I've gathered ideas from others and have come up with my own recommendations: a) avoid going to the computer if you can be doing something else and don't need to be there (once I'm at a computer, there is always something that can make me stay there so avoiding being there in first place is important) b) stand up and take brief walks for whatever at least once/hour while working c) recent research suggests that taking vitamin d tablets starting in ones thirties can have a significant impact on relieving some of the sunlight/lack of being outdoor issues d) try to go to the gym or do some signficant exercise to start the day, this can possibly trick your metabolism to run faster all day long e) what many people do, I find, is simply have days where you don't touch the computer (briefly check cell phone but thats it) f) try to find something in your daily routine that will take you outdoors for at least an hour/day, preferably longer (can be harder for those of us who telecommute) g) try to build regular activities with your family/friends that involve outdoor recreation (build a home pool/take up swimming laps/etc) Nothing will completely remove the fact that modern life is increasingly unhealthy, but the above is at least a good start. Matt -- Matthew Marlowe m...@professionalsysadmin.com Senior Internet Infrastructure Consultant DevOps/VMware/SysAdmin https://www.twitter.com/deploylinux Gentoo Linux Dev "Courage is not simply one of the virtues, but the form of every virtue at the testing point." -- C.S. Lewis
Re: [gentoo-user] {OT} Computers and mental/emotional health
On Oct 4, 2011 7:01 AM, "Matthew Marlowe" wrote: > > > > >> Has anyone dealt with this successfully? I'd love to > >> know how you did it. > > > > You're right to worry about thisand I suspect it's also aging > related. The older I get, the more sensitive I am to how many > hours/day is healthy. > > I don't think there is a perfect solution, especially as more and more > important things in life require an internet device of some kind. > It's not uncommon for me to spend 2-3hrs researching something, up to > 8hrs working, and then have 2 hrs of other emails/social/community > stuff all in one day that involve computers. > 12 hours/day in a roughly fixed position indoors is never ever going > to be healthy. Especially if it must be kept up for years and years > as one gets older. > > So, I've gathered ideas from others and have come up with my own > recommendations: > a) avoid going to the computer if you can be doing something else and > don't need to be there (once I'm at a computer, there is always > something that can make me stay there so avoiding being there in first > place is important) > b) stand up and take brief walks for whatever at least once/hour while working > c) recent research suggests that taking vitamin d tablets starting in > ones thirties can have a significant impact on relieving some of the > sunlight/lack of being outdoor issues > d) try to go to the gym or do some signficant exercise to start the > day, this can possibly trick your metabolism to run faster all day > long > e) what many people do, I find, is simply have days where you don't > touch the computer (briefly check cell phone but thats it) > f) try to find something in your daily routine that will take you > outdoors for at least an hour/day, preferably longer (can be harder > for those of us who telecommute) > g) try to build regular activities with your family/friends that > involve outdoor recreation (build a home pool/take up swimming > laps/etc) > > Nothing will completely remove the fact that modern life is > increasingly unhealthy, but the above is at least a good start. If I may add: try a cup of normal (i.e. non-decaf) coffee about 1 hour after you start using the computer. I recently read in the newspaper that 2-4 cups of caf coffee per day significantly reduce the chance of getting a depression. Rgds,
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 03, 2011 at 12:03:47PM -0700, Canek Pel??ez Vald??s wrote > They have the advantage over UUID's in that you can set them and > therefore can be human readable. Also, if you use a desktop > environment, they look nice in file managers. I assume that name clashes can be avoided by using hostname-label. My question is... are there any circumstances where you can use UUIDs but not labels, or visa versa? If so, I'd prefer to go with the more robust option from day 1, rather than switch later. -- Walter Dnes
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 7:25 PM, Walter Dnes wrote: > On Mon, Oct 03, 2011 at 12:03:47PM -0700, Canek Pel??ez Vald??s wrote > >> They have the advantage over UUID's in that you can set them and >> therefore can be human readable. Also, if you use a desktop >> environment, they look nice in file managers. > > I assume that name clashes can be avoided by using hostname-label. My > question is... are there any circumstances where you can use UUIDs but > not labels, or visa versa? If so, I'd prefer to go with the more robust > option from day 1, rather than switch later. You can use whatever you want whenever you want. They are orthogonal. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
Walter Dnes wrote: On Mon, Oct 03, 2011 at 12:03:47PM -0700, Canek Pel??ez Vald??s wrote They have the advantage over UUID's in that you can set them and therefore can be human readable. Also, if you use a desktop environment, they look nice in file managers. I assume that name clashes can be avoided by using hostname-label. My question is... are there any circumstances where you can use UUIDs but not labels, or visa versa? If so, I'd prefer to go with the more robust option from day 1, rather than switch later. From what I know, they both seem to travel well. If you remove a drive and take it to another system, the UUID and LABELS will go with it. LABELS can be shorter and easier on the human to read tho. I see that as a positive that UUID doesn't have. Dale :-) :-)
Re: [gentoo-user] Strange partition on USB stick
If the data is important, I'd use ddrescue to create an image of the drive, then run testdisk over that image to see if it can untangle the partition table mess. Both are in portage.
