[gentoo-user] resize raid1 array
Hello, I had a raid1/mirror array of two 200G disks. Then one failed and I thought 'lets get two 500G disks and just ease them in, they are cheap'. So I added the full 500G partition from the first disk to the degraded array, watched the resync, removed the remaining 200G disk and added the last 500G disk. My hope now was that I would be able to tell the kernel that md0 now is 500G, not 200G. But alas, I have not been able to do so. Is there a way to do this? Or is the only way forward to shrink the partitions and create new 300G ones alongside? Thanks, Rasmus -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] resize raid1 array
On Thu, Feb 07, 2008 at 06:50:57AM -0500, Simon Turner wrote: > Hi Rasmus, > you will first need to resize your md device. Using mdadm, that > would be done with --grow (check the man, `mdadm /dev/md123 --grow > --size=500G` should do the trick). Ah yes, thanks for that. For some reason I had not noted that use of --grow... mdadm /dev/md0 --grow --size=max did the trick nicely thanks. > But that will leave your filesystem intact at the current size, so > you'll have to resize the filesystem next. I know ext2&3 support > "growing" but I don't know about other filesystems. I use LVM(2) and ext3 so I am ok there. > However, since it's a simple raid one, you could create a new raid > device with one drive, format it at 500G, copy the stuff over from the > old raid1, drop the old raid1 device and add it to the new one. Yes, thats true. But since this is my root, var, home, etc etc that would require me to go to single-user mode for the duration of the copy. The resync-in-background is much nicer :) Thanks for your help. Cheers, Rasmus -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] My portage is hosed (segfaults) and I dont know why...
On Thu, Feb 07, 2008 at 03:39:48PM +, Neil Bothwick wrote: > On Thu, 7 Feb 2008 16:31:29 +0100, Rasmus Andersen wrote: > > > How do I reestablish python > > and/or portage with the least pain? I have both in my PKGDIR, can I > > 'just' untar them on top of the existing installation? > > Yes, untar then in /, ignoring the warning about invalid data at the end > (this is metadata added by portage). Once things are working again, > re-emerge them properly. Thanks. That did it. Rasmus -- gentoo-user@lists.gentoo.org mailing list
[gentoo-user] My portage is hosed (segfaults) and I dont know why...
Hello, It seems that my portage or python installation or the combination thereof has become borked. Whenever I run a portage-related program (emerge, eclean, portageq, etc) I get segfaults. Due to these being scripts, gdb is not of much help... I am able to start python normally and get it to add numbers so python is not completely hosed. But I lack another python program of substance to test whats wrong. The last thing I emerged successfully was iptables and it seems unlikely that they had an influence on python/portage. ... Thinking again I have rdiff-backup around. Testing that shows a small backup to go ok and a largish backup to fail with sigsegv. The box as such is busy enough, memorywise, and survives parallel kernel compiles and untars fine. So, I guess I have two questions: One, do anyone have a reasonable opinion of what is wrong? Failing that, two: How do I reestablish python and/or portage with the least pain? I have both in my PKGDIR, can I 'just' untar them on top of the existing installation? Thanks, Rasmus -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Ghosting a Ext3 partition
On Sun, Mar 02, 2008 at 05:51:06PM +1300, Mark Kirkwood wrote: > I wrote: >> >> >> If you want to back the system up while it is running (in particular /), >> then you need to use a tool that understands how to create a backup image >> that is valid (i.e will boot) - something like xfsdump, *dumpe2fs* etc or >> smart tar/dump based tools like Amanda. > Hmm - dunno what I was thinking there - 'dumpe2fs' is completely wrong, > should have written 'dump', sorry! If you do backup live filesystems/data then dump is on par with dd; both read from the underlying device and might bypass the kernel's page cache. Ie., there might be unwritten data cached thats not on disk yet. Tar/rdiff-backup/etc reads through the pagecache and avoids this problem. The dump people talk a bit about this themselves on http://dump.sourceforge.net/isdumpdeprecated.html Note I dont want to dis dump, backing up live filesystems is just tricky (depending on your consistency requirements :) and dump adds another level to that. Cheers, Rasmus -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Ghosting a Ext3 partition
