Re: [gentoo-user] free -m under x86_64
Looks fine. Looks like you see nearly all of the 4GB of ram (3982104). You probably have a lot of memory reserved in BIOS shadows or other hardware (like video card). This looks perfectly normal. If you have an onboard video card, a chunk of your memory is probably being used for that. In the BIOS you can usually set how much memory should be reserved for your video card. On board sound card can do the same. On Nov 19, 2007 2:06 AM, Stefan G. Weichinger [EMAIL PROTECTED] wrote: Stefan G. Weichinger schrieb: Bryan Whitehead schrieb: Can you just run uname -a and cut/paste that to an email and send to us? I think you are still in 32bit land. # uname -a Linux hiro 2.6.22-gentoo-r9 #4 SMP PREEMPT Sat Nov 17 02:42:03 CET 2007 x86_64 Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz GenuineIntel GNU/Linux And additionally: # free total used free sharedbuffers cached Mem: 398210410446602937444 0 21200 548428 -/+ buffers/cache: 4750323507072 Swap: 1052216 01052216 Looks OK to me already. Stefan -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] memtest86+ taking too long
memtest86 will keep retesting your memory over and over. Look at the Pass and Errors column. If Errors is 0 and Pass is more than 1, your memory is fine. If Pass is 0 and Errors is 0, then it is still on the first run. It shouldn't take this long. In that case make sure you have the latest version. Example: http://www.memtest.org/pics/i875-big.gif On Nov 19, 2007 12:47 PM, de Almeida, Valmor F. [EMAIL PROTECTED] wrote: Wow. In fact I don't have a clue when it is going to end. There are 2 fields with varying percentage values; but they get cleared from time to time. Another field is called Test and it was #3 (I think), and now it is at #4. So far it has taken 20:35h. On the upside, no errors detected. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] free -m under x86_64
Can you just run uname -a and cut/paste that to an email and send to us? I think you are still in 32bit land. On Nov 16, 2007 2:17 PM, Stefan G. Weichinger [EMAIL PROTECTED] wrote: Greets, following that Core2Duo-thread from a few days ago I now set up a new installation of my current 32bit-x86-setup on a second 64bit machine. I started a fresh install with the amd64 cd, used CHOST=x86_64-pc-linux-gnu and emerged everything fresh from the world file on the 32bit machine. I built the kernel for a generic 64bit CPU (just to make sure that it will be easy to move that partition to the target system, I will have to move from a Pentium D950 to a Core2Duo E6600). #zgrep 64 /proc/config.gz CONFIG_X86_64=y CONFIG_64BIT=y [...] Now I wonder why free -m still shows only 3.2 GB of RAM when I have 4 gigs in the box ... Do I have to set/remove some specific kernel-flag in /usr/src/linux/.config ? Do I misunderstand something? Thanks for any pointer, Stefan -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Re: problems with clipboard separation
X11 programs have a second way of copying and pasting text, so the first method is not a hack (sorry), however, many X11 applications do not bother with the first method. For example, xterm doesn't have an edit, copy, or paste on all flavors of unix - try using them in dtterm on Solaris and you'll see how useless the first method is when you can't cut/paste consistently between different programs (cut/copy some text, then try to paste it into gnome/kde/gtk/qt applications). xchat is typical software that doesn't do the edit menu. http://xchat.org/faq/#q24 (nor does it provide keyboard mapping for cut/copy/paste - your WM or OS must do that). The standards doc might be anal about what is first and second, but in the real world the second way is what seems to be universal. On Nov 15, 2007 10:16 PM, Miernik [EMAIL PROTECTED] wrote: Bryan Whitehead [EMAIL PROTECTED] wrote: This is the default behavior of X. Highlighting IS copying to the clipboard. Also, middle-click (or whatever is mapped to your 3rd mouse button) is paste. This is just how X works. Getting around this is a hack in itself. No, read this: http://www.jwz.org/doc/x-cut-and-paste.html -- Miernik http://miernik.name/ -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] problems with clipboard separation
It boils down to underneath everything is Xlib and the guts of X. The guts of X have many ways to do the same thing and the result is QT, GTK, KDE, GNOME, etc all end up messing with a different piece of how X should handle cut/paste. As another post points out - there seems to be 2 different ways of doing cut/copy/paste. It seems that many like the second way - or just completely ignore changing how the second way operates. I would gamble you might have a couple of applications that specifically change how each of the cut/paste methods work because the programmer is a nazi and thinks only one way is good (for example maybe he hates using the primary way so he makes the second way fudge the first way). As an example, it looks like the Firefox/Mozilla people want to enforce the first way. So people like me (who love highlighting and clicking) get pissed off because the behavior is changed in JUST firefox. It doesn't feel consistent. I can also see why someone would be a nazi (like the realplayer people are) for the second way. Do you know how annoying it is to highlight something in a dtterm in Solaris and then you can't paste it into something else? For example, I highlight in firefox, but it doesn't paste in my xterm... and my xterm doesn't seem to like any combination of Ctrl-V, etc The thing is, mozilla/firefox are GTK apps. It could be that GTK just doesn't give a crap about cut/paste which leaves the programmer to do whatever he/she wants. Which resultes in randome GTK applications acting different. Yes, it is a bunch of BS and it is why many people just stick to KDE or GNOME... rejecting other software. However, it is this fully tweakable aspect of X11 that gives us the many different windows managers, applications, headaches, and flaming mailing lists. On Nov 16, 2007 8:12 AM, Crayon Shin Chan [EMAIL PROTECTED] wrote: On Friday 16 November 2007, Bryan Whitehead wrote: This is the default behavior of X. Highlighting IS copying to the clipboard. My point is that text which I did not *specifically* highlighted should never be placed in the clipboard (whether primary/secondary/whatever). Real life example: 1) in firefox/mozilla using CTRL-L will highlight the address url so you can quickly replace it with something else, you can also use CTRL-V to paste in something off the clipboard because firefox/mozilla does not affect the clipboard when the address url is highlighted. 2) in Realplayer using CTRL-L will bring up a dialog where you can type in a url, the current url is displayed in the dialog and is already highlighted. However realplayer has also overwritten the clipboard with the current url, which in 99.% of cases is NOT what a user wants, because now I cannot paste in a new url without having to first delete the current url, then go back and copy the new url and finally paste it into realplayer. This is just how X works. Getting around this is a hack in itself. But how is it that all KDE programs have hacked it so that it behaves correctly (IMO), whereas some gtk based programs like realplayer are just so clumsy (to put it charitably). -- Crayon -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] problems with clipboard separation