[gentoo-user] What is the most error resistant filesystem?
Hello people! Now, I have the same question as this guy: https://bbs.archlinux.org/viewtopic.php?id=66651 I.e., what is the most robust filesystem for Linux? The box will be used as a gateway/firewall for a branch office, so I really couldn't care less about filesystem performance. But the utility power there is horrendous, so I need something that can shrug off a catastrophic power loss, and/or very fast fsck. I'd also appreciate any tips on mount options to further enhance robustness. TIA! Rgds,
Re: [gentoo-user] {OT} Computers and mental/emotional health
On 10/03/2011 10:19 PM, Pandu Poluan wrote: > > > If I may add: try a cup of normal (i.e. non-decaf) coffee about 1 hour > after you start using the computer. > Ok, but how do you survive the first hour?
Re: [gentoo-user] What is the most error resistant filesystem?
On Mon, Oct 3, 2011 at 11:47 PM, Pandu Poluan wrote: > Hello people! > > Now, I have the same question as this guy: > > https://bbs.archlinux.org/viewtopic.php?id=66651 > > I.e., what is the most robust filesystem for Linux? > > The box will be used as a gateway/firewall for a branch office, so I really > couldn't care less about filesystem performance. But the utility power there > is horrendous, so I need something that can shrug off a catastrophic power > loss, and/or very fast fsck. ISO9660? Read-only, error correction, and have logging go over the network to something else. (Well, ISO9660 isn't required; any read-only media with a read-only filesystem would probably do.) -- :wq
Re: [gentoo-user] {OT} Development framework with access restriction?
On 10/03/2011 05:54 PM, Grant wrote: > > Would multiple repos work in a scenario where different developers > have access to different stuff and some stuff should be accessible to > multiple devs? I don't think you want the same stuff in more than one > repo. It seems like managing multiple repos would get out of hand in > that sort of situation and I might be better off with config files and > a single repo. (for the tl;dr, see the last paragraph) Subversion separates authentication and authorization: http://svnbook.red-bean.com/en/1.6/svn.serverconfig.svnserve.html#svn.serverconfig.svnserve.auth You'll hear security people say that a lot, but hopefully an example makes the difference clear. I'll use Apache in my example, because that's what we use, and I'm mostly sure I'm not talking out of my ass this way =) The "authentication" part is your usernames and passwords. Authentication is proving who you are. Each developer has his own username and password -- these only need to be stored once. When you go the Apache route, Apache itself controls the authentication. In the "website" definition, we have, # The SVN "root" which lists all repos, assuming you're allowed to do # that. This would be offered up as e.g. https://svn.example.org/ # Allow from all DAV svn SVNParentPath /var/svn/repos SVNListParentPath on AuthType Basic AuthName "Subversion Repository" AuthUserFile /var/svn/auth/svnusers Require valid-user SSLRequireSSL # Accessible via https://svn.example.org/repo1 # Allow from all DAV svn AuthType Basic AuthName "Repository One" AuthUserFile /var/svn/auth/svnusers AuthzSVNAccessFile /var/svn/auth/authz-repo1 Require valid-user SSLRequireSSL # Accessible via https://svn.example.org/repo2 # Allow from all DAV svn AuthType Basic AuthName "Repository Two" AuthUserFile /var/svn/auth/svnusers AuthzSVNAccessFile /var/svn/auth/authz-repo2 Require valid-user SSLRequireSSL You'll notice that both repos (and the root) use the same AuthUserFile. That's just an Apache 'htpasswd2' file with usernames and encrypted passwords. Some of our developers have access to every repo, but they still go in that file just once. The "authorization" part defines what you're allowed to do once you've authenticated (i.e. we know who you are). Apache calls this "authz" as opposed to "auth" everywhere, and is a subtle distinction that took me embarrassingly long to realize. Each Subversion repository can have its own AuthzSVNAccessFile, and that format is specified somewhere in the Subversion book. Basically, you list which users (from the AuthUserFile) can do what. In the example above, the two repos use different authorization files, because our devs have different permissions in repo1 than they do in repo2. So, to answer your question: you separate your projects into repositories logically, in whatever way makes sense. Then, you define users and permissions to match that. The authentication and authorization are flexible enough that you shouldn't have to duplicate anything.