On Sun, Mar 02, 2008 at 09:51:47PM +1300, Mark Kirkwood wrote: > Understood - I have seen that article too. I must say, I've mainly had > experience with 'dump' on Freebsd and 'xfsdump' on Linux, and never had > restore issues with *either* of these. Now I'm not sure whether these are > supposed to be better than 'dump' on Linux aimed at ext2|3 filesystems - > certainly Freebsd's 'dump' has an option to tell it that it is dumping a > 'live' filesystem, and the man pages for xfsrestore have notes concerning > what happens when restoring an (xfs)dump from a 'live' filesystem - so they > may well be! FreeBSD's softupdates should make filesystem state always consistent, metadatawise. Or so I think I remember, its been a while. That might aleviate some of the problems noted on the dump page I referenced. > On the other hand I've certainly routinely seen cases of people using dd > (rsync, cpio, tar etc) and coming to grief at restore time. I am reluctant > to suggest that folks use xfs and hence get access to xfsdump, as one of > the nice things about Linux is the choice of a variety of filesystems - > but it is pretty important to get able to backup of (for instance ) / ... > and you usually don't have much option other than doing it live! I use rdiff-backup for my backups but then again I have low requirements wrt. consistency outside file-level. I have considered LVM snapshots since I use LVM already but havent bothered so far. Cheers, Rasmus -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Limiting portage sync
On Tue, May 20, 2008 at 02:09:26AM +0100, Matt Harrison wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > I was just wondering, while watching an emerge --sync, is it possible (or > advisable) to limit what portage syncs? For example, I have a server here > that doesn't use X, kde, gnome, java, and a number of other fairly large > areas of portage. It would be nice if I could speed up syncing by excluding > these categories from the process. I do this and find it works like a charm. The actual sync time is not an issue for me since I have a decent connection but the 'updating cache' time went down by at least 75% for me (old computer). Which made 'emerge --sync' a lot less of a 'start this, go for lunch' command for me. I have had a tiny amount of issues with this since I started approx. 4 years ago, the biggest issue is that since you lobotomize the metadata tree you get some odd errors. But I find it worth it. Cheers, Rasmus -- gentoo-user@lists.gentoo.org mailing list
Re: [gentoo-user] Re: thunar won't build?
On Tue, Feb 13, 2007 at 04:30:54PM +, Grant Edwards wrote: > http://bugs.gentoo.org/buglist.cgi?quicksearch=thunar > > > http://bugs.gentoo.org/show_bug.cgi?id=166573 > > How in hell are you supposed to find bugs for thunar when the > don't show up when you search on "thunar"? If you include 'ALL', the bug is shown: http://bugs.gentoo.org/buglist.cgi?quicksearch=ALL+thunar Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [slightly OT] kernel .config help pages
On Tue, Mar 20, 2007 at 06:23:40PM +, Mick wrote: > Hi All, > > Is there a quick way to access the help page of a singular module in the > kernel .config list, without having to fire up make menuconfig? > > Something on the CLI to access just one particular help page. I dont think there is a dedicated CLI tools for this, but the Kconfig files in the kernel directory contains the help text. So if you know the name of the config option is question (or its placement in the directory hierarcy) you can grep -r for it and read/grep/parse the Kconfig file in question. Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT - mbox to maildir conversion
On Mon, Oct 09, 2006 at 04:21:12AM -0500, Michael Sullivan wrote: > I want to reconfigure my SMTP server from mbox to maildir. The problem > is that I have hundreds of messages saved in imap folders in the mbox > style. Is there a way to convert those folders to maildir style? This one worked fine for me: http://untroubled.org/mbox2maildir. Backups recommended. Rasmus -- gentoo-user@gentoo.org mailing list
Re: Re: [gentoo-user] procmail, formail and maildir for digests...