This is the default behavior of X. Highlighting IS copying to the clipboard. Also, middle-click (or whatever is mapped to your 3rd mouse button) is paste. This is just how X works. Getting around this is a hack in itself. Next time you are on an Solaris or AIX workstation - know that cut/paste is the same (as X intended): highlight and 3rd button click. :) On Nov 15, 2007 8:28 PM, Crayon Shin Chan [EMAIL PROTECTED] wrote: On Friday 16 November 2007, [EMAIL PROTECTED] wrote: If so, then it seems that for me mouse-selection and Ctrl-c write into the same buffer. Can anyone give me a hint, where to look for the possibility to change this behaviour? I use Klipper and have it configured so that both clipboard buffers are synced. Normally this works fine. However some GTK based programs *always* puts whatever is highlighted onto the clipboard - it doesn't matter *how* it was highlighted - ie whether I specifically mouse dragged, or shift cursor, or even when the program itself highlighted it (eg usually when you TAB within a dialog the text in a text input is automatically highlighted). It is this last behaviour which is the most annoying - if I didn't specifically highlighted then I don't want it on the clipboard, but gtk based programs thinks otherwise. Another reason why I hate gtk and gnome :) -- Crayon -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] OT - Need help enabling iptables support in kernel
I'm pretty sure those changes are from the kernel devs - you would need to ask the lkml people. if it is from the gentoo guys, I find it less annoying than the default editor being nano instead of vi... :) On Nov 13, 2007 11:21 PM, Walter Dnes [EMAIL PROTECTED] wrote: I've been running Gentoo for a few years, and I remember earlier versions of iptables, where everything was on one page. Why do we have to activate the same feature on two separate pages now? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] OT - Need help enabling iptables support in kernel
I don't see what the big deal is - you are choosing to do everything manually by running gentoo and compiling your own kernel. If you don't like having to learn things like this why not use Ubuntu or Fedora? On Nov 12, 2007 8:35 PM, Walter Dnes [EMAIL PROTECTED] wrote: On Sat, Nov 10, 2007 at 10:53:52AM -0600, Michael Sullivan wrote On Sat, 2007-11-10 at 15:40 +0200, Daniel Iliev wrote: I believe your problem comes from: # CONFIG_IP_NF_CONNTRACK_SUPPORT is not set Build this module and try again. This option isn't even available in my config. Should I add it? Will it work with the kernel I'm running (2.6.22-hardened-r8) I'm beginning to long for the good ole days of ipchains. Is it still maintained? iptables has been scattered all over hell's-half-acre, and you need to run around enabling things all over the place to make it work. Here are some things enabled in my setup via make menuconfig. Note that this is just for filtering out the bad guys. I do not do any masq/nat/mangling/etc with iptables. *IMPORTANT NOTE* you *MUST* enable the item... IPv4 connection tracking support (required for NAT) in order for state matching to work. I found this out the hard way. Networking --- [*] Networking support Networking options --- [*] Network packet filtering framework (Netfilter) --- Core Netfilter Configuration --- * Netfilter connection tracking support --- Netfilter Xtables support (required for ip_tables) * CLASSIFY target support * MARK target support * NFQUEUE target Support NFLOG target support TCPMSS target support * comment match support connbytes per-connection counter match support connmark connection mark match support conntrack connection tracking match support * DCCP protocol match support DSCP match support ESP match support helper match support * length match support * limit match support * mac address match support * mark match support * Multiple port match support * pkttype packet type match support quota match support * realm match support * sctp protocol match support (EXPERIMENTAL) * state match support statistic match support * string match support IP: Netfilter Configuration --- * IPv4 connection tracking support (required for NAT) [*] proc/sysctl compatibility with old connection tracking IP Userspace queueing via NETLINK (OBSOLETE) * IP tables support (required for filtering/masq/NAT) * IP range match support * TOS match support * recent match support ECN match support AH match support * TTL match support * Owner match support * address type match support * Packet filtering * REJECT target support * LOG target support ULOG target support Full NAT Packet mangling raw table support (required for NOTRACK/TRACE) ARP tables support -- Walter Dnes [EMAIL PROTECTED] In linux /sbin/init is Job #1 Q. Mr. Ghandi, what do you think of Microsoft security? A. I think it would be a good idea. -- [EMAIL PROTECTED] mailing list -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Excessively simple flatfile database
Sounds like you want to use a spreadsheet. You try OpenOffice? On Nov 10, 2007 6:55 PM, Alan E. Davis [EMAIL PROTECTED] wrote: Maybe I'm barking up the wrong tree. I want to have a record of my specimens and preparations, in a format that can be retrieved in various ways, sorted, and printed. A record, like a card file. Nothing compilcated, don't need a server. I'm thinking emacs data base. I guess I want database functionality without the complexities. Alan On Nov 11, 2007 11:50 AM, [EMAIL PROTECTED] wrote: On Sun, Nov 11, 2007 at 11:23:42AM +1000, Alan E. Davis wrote: May I elicit suggestions? Do you want simple key=value pair (perl hash, associative array, etc) where one key gives one value? Your value may of course be tab separted sub values or anythig really. They are hard to search, but if you know the keys, always or often enough, they are simple. I don't know any package names, but Berkeley DB is one free source product. If you need to search on anything very often, key or value, they probably aren't it. -- ... _._. ._ ._. . _._. ._. ___ .__ ._. . .__. ._ .. ._. Felix Finch: scarecrow repairman rocket surgeon / [EMAIL PROTECTED] GPG = E987 4493 C860 246C 3B1E 6477 7838 76E9 182E 8151 ITAR license #4933 I've found a solution to Fermat's Last Theorem but I see I've run out of room o -- [EMAIL PROTECTED] mailing list -- Alan Davis, Kagman High School, Saipan [EMAIL PROTECTED] It's never a matter of liking or disliking ... ---Santa Ynez Chumash Medicine Man -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Multiple Routes to the same Subnet