Re: [gentoo-user] What is the most error resistant filesystem?
On 3 October 2011 20:47, Pandu Poluan wrote: > Hello people! > > Now, I have the same question as this guy: > > https://bbs.archlinux.org/viewtopic.php?id=66651 > > I.e., what is the most robust filesystem for Linux? The *most* robust? Probably something seriously expensive from IBM or similar. I'd go with ReiserFS or Ext3. Both are very good. I use ReiserFS pretty much everywhere. No running-out-of-inode problems with ReiserFS so I prefer it over Ext3. > The box will be used as a gateway/firewall for a branch office, so I really > couldn't care less about filesystem performance. But the utility power there > is horrendous, so I need something that can shrug off a catastrophic power > loss, and/or very fast fsck. Then why not simply use a LiveCD like OpenWall? Unbreakable file system as it's all read-only (or RAM). Can't beat read-only. :-)
Re: [gentoo-user] What is the most error resistant filesystem?
On Oct 4, 2011 11:30 AM, "Michael Mol" wrote: > > On Mon, Oct 3, 2011 at 11:47 PM, Pandu Poluan wrote: > > Hello people! > > > > Now, I have the same question as this guy: > > > > https://bbs.archlinux.org/viewtopic.php?id=66651 > > > > I.e., what is the most robust filesystem for Linux? > > > > The box will be used as a gateway/firewall for a branch office, so I really > > couldn't care less about filesystem performance. But the utility power there > > is horrendous, so I need something that can shrug off a catastrophic power > > loss, and/or very fast fsck. > > ISO9660? Read-only, error correction, and have logging go over the > network to something else. > > (Well, ISO9660 isn't required; any read-only media with a read-only > filesystem would probably do.) Indeed that thought occurred in my mind. But I still need to keep some logs, and have read-write access to /etc Rgds,
Re: [gentoo-user] What is the most error resistant filesystem?
On Mon, Oct 3, 2011 at 9:34 PM, Pandu Poluan wrote: > > On Oct 4, 2011 11:30 AM, "Michael Mol" wrote: >> >> On Mon, Oct 3, 2011 at 11:47 PM, Pandu Poluan wrote: >> > Hello people! >> > >> > Now, I have the same question as this guy: >> > >> > https://bbs.archlinux.org/viewtopic.php?id=66651 >> > >> > I.e., what is the most robust filesystem for Linux? >> > >> > The box will be used as a gateway/firewall for a branch office, so I >> > really >> > couldn't care less about filesystem performance. But the utility power >> > there >> > is horrendous, so I need something that can shrug off a catastrophic >> > power >> > loss, and/or very fast fsck. >> >> ISO9660? Read-only, error correction, and have logging go over the >> network to something else. >> >> (Well, ISO9660 isn't required; any read-only media with a read-only >> filesystem would probably do.) > > Indeed that thought occurred in my mind. But I still need to keep some logs, > and have read-write access to /etc Set / to read-only and put /var in another partition. When you need to modify /etc, you remount / rw, modify, and then remount rw. A a gateway/firewall should not need config changes very often. With a ro filesystem, it doesn't really matter what filesystem do you use. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
Windows gives partitions shorter UUID's too, so that's a non-standard thing on your /etc/fstab. I opted for LABELs. By the way, is it possible to use LABELs without and initrd? I'll start using an initrd before too long, I'll also mess with decorations, but for now, I'd like to keep my setup simple, no initrd. --Spidey
Re: [gentoo-user] What is the most error resistant filesystem?