On Wed, Oct 11, 2006 at 11:32:08AM +0100, Steve [Gentoo] wrote: > Is formail the right tool for me to use here? Is there a tool to > deliver an mbox of messages to a maildir that I can use in place of '>> > gentoo_user/' above? To googling for mbox2maildir or mb2md(?). With a bit of work I think they could do it. Rasmus -- gentoo-user@gentoo.org mailing list
[gentoo-user] [uOT] Thunderbird, Mozilla and HELO
Hi list, I have a slightly offtopic question that I hope I can get help with here. I have a home server, running an MTA for my domain. As of the last month or so, I have experienced a huge increase in spam and spam bounces. To combat this, I have upped my MTA's pickyness quite a bit but would like to up it more. Specifically, I would like to reject mail where sender says HELO jaquet.dk and/or where the Received line looks like [EMAIL PROTECTED] (my MTA's Received stamps are of the form [EMAIL PROTECTED]). Spammers like to use these to 'fake' their way through. This brings me to the point, because it seems like mozilla and thunderbird (my GF uses one, I use the other) grabs the domain information they use in their SMTP exchanges from the sender email address. So, when my GF sends from work via their MTAs to my home server as [EMAIL PROTECTED], that mail will run afoul of this rule since the first Received will look like Received: from jaquet.dk () by mail.XXX.dk (Postfix) with ESMTP id 75FE532CD4C etc even though her laptop at the point of sending isnt connected to my network. So, the question. Do anyone know how to fix this? I found a bugzilla entry for this, https://bugzilla.mozilla.org/show_bug.cgi?id=68877, with a final comment that this have been committed to tbird 0.6 and mozilla 1.7 in 2005. I am running tbird 1.5.0.7 and still sees this. Any clues? Thanks, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Running "hddtemp" with plain user rights
On Sat, Jan 21, 2006 at 04:06:44PM +0100, Alexander Skwar wrote: > > Does the group have the > > right to execute hddtemp? > > Yep: > > [EMAIL PROTECTED] ~ $ ls -la `which hddtemp` > -rwxr-xr-x 1 root root 27712 21. Jan 08:02 /usr/sbin/hddtemp > > BTW: sudo is also not what I'm after :) I'm after the > reason, why I get that "Permission denied" message > when I've got the necessary rights on the device file. I get this too. stracing the hddtemp process shows that it tries to perform some ioctls on the device and get EACCESS back, probably from some uid check in the kernel. The check itself can probably be found but I dont have the time for that now. Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Installed vs. Running (new feature suggestion)
On Thu, Feb 02, 2006 at 04:37:53AM -0500, Robin wrote: > bash: lsof: command not found. > > I must be missing something Have you emerged sys-process/lsof? The binary is in /usr/sbin/, so it might not be in your path. Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Tuning SCSI disks-App like hdparm available?
On Mon, Mar 13, 2006 at 06:24:38PM +0800, Ow Mun Heng wrote: > I'm wondering if there are applications which can be used to tune the > disks for better performance? There is a hdparm for SCSI, sdparm. Never tried it though. Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] dd if=/dev/dvd of=backup.iso
On Thu, Mar 16, 2006 at 11:49:00AM +0100, Alex Schuster wrote: > Or use dvdisaster (free, get it from http://www.dvdisaster.com/). It > is especially useful for bad disks, and doesn't stop when encountering > an error. I havent followed this thread closely but have the 'conv=noerror' option to dd been tried? Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] binary comparison
On Mon, Oct 17, 2005 at 09:01:35PM +0930, Shawn Haggett wrote: > >'diff' is text oriented tool. I there some kind of such tool > >oriented to binary files/subtrees comparison? > > > >Thanks! > > > > > 'od | diff' 'cmp' Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] "top" for disk access
On Fri, Oct 28, 2005 at 05:16:54PM +0930, Iain Buchanan wrote: > > > is there a top-like utility that can tell me what processes are doing > > > the most disk reads/writes? > > > > run 'vmstat 1' > > that doesn't seem to tell me what processes are running, only what the > system totals are. I wanted to see a top like output, eg > process1 xMb/s > process2 yMb/s > etc I dont think you can get that out-of-the-box. I have not played with accounting stuff (sar etc) so that might be able to do something like this for you but AFAIK then vanilla kernel wont try to track this. The reason is that short-lived processes might die before their disk request gets to the block layer, making accounting impossible. A bit more farfetched you might picture PID wrap and get disk access accounted to the wrong process. I am not an expert on this so I am happily corrected. For disk IO I find iostat to be better than vmstat. Its part of the sysstat package. Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT - Linewrap in vim
On Fri, Mar 24, 2006 at 02:19:19PM -, Michael Kintzios wrote: > While on this topic, when I cut & paste in Vim it automatically inserts > indents on the front of each pasted line which messes up my config > files. I had it fixed some time ago and now I noticed it's back - would > you know how I can switch it off again? I use :set paste, , :set nopaste. Let me know if there is a better way. Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] what does NR stands for?