Google bonding linux. Basically at the ethernet level you make eth[0-3] = bond0. You'll then have the bandwidth of all the nics as one nic. Your switch might need some extra setup - but this is the best way to go. On 11/8/07, kashani [EMAIL PROTECTED] wrote: Dan Farrell wrote: Thanks for your responses, all. On Wed, 07 Nov 2007 10:30:22 -0800 kashani [EMAIL PROTECTED] wrote: First off don't assign separate IPs to each port on your four port card, bond them into a single interface. That will simplify your config and perform better. Perhaps I will; that's not a bad idea. However, I will still have another interface that is to handle non-NFS traffic. (The reason I split it this way, by the way, is that NFS is the only network service that might potentially be limited by bandwith. Second, what sort of routing are you doing? If all the clients are on the same subnet as the four port card you should not need routing. Additionally if they are on the same subnet you should not be limited by the speed of your gateway which may or may not be able to route at 4 Gb/s whereas your switch may actually have that sort of performaance. Are the clients on a separate subnet and if so can you put them on the same subnet? No, they're all on the same subnet. Each of the 5 interfaces adds a route to that subnet (no gateway, as you said, it's the same broadcast domain) but the routes all have different metrics. The first such route chosen is the one that gets all the traffic. The NFS server is used primarily for Read access, so this routing problem does a pretty good job mitigating any benefit of having so many interfaces. Oh, by the way, this is 100-T, not Gigabit. Do I sound rich to you : ) ? Buying a single GigE card would appear to be simpler and cheaper unless you don't have a GigE switch. :-) So, let's say I bond the 4 together. Now I have 2 interfaces, a bond and eth0. I still need to route through one or the other, so I still have the problem. I am reading about policy routing, which should be able to solve the problem by allowing routing based on the source rather than the destination. I will keep the lists informed... You should not need to do any routing and I'd be surprised if Linux is actually doing any routing in this case. However depending on how you are testing you might see some issues. Let's assume you've got this network. server eth0 10.11.12.21/24 server eth1 10.11.12.22/24 client1 eth0 10.11.12.101/24 client2 eth0 10.11.12.102/24 The server will have all sorts of nonsense about 10.11.12.21 255.255.255.255 routes and you can ignore all that. Additionally when you initiate a connection from your server it will always originate from eth0 because 0 comes before 1 IIRC. Just one of those things. However when you initiate a connection from a client to eth1 the server should respond out the same interface. I'd play around with tcpdump on a client and see if this is happening like it should be. You might also try forcing portmap to bind to one IP in /etc/conf.d/portmap. If for some reason I'm completely off base and Linux is defaulting out eth0 for connections coming into eth1 you can always do the lo tech solution. Assuming the above network we then assign a separate subnet to eth1 and an alias to each client. server eth0 10.11.12.21/24 server eth1 10.11.88.21/24 client1 eth0 10.11.12.101/24 client1 eth0:0 10.11.88.101/24 client2 eth0 10.11.12.102/24 client2 eth0:0 10.11.88.102/24 The machines connect on 10.11.88.0/24 and you avoid any interface confusion. kashani -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] emerge update and time skell
Wait 2 or more hours, reboot for good measure, and try again. On 11/7/07, Mateus Interciso [EMAIL PROTECTED] wrote: Hello, I've recently made a change on my gentoo boxes on the time, which was 2 hours ahead, now when I'm trying to make some emerge --update, I always get a LOT of date mismatches, and some emerges are caught on a loop, how can I fix this? I've already made a new emerge --sync, and it didn't fixed. Also, when I try to start a service, like for instance mrtg, by using / etc/init.d/mrtg start, it complains about some files either on /etc/ conf.d or /etc/init.d having a future timestamp, but the timestamp are months or days behind... Thanks a lot. Mateus -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] OT - Why won't my computers use their own DNS servers? [SOLVED]
Why not do this: nameserver 127.0.0.1 Michael Sullivan wrote: On Mon, 2007-05-14 at 16:58 +0200, Etaoin Shrdlu wrote: On Monday 14 May 2007 16:18, Michael Sullivan wrote: On camille: camille ~ # cat /etc/resolv.conf search espersunited.com nameserver 70.24.122.250 Shouldn't this be 70.234.122.250? That did it. Thank you! I wonder how that happened, and why I didn't see that the addresses didn't line up? -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] RAID-0 with LVM - is there any point?
I think you need to try running a real benchmark like bonnie++ against both. For example, you run time dd but you don't include the sync in the time... Daniel Iliev wrote: Neil Bothwick wrote: Hello Daniel Iliev, Actually I'd be glad to read some results from a Fake RAID-0 vs LVM tests. My bet would be that RAID-0 w/o LVM would give the best speeds Omitting LVM isn't an option, I'd lose all the flexibility that LVM offers. I don't see why RAID-0 should be necessarily more efficient than LVM, unless there's something superior about RAID-0's striping algorithms. I could do some before and after tests, but I'd first have the reformat the filesystems to remove any effects of fragmentation. If no one comes up with a good reason for keeping the RAID, I'll get rid of it, running bonnie++ before and after. Hi, Neil! Out of curiosity I made some tests which confirmed my expectations. What about you - did you have time (and wish) to take some performance benchmarks? I would be glad to see some additional results. I'm attaching my tests in file called bench.txt. echo y | mdadm -C /dev/md9 -n2 /dev/sda11 /dev/sdb11 -l0 mkfs.xfs /dev/md9 mkdir /test mount /dev/md9 /test dd if=/dev/urandom of=/test.rnd bs=1M count=1500 time cp /test.rnd /test real0m44.981s user0m0.036s sys 0m6.967s sync time mv /test.rnd /test real0m47.514s user0m0.047s sys 0m7.077s sync time mv /test/test.rnd / real0m53.863s user0m0.060s sys 0m8.885s mdadm --stop /dev/md9 pvcreate /dev/sda11 pvcreate /dev/sdb11 vgcreate test /dev/sda11 vgextend test /dev/sdb11 vgdisplay | grep 'Total PE' Total PE 1686 lvcreate -i2 -l1686 -nlogvol test mkfs.xfs /dev/test/logvol mount /dev/test/logvol /test time cp /test.rnd /test real1m12.183s user0m0.039s sys 0m9.570s sync time mv /test.rnd /test real0m51.643s user0m0.044s sys 0m7.275s sync time mv /test/test.rnd / real1m54.937s user0m0.047s sys 0m9.556s = BOTTOM LINE: cp /test.rnd /test LVM:20.78 [MB/s] RAID-0: 33.41 [MB/s] mv /test.rnd /test LVM:29.04[MB/s] RAID-0: 31.56[MB/s] mv /test/test.rnd / LVM:11.11[MB/s] RAID-0: 27.84[MB/s] Strange: I repeated the last LVM test because it seemed to me as a low performance peak, but the result was again very low: time mv /test/test.rnd / real1m27.775s user0m0.050s sys 0m9.813s which is: 1500/87.775 = 17.089 [MB/s] -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Why are gentoo people so in love with colorized output?!?