On 10/03/2011 11:47 PM, Pandu Poluan wrote: > Hello people! > > Now, I have the same question as this guy: > > https://bbs.archlinux.org/viewtopic.php?id=66651 > > I.e., what is the most robust filesystem for Linux? > > The box will be used as a gateway/firewall for a branch office, so I > really couldn't care less about filesystem performance. But the utility > power there is horrendous, so I need something that can shrug off a > catastrophic power loss, and/or very fast fsck. > > I'd also appreciate any tips on mount options to further enhance robustness. Journaling filesystems are as safe as you'll get on commodity hardware. I would choose ext4 because I'm familiar with it, but I'm sure others offer the same options. It's also got the fastest fsck that I'm aware of. >From `man tune2fs`: journal_data When the filesystem is mounted with journalling enabled, all data (not just metadata) is committed into the journal prior to being written into the main filesystem. block_validity (I haven't used this, but spotted it in the man page) The file system will be mounted with the block_validity option enabled, which causes extra checks to be per‐ formed after reading or writing from the file system. This prevents corrupted metadata blocks from causing file system damage by overwriting parts of the inode table or block group descriptors. This comes at the cost of increased memory and CPU overhead, so it is enabled only for debugging purposes. (This option is currently only supported by the ext4 file system driver in 2.6.35+ kernels.)
Re: [gentoo-user] Blank screen in X, no errors in Xorg.log
On Mon, Oct 3, 2011 at 19:01, Adam Carter wrote: > FYI, this is fixed. I ran 'strace startx > startx.strace.out 2>&1' and > this showed; > /usr/bin/X: symbol lookup error: > /usr/lib64/xorg/modules/drivers/fglrx_drv.so: undefined symbol: > GlxInitVisuals2D > > Which means that i forget to eselect opengl set 1, though it wasn't > logged to Xorg.0.log (so it was logged to STDOUT, but i couldn't see > that as the screen was blank). > > Then after fixing that; > which: no keychain in > > (/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/opt/bin:/usr/x86_64-pc-linux-gnu/gcc-bin/4.5.3) > /etc/X11/xinit/xinitrc: line 63: exec: xterm: not found > /etc/X11/xinit/xinitrc: line 61: xterm: command not found > xinit: connection to X server lost > > So i emerged keychain and xterm and it now works. Of course i didn't > need xterm as i'm running gnome, so i could have just kicked that off > instead. > > The default xinitrc or something starts twm, xterm and xclock or something like that, that's normal behavior for first time X.org users. You can toggle that to your needs, or you emerge gnome and add gdm to your default runlevel, and forget about startx. About what kernel configurations you have to check and what you have to make sure to not have checked, you can see it here: http://en.gentoo-wiki.com/wiki/Fglrx You'll want to disable everything related to DRM, basically. Claudio Roberto França Pereira (a.k.a. Spidey) hardMOB - HTForum - @spideybr Engenharia de Computação - UFES 2006/1
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 9:54 PM, Spidey wrote: > Windows gives partitions shorter UUID's too, so that's a non-standard thing > on your /etc/fstab. I opted for LABELs. > By the way, is it possible to use LABELs without and initrd? I'll start > using an initrd before too long, I'll also mess with decorations, but for > now, I'd like to keep my setup simple, no initrd. Yeah, labels are a feature of mount, an initramfs (really, totally different from an initrd) has nothing to do with it. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
> By the way, is it possible to use LABELs without and initrd? I'll start > using an initrd before too long, I'll also mess with decorations, but for > now, I'd like to keep my setup simple, no initrd. AND what bootloaders can use LABEL/UUID? Can grub's device.map use them?
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 10:02 PM, Adam Carter wrote: >> By the way, is it possible to use LABELs without and initrd? I'll start >> using an initrd before too long, I'll also mess with decorations, but for >> now, I'd like to keep my setup simple, no initrd. > > AND what bootloaders can use LABEL/UUID? Can grub's device.map use them? I believe grub uses its own partition scheme (hd(0,0)), so fstab should not matter to it unless you use the root= option. But even with the root= option, grub can understand labels, and so grub2 (I'm using it with labels right now). LILO I don't know, I haven't used it in years. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Tue, Oct 4, 2011 at 12:02 AM, Adam Carter wrote: >> By the way, is it possible to use LABELs without and initrd? I'll start >> using an initrd before too long, I'll also mess with decorations, but for >> now, I'd like to keep my setup simple, no initrd. > > AND what bootloaders can use LABEL/UUID? Can grub's device.map use them? I think you need to use grub2 for that.