On Wed, May 10, 2006 at 05:18:22PM +0800, fei huang wrote: > anybody who has some ideas about this prefix of numerous macros within linux > source code? > NR_TASKS for instance, I just could not find any explanation. number Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] glsa-check script
On Thu, May 11, 2006 at 04:35:46PM -0400, Nick Smith wrote: > has anyone written a script that checks for glsa security updates and > then has it auto emerge the packages it needs to get rid of the > security risk? if so i would be very interested in looking at that > script. Keeping 5 gentoo machines up to date security wise is > becoming very time consuming. There is glsa-check from gentoolkit. I never used it, though. Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Trying to do a world update! (Blocked by... )
On Thu, May 11, 2006 at 04:58:30PM -0400, Christopher E wrote: > Hello All, > > I am geting this at the top of my world update, here is the command > and the very beggining of the out put: > Mr-Eyes ~ # emerge -aqDu --newuse world > [blocks B ] sys-apps/coldplug (is blocking sys-fs/udev-090) > [blocks B ] sys-apps/pam-login (is blocking sys-apps/shadow-4.0.15-r1) > [blocks B ] dev-java/ant-core-1.6.5-r2) > [ebuild U ] sys-devel/patch-2.5.9-r1 [2.5.9] > > What should I do any help is greatful! Generally, you should just unmerge the blocking packages (emerge -C). Rasmus -- gentoo-user@gentoo.org mailing list
Re: [OT] Was: [gentoo-user] Finding packages which provide a file
On Mon, May 22, 2006 at 01:21:31PM -0700, Ryan Tandy wrote: > Nothing to do with the original topic, BUT... > > I'd be very interested to know how a flag called GAPING_SECURITY_HOLE > gets into one's USE ;) I'm assuming you didn't put it there yourself! Thats only used by netcat, iirc, and enables the running of commands from netcat. I guess it could be useful for testing but it is rightly named something hair-raising :) Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] revdep-rebuild failure
On Mon, May 29, 2006 at 09:22:11AM -0700, Mark Knecht wrote: > >if revdep-rebuild complains that some of the stuff is no longer in portage, > >how to find out which one? emaint, I think. Rasmus -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] testing SATA drives
On Fri, Aug 04, 2006 at 09:36:16AM -0400, Timothy A. Holmes wrote: > if it is a physical or electrical problem with the disks. The disks are > SATA drives 2x 250gb -- I am not sure how to proceed and I googling has > not been helpful -- it may be there, I just haven't found it. Use a livecd/knoppix thing and run badblocks(8) on them. Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
Re: Stefans posts (was: [gentoo-user] Re: looking for a terminal w/ url activation capability)
On Fri, Aug 18, 2006 at 10:11:44AM +0200, Alexander Skwar wrote: > Uwe Thiem schrieb: > >On 17 August 2006 20:04, Stefan Wimmer wrote: > > > >How do you always avoid my filter that stores emails coming from this list > >into my folder "gentoo-user". You are the only one. > > On my system, he's "avoiding" the filters, as his messages don't have > a *HEADER* List-Id. And it's this header, on which I filter. Maybe you > do the same? FWIW, I see the same as Alexander. Cheers, Rasmus -- gentoo-user@gentoo.org mailing list