Go back to using Solaris ya old fart! ;) [EMAIL PROTECTED] wrote: On Wed, Apr 04, 2007 at 07:36:49AM +0200, Bo ?rsted Andresen wrote: Weren't you talking about portage? In that case you should obviously file it against portage.. But yeah, any app that has a --nocolor equivalent that doesn't work deserves a bug report.. Even for apps that don't it's reasonable to file it as an enhancement request. Oh for pete's sake, don't be so literal. Esearch has screwed up. Emerge has screwed up. Revdep-rebuild has screwed up. Stop reading the leaves on the trees and paya ttention to the forest. Your quibbly attitude is exactly the petulant behavior which makes me not want to waste my time filing bug reports on somebody's pet eye candy. First of all I believe most people (including myself) very much prefer colors over no colors (no I cannot qualify with any numbers..). That does not, however, mean that the pipe detection and --color switch etc. shouldn't be honoured. It should (and it does here). Secondly, how did you come up with the idea that a bug report would be dismissed if you never filed one? The UNIX standard for ages has been simple text output. Why must gentoo add trendy colors which change every time some eye candy fanatic gets a bug up his butt to change colors when he gets bored with the old fashioned colors? the default ought to be colors OFF and you have to ask to get them. I choose fonts small enough to get maximum density with minimum eye strain. The only way I could read these colors would be to increase the font size and decrease the density. If gentoo developers think that a wise trade off when almost no other utility uses colors so much and so horribly, then gentoo is broken by design and no amount of bug reportage will change a damned thing. Harmony is a nice design feature. You ought to try it sometime. As long as I am ranting, I may as well throw in a few rants on the amateur kids who run gentoo; those who think the world should be thankful for their color choices are the same idiots who linked ls against a /usr/lib library and made my system ubootable, who removed libraries which LVM linked against during boot and made my system unbootable. Gentoo has good points, starting with portage, but it also has innumerable insufferable knowitalls who make me gnash my teeth at their inconsiderate unthinking fad-of-the-week behavior. -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] OT im more just curious
Nearly 28. Been using gentoo since version 1.0 (maybe pre-1.0 but can't remember). I started using linux back when slackware fit on a bunch of 5.25 floppies. I now work full time at a startup in the silicon valley watching over 4 datacenters full of CentOS machines (and some Solaris). [EMAIL PROTECTED] wrote: What is the average age of the gentoo user here? Sent via BlackBerry® from Vodafone �éí¢‹¬z¸ žÚ(¢¸j)bž bst== -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] lilo and SW-RAID-boot-partition
This is what I have for sw-raid for lilo: disk=/dev/hda bios=0x80 disk=/dev/hdc bios=0x81 menu-scheme=Wb boot = /dev/md0 prompt map = /boot/System.map raid-extra-boot=/dev/hda,/dev/hdc lba32 timeout=150 delay = 50 default = 2.6.14 serial = 0,9600n8 vga = normal# Normal VGA console image = /boot/vmlinuz-2.6.14-gentoo append= console=ttyS0,9600 console=tty0 root = /dev/md1 label = 2.6.14 read-only # read-only for checking My /etc/fstab: /dev/md0/boot ext2noatime 1 1 /dev/md1/ ext3noatime 0 0 /dev/md3noneswap sw 0 0 /dev/md2/tmpext3noatime 0 0 /dev/md4/varext3noatime 0 0 /dev/md5/export/homexfs noatime 0 0 none/proc proc defaults0 0 none/dev/shmtmpfs defaults0 0 /export/home/home xfs bind0 0 Stefan G. Weichinger wrote: Mark Kirkwood schrieb: I think you might need a root=/dev/md3 inside the specification for 'Gentoo': image=/boot/kernel-genkernel-x86-2.6.19-gentoo-r5 label=Gentoo vga=0x314 initrd=/boot/initramfs-genkernel-x86-2.6.19-gentoo-r5 root=/dev/md3 Thanks for that, but it didn't work. Same error as before. S -- gentoo-user@gentoo.org mailing list
[gentoo-user] (offtopic) old RH7.1 disks?
I have some commercial software that doesn't work on any version of linux except rh7.1. However, I don't have the disks. The only mirror we have been able to find is only moving at a steady 2.5k/sec FWIW, The software is CSPS: http://www.cisco.com/univercd/cc/td/doc/product/voice/sipproxy/relnotes/stnlnxrn.htm Cisco wants ~$40k per install to upgrade to a version that runs on a modern OS. So while we write our own SIP Proxy Server... we are stuck using rh7.1. Any help would be great! Thanks! (and yes, we have looked at SER but there are a number of key features missing for us to use it) -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] BIG reiserfs problem
If you are so concerned with the awesomeness of XFS's caching... why not turn on data-journaling? Then data (not just meta-data) is committed to the journal. You can also tune XFS to not wait so long to hold cached data. Alan McKinnon wrote: On Saturday 28 October 2006 16:41, b.n. wrote: Dale ha scritto: If you use XFS, make sure you have good power. XFS does not like power failures at all. I have had to reinstall on a second rig because of this very problem. If you have a UPS, that may be OK. Thanks a lot for the advice. Power outages do happen and I don't have an UPS. Why does it happen? Isn't XFS journaled? Yes it is journaled but it also allows data to be very aggressively cached. Make that VERY aggressively cached. With the result that data can be held in a huge cache somewhere and the kernel can be convinced it has been written to disk. Consider XFS's pedigree - SGI wrote it for their graphics machines. These were big monsters backed up with high grade UPSs and such - the logic was that if you spend a brazillion bucks on hardware, a mega UPS is part of the deal, along with the wages to pay the army of admins you also need. And, when doing video rendering, it turns out that it's easier to simply re-render a frame when the filesystems does something odd with the data rather than go to the effort of writing an FS that is 100% reliable. So SGI sacrificed something that doesn't actually matter for their use case to gain a significant performace increase (which does matter a great deal) alan -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] BIG reiserfs problem
After sending this I realized that XFS doesn't support journal=data... I thought journal=data was a general VFS part of the linux kernel... my bad. :) I guess you are just left with in kernel tuning (someone previously posted a link to). Bryan Whitehead wrote: If you are so concerned with the awesomeness of XFS's caching... why not turn on data-journaling? Then data (not just meta-data) is committed to the journal. You can also tune XFS to not wait so long to hold cached data. Alan McKinnon wrote: On Saturday 28 October 2006 16:41, b.n. wrote: Dale ha scritto: If you use XFS, make sure you have good power. XFS does not like power failures at all. I have had to reinstall on a second rig because of this very problem. If you have a UPS, that may be OK. Thanks a lot for the advice. Power outages do happen and I don't have an UPS. Why does it happen? Isn't XFS journaled? Yes it is journaled but it also allows data to be very aggressively cached. Make that VERY aggressively cached. With the result that data can be held in a huge cache somewhere and the kernel can be convinced it has been written to disk. Consider XFS's pedigree - SGI wrote it for their graphics machines. These were big monsters backed up with high grade UPSs and such - the logic was that if you spend a brazillion bucks on hardware, a mega UPS is part of the deal, along with the wages to pay the army of admins you also need. And, when doing video rendering, it turns out that it's easier to simply re-render a frame when the filesystems does something odd with the data rather than go to the effort of writing an FS that is 100% reliable. So SGI sacrificed something that doesn't actually matter for their use case to gain a significant performace increase (which does matter a great deal) alan -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] cyrus-sasl error.