Re: [gentoo-user] Strange partition on USB stick
On Tuesday 04 Oct 2011 04:39:45 Adam Carter wrote: > If the data is important, I'd use ddrescue to create an image of the > drive, then run testdisk over that image to see if it can untangle the > partition table mess. Both are in portage. Well, that's the thing: I'm not sure that there is a mess. At least not as far as parted is concerned, which can read the partition table properly. I suspect that fdisk (unlike parted) is not capable of reading the device correctly. I forgot to say that when mounted the USB stick shows not partitions (i.e. there is no sdb1, sdb2, etc.) To access the fs I must do something like: pmount /dev/sdb and then all is lists under /media/sdb. It is like a big floppy. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 10:10 PM, Paul Hartman wrote: > On Tue, Oct 4, 2011 at 12:02 AM, Adam Carter wrote: >>> By the way, is it possible to use LABELs without and initrd? I'll start >>> using an initrd before too long, I'll also mess with decorations, but for >>> now, I'd like to keep my setup simple, no initrd. >> >> AND what bootloaders can use LABEL/UUID? Can grub's device.map use them? > > I think you need to use grub2 for that. You are right: for grub-legacy you need to use the old hd(x,y) thingy. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] What is the most error resistant filesystem?
On Mon, Oct 3, 2011 at 10:47 PM, Pandu Poluan wrote: > The box will be used as a gateway/firewall for a branch office, so I really > couldn't care less about filesystem performance. But the utility power there > is horrendous, so I need something that can shrug off a catastrophic power > loss, and/or very fast fsck. I've lost XFS and JFS filesystems in the past due to their failure to recover after sudden power loss. Ext3/4 have not failed me (yet). But my question is, why don't you use a UPS and monitoring software to perform a proper (clean) shutdown when power's off and battery is running low. Some UPS also support automatic power-on once things are normal again, in case this is an unattended box that locals can't be bothered with rebooting themselves. I can think of making a complicated system with read-only boot media (cd/dvd/mmc/whatever) which attempts recovery of important data (logs created since last backup) to a spare partition, RAM drive or the Internet, then repartitions & reinstalls itself to the harddrive and restores the recovered data. Optionally downloading updated configs from Internet. (think kiosk distros).
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
> You are right: for grub-legacy you need to use the old hd(x,y) thingy. Which i assume suffers from the same reassignment risk as the kernel's /dev/sdX naming that prompted this discussion. Looks I'll be moving to grub2.
Re: [gentoo-user] Strange partition on USB stick
On Tue, Oct 4, 2011 at 12:18 AM, Mick wrote: > On Tuesday 04 Oct 2011 04:39:45 Adam Carter wrote: >> If the data is important, I'd use ddrescue to create an image of the >> drive, then run testdisk over that image to see if it can untangle the >> partition table mess. Both are in portage. > > Well, that's the thing: I'm not sure that there is a mess. At least not as > far as parted is concerned, which can read the partition table properly. > > I suspect that fdisk (unlike parted) is not capable of reading the device > correctly. > > I forgot to say that when mounted the USB stick shows not partitions (i.e. > there is no sdb1, sdb2, etc.) To access the fs I must do something like: > > pmount /dev/sdb > > and then all is lists under /media/sdb. It is like a big floppy. I think that's your answer. The "partition table" looks funny because it isn't one. :) It is somewhat common. I've had some myself that are like that.