Can you do a emerge -upv dev-libs/cyrus-sasl and post? I'd like to see your use flags. Leandro Melo de Sales wrote: Hi, I'm trying to update openldap and one of package dependency is cyrus-sasl, but when emerge compile cyrus-sasl package, it shows the following message: # emerge cyrus-sasl Calculating dependencies... done! Emerging (1 of 1) dev-libs/cyrus-sasl-2.1.22 to / * cyrus-sasl-2.1.22.tar.gz MD5 ;-) ... ... ... ... x86_64-pc-linux-gnu-gcc -shared .libs/sasldb.o .libs/sasldb_init.o .libs/plugin_common.o -Wl,--whole-archive ../sasldb/.libs/libsasldb.a -Wl,--no-whole-archive /usr/lib64/libdb- 4.2.so http://4.2.so -lresolv -march=athlon64 -Wl,-soname -Wl,libsasldb.so.2 -o .libs/libsasldb.so.2.0.22 /usr/lib/gcc/x86_64-pc-linux-gnu/3.4.6/../../../../x86_64-pc-linux-gnu/bin/ld: ../sasldb/.libs/libsasldb.a(allockey.o): relocation R_X86_64_32S against `a local symbol' can not be used when making a shared object; recompile with -fPIC ../sasldb/.libs/libsasldb.a(allockey.o): could not read symbols: Bad value collect2: ld returned 1 exit status make[2]: *** [libsasldb.la http://libsasldb.la] Error 1 make[2]: Leaving directory `/var/tmp/portage/cyrus- sasl-2.1.22/work/cyrus-sasl-2.1.22/plugins' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/var/tmp/portage/cyrus-sasl-2.1.22/work/cyrus-sasl-2.1.22' make: *** [all] Error 2 !!! ERROR: dev-libs/cyrus- sasl-2.1.22 failed. Call stack: ebuild.sh, line 1546: Called dyn_compile ebuild.sh, line 937: Called src_compile cyrus-sasl-2.1.22.ebuild, line 166: Called die Any clue/suggestion will be greatfully accepted. Thank you. Leandro -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vim-in-pine problem
For me it was already having known all the pine keys... mutt isn't that much better to justify months of readjusting to a new email reader... Jorge Almeida wrote: On Sat, 21 Oct 2006, Philip Webb wrote: I don't really remember why I didn't stick with mutt, but I know there were some reasons: colors? difficult to configure? Yes, you need to read 'man muttrc' set up ~/.muttrc to your tastes. There was something else...Asking for password too many times? -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] SMTP Authentication
I use stunnel to wrap imap and pop3. For SMTP I have TLS authentication
using the saslauthd deamon and postfix. I have many virtual domains, but
all accounts are local/plain unix accounts. Most do not have shell access.
I use the following use flags
mail-mta/postfix pam sasl ssl
dev-libs/cyrus-sasl berkdb crypt ntlm_unsupported_patch pam ssl
emerge --unmerge qmail
emerge postfix
in main.cf I have:
smtpd_sasl_auth_enable = yes
smtpd_sasl_security_options = noanonymous
broken_sasl_auth_clients = yes
smtpd_use_tls = yes
smtpd_tls_auth_only = yes
smtpd_tls_key_file = /etc/stunnel/stunnel.pem
smtpd_tls_cert_file = /etc/stunnel/stunnel.pem
smtpd_tls_CAfile = /etc/stunnel/stunnel.pem
smtpd_tls_received_header = yes
smtpd_tls_session_cache_timeout = 3600s
tls_random_source = dev:/dev/urandom
(I just use the same key from stunnel for TLS... I duno if it is even
needed or not)
~ # cat /etc/sasl2/smtpd.conf
pwcheck_method: saslauthd
log_level: 3
mech_list: PLAIN LOGIN
authdaemond_path:/var/lib/sasl2/mux
~ # grep -v -- \# /etc/conf.d/saslauthd
SASLAUTHD_OPTS=${SASLAUTH_MECH} -a shadow -r
If you guys have any problems with the above... I lurk on this list if
you need help.
Tito Valentin wrote:
Hello list:
I have been trying to get SMTP authentication on Gentoo to work with no
success. I am running qmail (netqmail package) with dovecot and
saslauth.
I am able to connect and read my mail through IMAP fine. The problem is
that I cannot send email using my own SMTP server due to an
authentication
issue. Every time we try to send email out we get an error saying
that it
cannot establish a connection because I am not part of the allowed
rcphost
list. I tried adding the remote host in the rcpthosts database but the
problem persists. The only error I see in the log files for dovecot is
Disconnected
Can anyone point me in the right direction as to why I can't be
authenticated when sending email out? Another thing is, it keeps
prompting me for my passwd.
GV
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Bind (Named) SOA Records
You are missing semi-colons after each record... Also you might want to just make a dummy domain on a test dns - see if it works. On Wed, 23 Aug 2006, Michael Crute wrote: I am trying to write a scrip to maintain my Bind zone files and I was wondering if it is valid to cram the SOA all onto one line. I have scoured the internet and can't find anyone who does it that way. My question is... will it break stuff? Right now my SOA looks like: @ IN SOA ns1.domain.com. hostmaster.domain.com. ( 2006082201 ; serial number 12h ; Refresh 1h ; Retry 2w ; Expire 1h ; Min TTL ) what I am proposing is this: @ IN SOA ns1.domain.com. admin.domain.com. ( 2006082201 12h 1h 2w 1h ) Are the two equivalent? -Mike -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge problems
It is most likely a problem with your own build enviroment or bad use flags. Post the errors you are having to this group so we can see the build problems. Before you do that, try running repdev-rebuild. On Sat, 26 Aug 2006, Martin S wrote: I seem to find lots of bad repository servers when running emerge. emerge mozilla-firefox failed for quite some time (at least a day), emerge kuroo has failed for two days, just like emerge crafty. Does anyone know if there are any general problems with servers? Regards, Martin S -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] NIS configuration
If you are worried about the users getting the ldap bind password - then why on earth would you bother with NIS? NIS has no password to block binding at all! At least ldap has a binding password... On Tue, 22 Aug 2006, Leandro Melo de Sales wrote: Hi! I configured my gentoo server box to authenticate users through LDAP (nsswicth+pam_ldap). Everything are working fine and now I want to run a ypserv and as a result I want the nis clients authenticating against my server. I following the instructions in http://gentoo-wiki.com/HOWTO_Setup_NIS but the clients (gentoo and ubuntu boxes) don't find the ypserver, any clue? I was planning to make the clients authentication directly to LDAP server, but in this case I have to install pam_ldap and create a /etc/ldap.conf and put ldap bindpw into this file or in /etc/ldap.secret. Even if I protect it with 600 the use can get this file installing a windows application (the client machines have dual boot) that read ext3 partitions and read the files, finally getting the bindpw. Any comment about this? Am I thinking wrong about that? Thank you. BTW, all suggestion will be greatfully accepted. Leandro. -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] netqmail and qmail
Anyway - I stand to what I wrote. I'd suggest any MTA, *BESIDES* qmail and sendmail. qmail, as it's too buggy, too few features and too complicated. sendmail, as the configuration is a nightmare (compared to easier systems available nowadays). I object to this statement. Sendmail is a solid MTA. Calling it as bad as qmail is just going to far... I've maintained some pretty large sendmail systems without a problem. That said - I now use postfix if given a choice. -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: Change Date (was Re: [gentoo-user] Seamonkey blocking mozilla and mozilla blocking seamonkey)
Yes. please fix you date... emerge ntp man ntp :) On Sun, 16 Jul 2006, Brad Camroux wrote: Jacob, You need to change the date on your computer. It's set for December 2006, and so this thread appears before all properly-dated mail. Brad -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT SATA card recommendations
You might want take a look at your numbers again. Using 500GB SATA disks or bigger means you don't need such expensive raid cards (or multiple raid cards)... Also, you have fewer moving parts to break. On Mon, 17 Jul 2006, Mike Williams wrote: Hey, In the next month or so I wish/need to make some storage modifications, I'm running out of space, quickly. Currently it's a horrible hodge podge of 2 RAID5s in one VG, and 1 RAID10 in another VG. 1 of 3 is PATA other 2 SATA (RAID5), 5 of 5 SATA (RAID5), and 4 of 4 PATA (RAID10), 1 of 1 PATA (boot and OS). Spread over the on-board PATA controller, a PCI 2 port PATA (for the 4 drives in RAID10), and an 8 port marvell SATA card (which doesn't have a usable driver I can find after 2.6.13). That lot in a very old globalwin 302 (??) midi tower. Yes, it's *hot*. I already know I'm going for a Antec P180, so I'm limited to a maximum of 10 drives, one boot disk, and a CD (something I'm missing now). 320GB drives are the best £/GB at the moment, even if they only give 298GB usable space. 10 x 298GB / RAID6 == just under twice what I've got already, and more redundancy. So, the point in my post. I need a 12 port SATA card that works properly with open-source drivers (preferable in kernel). Cost isn't the primary concern, even if this is only for home use. I don't need, or in fact want, on-board RAID, software RAID is better in my opinion. 3x 4 ports is an option, at a push. I'd rather not lose the gigabit nic though. I see that 3ware do a 12 port card, but it's over 500 quid! Cost may not be the primary concern, but 500 quid is still 500 quid! Anyone with any experience? Ta -- Bryan Whitehead Email:[EMAIL PROTECTED]
Re: [gentoo-user] Reiserfs meltdown
On Mon, 17 Jul 2006, Ralph Slooten wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hiya list, Just need some opinions here, and am not looking for a raving flame-war regarding which file system is better etc ;-) ~ Oh and please excuse the long mail, but I need explain my situation clearly to avoid confusion. [snip] I've had this happen to me a number of times... I'm now a happy XFS user. :) /flamestart ;) -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] enable verbose-debug on iptable
What you're really looking for is how to log packets that are dropped? At the end of your chain add a rule to log. see www.netfilter.org packet filtering HOWTO. On Fri, 31 Mar 2006, El Nino wrote: Dear Group, how to enable verbose-debug on iptables? -- ... (((o)))~--~--~-- Proud to be a Sinhalese. SINHALESE ARE GENIUSES OF IRRIGATION http://easyweb.easynet.co.uk/~sydney/sinhales.htm -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: php + apache
Sounds like an apache.conf problem... Of course it could be a stray php.so isn't linked correctly, try running revdep-rebuild? On Sat, 1 Apr 2006, James wrote: Michael Sullivan michael at espersunited.com writes: Ive installed php-4.4.2 and apache2. It's mostly working except for when I pull up the url of a php file, like setup.php, I see the actual php code instead of the gui form that the setup.php is suppose to render. I know the file works as the software is also built on a debian system and the indentical setup.php file work fine on debian + apache Any ideas what's wrong? Do you have -D PHP4 on the APACHE2_OPTS line in /etc/conf.d/apache2? Yes, I've unmerged everthing the package used, and delete any remaining files. Complete fresh installation in underway. If that does not fix it, I'll post again thx, James -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ssh2 and xover LAN
What do you mean by xover? FTP? SCP? NFS? Coda? HTTP? On Sat, 1 Apr 2006, maxim wexler wrote: Hi everybody, Anybody got this working between two PCs and can tell me how to edit the conf files? The NICs are found, the modules are loaded, the net is up and both PCs can ping each other. But transferring files fails. FWIW tcpdump indicates the two PCs are aware of each other. ssh2 is probably overkill since security is not an issue but ftp didn't work either. -Maxim __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ssh2 and xover LAN
He said it could ping... I dunno... seems he just needs to figure out how he wants to transfer files. On Sat, 1 Apr 2006, Matt Richards wrote: Bryan Whitehead wrote: What do you mean by xover? FTP? SCP? NFS? Coda? HTTP? I guess by xover they mean as in 2 NIC's one cable cross wired! Matt. On Sat, 1 Apr 2006, maxim wexler wrote: Hi everybody, Anybody got this working between two PCs and can tell me how to edit the conf files? The NICs are found, the modules are loaded, the net is up and both PCs can ping each other. But transferring files fails. FWIW tcpdump indicates the two PCs are aware of each other. ssh2 is probably overkill since security is not an issue but ftp didn't work either. -Maxim __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] DNS Expert Required
You might want to set your serial to a date format like:
2006032600
When you do an update on the same day you just increment the last
digit(s).
2006032601
You are more likely to remember when you have done updates with a date.
On Sun, 26 Mar 2006, Hiren Dave wrote:
Hi Uwe,
The main problem in my configuration was this:
search server1.guru.com server2.guru.com
search guru.com
zone server1.guru.com IN {
make this: zone guru.com IN {
After changing this, my dns server is working fine. Thanks a lot dude.
Hiren
On 3/26/06, Uwe Thiem [EMAIL PROTECTED] wrote:
On 26 March 2006 10:01, Hiren Dave wrote:
Alright, I'll bite. ;-)
Some small errors.
server1.guru.com (192.168.0.2) Primary DNS Server
server2.guru.com (192.168.0.3) Secondary DNS Server
= Here is my configuration file for server1.guru.com machine.
###/etc/resolve.conf##
domain guru.com
search server1.guru.com server2.guru.com
search guru.com
(the search string gets appended to non-qualified names)
nameserver 192.168.0.2
nameserver 192.168.0.3
##
/var/named/chroot/etc/named.conf##
options {
directory /var/named;
dump-file /var/named/data/cache_dump.db;
statistics-file /var/named/data/named_stats.txt;
allow-transfer { 192.168.0.3; 192.168.0.1; };
allow-query { 192.168.0.0/24; localhost; };
allow-recursion { 192.168.0.0/24; localhost; };
};
allow-tansfer: take out 192.168.0.1; it's unnecessary.
[ snip ]
zone server1.guru.com IN {
make this: zone guru.com IN {
type master;
file server1.guru.com.zone;
make this: file guru.com; (not strictly necessary)
allow-query { any; };
In the general options, you restrict queries to your local network. Why do
you
now allow queries from anywhere?
allow-update { key rndckey; };
};
#include /etc/rndc.key;
##
###/var/named/chroot/var/named/server1.guru.com.zone##
make the file guru.co
$TTL 86400
@ IN SOA server1.guru.com. root.server1.guru.com. (
BTW, *not* a good idea to make root the technical contact for your
domain.
Now most stuff below can be much easier.
1 ; serial
300; refresh (5 minutes)
10 ; retry (10 seconds)
86400 ; expire (1 day)
600; minimum (10 minutes)
)
@ IN NS server1.guru.com.
@ IN MX 10 server1.guru.com.
server1.guru.com. IN A 192.168.0.2
server2.guru.com. IN A 192.168.0.3
win2k.guru.com. IN A 192.168.0.1
www1 CNAME server1.guru.com.
www2 CNAME server2.guru.com.
www3 CNAME win2k.guru.com.
server2 IN MX 0 server1.guru.com.
win2k IN MX 0 server1.guru.com.
The lines above can now look this way:
IN NS server1.guru.com.
IN NS server2.guru.com.
IN MX 10 server1.guru.co.
*.guru.com IN MX 10 server1.guru.com.
server1 IN A 192.168.0.2
server2 IN A 192.168.0.3
win2k IN A 192.168.0.1
www1CNAME server1
www2CNAME server2
www3CNAME win2k
(Note where I put a fullstop at the end of a name and where not. It's
important. Your MX statements above are contradictory. So I don't know
exactly which box your email server is and if you really have more than
one.