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 10:25 PM, Adam Carter wrote: >> You are right: for grub-legacy you need to use the old hd(x,y) thingy. > > Which i assume suffers from the same reassignment risk as the kernel's > /dev/sdX naming that prompted this discussion. Looks I'll be moving to > grub2. That's a good idea anyway, given that grub is in life support by its maintainers. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
Am 04.10.2011 07:09, schrieb Canek Peláez Valdés: > On Mon, Oct 3, 2011 at 10:02 PM, Adam Carter wrote: >>> By the way, is it possible to use LABELs without and initrd? I'll start >>> using an initrd before too long, I'll also mess with decorations, but for >>> now, I'd like to keep my setup simple, no initrd. >> >> AND what bootloaders can use LABEL/UUID? Can grub's device.map use them? > > I believe grub uses its own partition scheme (hd(0,0)), so fstab > should not matter to it unless you use the root= option. But even with > the root= option, grub can understand labels, and so grub2 (I'm using > it with labels right now). > > LILO I don't know, I haven't used it in years. > > Regards. In my experience, grub's partition numbering is more stable than /dev. hd(0,*) tends to be the device on which grub is installed, even if device numbering in /dev changes. The more tricky stuff is defining the root=/dev/* kernel parameter. Fortunately, starting with 2.6.37, you can use a UUID here as well (but not label, that support was removed a few years ago): http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=b5af921ec02333e943efb59aca4f56b78fc0e100 Regards, Florian Philipp signature.asc Description: OpenPGP digital signature
Re: [gentoo-user] Re: Hard drives not detected in repeatable order.
On Mon, Oct 3, 2011 at 10:29 PM, Florian Philipp wrote: > Am 04.10.2011 07:09, schrieb Canek Peláez Valdés: >> On Mon, Oct 3, 2011 at 10:02 PM, Adam Carter wrote: By the way, is it possible to use LABELs without and initrd? I'll start using an initrd before too long, I'll also mess with decorations, but for now, I'd like to keep my setup simple, no initrd. >>> >>> AND what bootloaders can use LABEL/UUID? Can grub's device.map use them? >> >> I believe grub uses its own partition scheme (hd(0,0)), so fstab >> should not matter to it unless you use the root= option. But even with >> the root= option, grub can understand labels, and so grub2 (I'm using >> it with labels right now). >> >> LILO I don't know, I haven't used it in years. >> >> Regards. > > In my experience, grub's partition numbering is more stable than /dev. > hd(0,*) tends to be the device on which grub is installed, even if > device numbering in /dev changes. > > The more tricky stuff is defining the root=/dev/* kernel parameter. > Fortunately, starting with 2.6.37, you can use a UUID here as well (but > not label, that support was removed a few years ago): > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=b5af921ec02333e943efb59aca4f56b78fc0e100 Mmmh. This overrides the root option from grub? Then it will also work in LILO. Regards. -- Canek Peláez Valdés Posgrado en Ciencia e Ingeniería de la Computación Universidad Nacional Autónoma de México
[gentoo-user] Kernel source servers compromised?
Fair enough, but chkrootkit is not the most maintained package. Last version was released in July 2009. http://www.theregister.co.uk/2011/10/04/linux_repository_res/ -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Kernel source servers compromised?
On Tue 04 Oct 2011 11:11:22 AM IST, Mick wrote: > Fair enough, but chkrootkit is not the most maintained package. Last version > was released in July 2009. > > http://www.theregister.co.uk/2011/10/04/linux_repository_res/ This is a quite old news and since then Linus has moved the kernel to github -- Nilesh Govindarajan http://nileshgr.com
Re: [gentoo-user] Strange partition on USB stick
On Tue, 4 Oct 2011 00:27:50 -0500 Paul Hartman wrote: > On Tue, Oct 4, 2011 at 12:18 AM, Mick > wrote: > > On Tuesday 04 Oct 2011 04:39:45 Adam Carter wrote: > >> If the data is important, I'd use ddrescue to create an image of > >> the drive, then run testdisk over that image to see if it can > >> untangle the partition table mess. Both are in portage. > > > > Well, that's the thing: I'm not sure that there is a mess. At > > least not as far as parted is concerned, which can read the > > partition table properly. > > > > I suspect that fdisk (unlike parted) is not capable of reading the > > device correctly. > > > > I forgot to say that when mounted the USB stick shows not > > partitions (i.e. there is no sdb1, sdb2, etc.) To access the fs I > > must do something like: > > > > pmount /dev/sdb > > > > and then all is lists under /media/sdb. It is like a big floppy. > > I think that's your answer. The "partition table" looks funny because > it isn't one. :) It is somewhat common. I've had some myself that are > like that. > I have a 4G Sandisk that does that too. It does everything a regular USB stick does except a) create a proper partition table and b) be booted from -- Alan McKinnnon alan.mckin...@gmail.com