If so you have to adjust my two MX entries above.)
Adjust server2 accordingly.
I cannot guarantee that I caught all mistakes. Just try it out.
Uwe
--
Why do consumers keep buying products they will live to curse?
--
gentoo-user@gentoo.org mailing list
--
Bryan Whitehead
Email:[EMAIL PROTECTED]
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ftp connection refused
I really don't understand the problem... you can't ftp from one server to the other? if so, which server from what client? What IP to what IP? Can you show me the output of netstat -tlnp on each? On Fri, 24 Mar 2006, maxim wexler wrote: Hello everybody, I keep casting this fly, hoping for a strike ;0 For a crossover lan. Ping OK. route -n confirms net setup on *both* machines iftraf indicates activity on remote machine when ftp command issued. adding debug switch returns: servname not supported for ai_socktype. googling points to bad /etc/hosts. here's /etc/hosts from present machine(192.168.0.3): 127.0.0.1 sarawak localhost 192.168.0.2 xlan yeti #remote pc # IPV6 versions of localhost and co ::1 ip6-localhost ip6-loopback fe00::0 ip6-localnet ff00::0 ip6-mcastprefix ff02::1 ip6-allnodes ff02::2 ip6-allrouters ff02::3 ip6-allhosts I think it was after emerging ftp I found the line starting with 192.168.0.3 on the machine with the fresh 2.6.15 install followed by 'gravity.twi-31o2.org gravity'(?!), but that didn't work either. in /etc/host.conf multi set to on and off. Neither works. Is it resolv.conf? But this is 'net stuff. Dynamic name finding or whatever, isn't it? The NICs are strictly non-web using static addresses. [EMAIL PROTECTED] ~ $ cat /etc/resolv.conf domain sarawak nameserver 206.47.244.52 nameserver 67.69.184.7 -Maxim __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] DNS trouble
[EMAIL PROTECTED] ~]$ dig -x 210.200.1.1 | grep AUTHORITY\ SEC -A 1 ;; AUTHORITY SECTION: 1.200.210.in-addr.arpa. 10733 IN SOA dns.ht.net.tw. root.dns.ht.net.tw. 1999090230 10800 3600 360 86400 Are you dns.ht.net.tw. ??? If not, then you need to contact them to get the reverse fixed. You can either allow them to give you authority over your address space, or they can add the names to their 1.200.210.in-addr.arpa file on the DNS. for your internal DNS, you setup your own .in-addr.arpa files for the 192.169.0.0/16 subnets that you use. On Mon, 14 Nov 2005, Allan Spagnol Comar wrote: still the problem [EMAIL PROTECTED] ~ $ host 210.200.1.230 Host 230.1.200.210.in-addr.arpa not found: 3(NXDOMAIN) [EMAIL PROTECTED] ~ $ host 210.200.1.232 232.1.200.210.in-addr.arpa domain name pointer shakan.commodity.com.br. and seems to me that both shakan.eng.com.br and thewho.eng.com.br is configured equal at the zones; the diference is the thewho ( 210.200.1.230 ) is my DNS server. On 11/14/05, Michael Kjorling [EMAIL PROTECTED] wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 2005-11-14 12:46 -0300, [EMAIL PROTECTED] wrote: does someone have a clue about what is going on ? Yes, `nslookup' is fundamentally broken. Try `host' or `dig' (with its `-x' option) instead. And don't expect RFC 1918 address space to be resolvable via a public name server... - -- Michael Kjörling, [EMAIL PROTECTED] - http://michael.kjorling.com/ * ASCII Ribbon Campaign: Against HTML Mail, Proprietary Attachments * * . No bird soars too high if he soars with his own wings . * -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFDeLeHdY+HSb3praYRAiahAJ4qLwX2kPTKMvIyuWwDTaPcrgYM3QCfTQhW nxFNE0o+QiYj1nsnU06LvGQ= =F5Xs -END PGP SIGNATURE- -- gentoo-user@gentoo.org mailing list -- An application asked: Requeires Windows 9x, NT4 or better, so I´ve installed Linux -- Bryan Whitehead Email:[EMAIL PROTECTED]
Re: [gentoo-user] fixing fstab
I still don't understand the logic of not having vi installed by default over nano... On Mon, 14 Nov 2005, Petteri R??ty wrote: Mark wrote: I made a mistake while creating my fstab on a new install, and I can't boot. If I use my Universal CD to boot up, what command(s) will I have to run to get access to the fstab to fix it? (I'm assuming I have to re-mount chroot but I don't know specifically what to do). Thanks! You only need to mount your root partition to /mnt/gentoo and then execute nano /mnt/gentoo/etc/fstab -w. You can change nano to another editor if you like. Regards, Petteri -- Bryan Whitehead Email:[EMAIL PROTECTED]
[gentoo-user] 14TB filesystem problems...
I need a big partition, and have a 14TB raid5 (with hot spares) I get this problem: [EMAIL PROTECTED] ~]# fdisk /dev/sdc The number of cylinders for this disk is set to 437698. There is nothing wrong with that, but this is larger than 1024, and could in certain setups cause problems with: 1) software that runs at boot time (e.g., old versions of LILO) 2) booting and partitioning software from other OSs (e.g., DOS FDISK, OS/2 FDISK) Command (m for help): n Command action e extended p primary partition (1-4) p Partition number (1-4): 1 First cylinder (1-170349, default 1): Using default value 1 Last cylinder or +size or +sizeM or +sizeK (1-170349, default 170349): 437698 Value out of range. Last cylinder or +size or +sizeM or +sizeK (1-170349, default 170349): -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ghosting(?) a drive
The best way would to be to create a new filesystem on the new disk. Then copy the data via rsync. startfsflamewar had you used a real fs, like xfs, you could do a dump/restore to generate an exact copy at the filesystem level (full acl, and other details embedded into a filesystem) /startfsflamewar On Fri, 11 Nov 2005, maxim wexler wrote: Hello everyone, Just received a new, unformatted SATA 120G HD with the intention of moving my entire gentoo OS over to it from a flaky 120G ATA drive(reiserfs). Hopefully, I can just boot up from the new drive as if nothing had changed. Can anybody recommend any tool(s) for the job? Gotchas? Does SATA prefer a certain fs? -mw __ Yahoo! FareChase: Search multiple travel sites in one click. http://farechase.yahoo.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Simple and lightweight SMTP server
Postfix. On Mon, 7 Nov 2005, Alexander Skwar wrote: Hi. I'm looking for a very lightweight SMTP server. It should (in order): - Support TLS or SSL for outgoing/incoming connections - Be VERY light on CPU usage - Store mails in Maildir format (or hand it off to an MDA like procmail/maildrop, for them to store the mails) - Relay mail to only one smart host It is important, that the system isn't too heavy on the CPU, as I've only got a MIPSel MIPS 4Kc V0.10. What can you recommend? On normal systems, I always use Postfix, but that's not really lightweight :) Thanks, Alexander Skwar -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 14TB filesystem problems...
So if I use a non-traditional partition... then I should be good? On Fri, 11 Nov 2005, Richard Fish wrote: On 11/11/05, kashani [EMAIL PROTECTED] wrote: Bryan Whitehead wrote: I need a big partition, and have a 14TB raid5 (with hot spares) I get this problem: [EMAIL PROTECTED] ~]# fdisk /dev/sdc fdisk is known to have issues with large volumes. I don't recall offhand what the best tool is, but I believe partd and cfdisk would work better than fdisk. However they may some quirks of their own with that much space. The largest partition possible in a traditional PC partition table on any disk with 512-byte sectors is 2TB. -Richard kashani -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] top for disk access
run 'vmstat 1' On Fri, 28 Oct 2005, Iain Buchanan wrote: Hi all, is there a top-like utility that can tell me what processes are doing the most disk reads/writes? Sometimes my disk will go crazy, and my system slow down. I usually know its updatedb, or something similar, but sometimes I have no idea why. Any ideas? thanks, -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] having a real mta + virtual one (ie postfix nbsmtp)
I don't understand why you would need to do this? If you need a MTA and a MUA, why would you bother with a MUA? All of what a MUA can do is in the MTA - postfix On Wed, 26 Oct 2005, Covington, Chris wrote: Hi all, I'd like to have both nbsmtp and postfix, so that I can run an MTA and use an MUA independently of each other. Although the two can co-exist peacefully, the combo seems to upset portage: videodrome ccovington # emerge -pvDu world These are the packages that I would merge, in order: Calculating world dependencies ...done! [blocks B ] mail-mta/nbsmtp (is blocking mail-mta/postfix-2.2.5) Is there something I can do? --- Chris Covington IT Plus One Health Management 75 Maiden Lane Suite 801 NY, NY 10038 646-312-6269 http://www.plusoneactive.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OpenOffice 2.0 + NFS = hang
Make sure all nfs services are running on client and server. OR you have iptables running and you are blocking random nfs service ports. Check both client and server. run rpcinfo -p on each machine to see what ports need to be open. It might be better to just allow anything to go between server-client and client-server. On Thu, 27 Oct 2005, Konstantin V. Gavrilenko wrote: As discussed in this thread http://www.openoffice.org/issues/show_bug.cgi?id=54586 QTE ++ 1 On certain Linux machines, file locking is known to fail due to the NFS lock demon not running. 2 On certain other Linux machines, it appears that file locking fails due to some other, not yet analyzed reason. + UNQTE I guess my machines fall into the second category. yours, kos Billy Holmes wrote: Konstantin V. Gavrilenko wrote: Apparently you need to disable file locking by commenting out the following lines: either that or run lockd... -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] text-only email parts in evolution 2.4
All flame wars about html vs text emails aside, how do I get it back? I've been trying to find out how to get / use plugins for evolution, but there doesn't seem to be much help about it. This is something to take up with the evolution developers. And also, if so many people want such a feature, why isn't it a patch from gentoo? Is there some fundamental understanding that I'm missing? flame Yep, someone like you never got pissed enough to add the patch or plugin as an ebuild. I think most (including me) just got sick of evolution being more or less crap, and started using different email clients... Any help, or enlightenment about the topic, would be much appreciated, Thanks, /flame -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] POSTFIX and SASL
After spending 5 mins trying to get SASL working myself... I just emerged stunnel and switched to simap/spop3... ;) On Wed, 5 Oct 2005, Joe Strusz wrote: Ive read every forum on the site, and even ventured over the Wiki for even more redundant reading. Ive read the entire postfix handbook, and still have not found an answer to this problem. Its been two weeks now I haven't been able to send mail from outside the network to non local users. Heres my /etc/postfix/main.cf: alias_database = hash:/etc/mail/aliases alias_maps = hash:/etc/mail/aliases biff = no broken_sasl_auth_clients = yes command_directory = /usr/sbin config_directory = /etc/postfix content_filter = smtp-amavis:[127.0.0.1]:10024 daemon_directory = /usr/lib/postfix debug_peer_level = 2 default_destination_concurrency_limit = 2 empty_address_recipient = MAILER-DAEMON home_mailbox = .maildir/ html_directory = no inet_interfaces = all local_destination_concurrency_limit = 2 mail_owner = postfix mailq_path = /usr/bin/mailq manpage_directory = /usr/share/man mydestination = op, op.$mydomain, $mydomain mydomain = mydomain.com myhostname = op.mydomain.com mynetworks = 192.168.1.0/24, 127.0.0.0/8, *.*.*.*(my WAN ip) newaliases_path = /usr/bin/newaliases queue_directory = /var/spool/postfix queue_minfree = 12000 readme_directory = /usr/share/doc/postfix-2.1.5-r2/readme sample_directory = /etc/postfix sendmail_path = /usr/sbin/sendmail setgid_group = postdrop smtp_use_tls = yes smtpd_data_restrictions = reject_unauth_pipelining smtpd_sasl_auth_enable = yes smtpd_sasl_local_domain = smtpd_sasl_security_options = noanonymous smtpd_tls_CAfile = /etc/ssl/postfix/server.pem smtpd_tls_cert_file = /etc/ssl/postfix/server.crt smtpd_tls_key_file = /etc/ssl/postfix/server.key smtpd_tls_loglevel = 3 smtpd_tls_received_header = yes smtpd_tls_session_cache_timeout = 3600s smtpd_use_tls = yes tls_random_source = dev:/dev/urandom unknown_local_recipient_reject_code = 450 virtual_alias_domains = myvirtual.com virtual_alias_maps = hash:/etc/postfix/virtual And heres /etc/sasl2/smtpd.conf: pwcheck_method: saslauthd mech_list: plain login And heres /etc/conf.d/saslauthd: # $Id: saslauthd.sysconfig,v 1.1 2001/05/02 10:55:48 wiget Exp $ # Authentications mechanism (for list see saslauthd -v) SASL_AUTHMECH=pam # Hostname for remote IMAP server (if rimap auth mech is used) # Ldap configuration file (if ldap auth mech is used) SASL_MECH_OPTIONS= # Extra options (for list see saslauthd -h) SASLAUTHD_OPTS= And last but not least, heres a description of the problem... TLS and everything else works great. However as soon as I check that little box in outlook express that says my smtp server requires authentication on the client whom worsk out of the office on home internet lines.. the password box keeps reappearing, and accepting any passwords... i ran the line saslpasswd2 -c username to no avail. Same issue. Postfix was compiled with SASL support: [ebuild R ] mail-mta/postfix-2.1.5-r2 -ipv6 -ldap -mailwrapper -mbox +mysql +pam -postgres +sasl (-selinux) +ssl -vda 0 kB And when i telnet to 25 and issue the ehlo command, i do get AUTH PLAIN lines... whats going on? please help im racking my brains. Joe Strusz IT Assistant Oxford Publishing, Inc. 307 West Jackson Avenue Oxford, MS 38655-2154 800-247-3881 662-236-5510x40 [EMAIL PROTECTED] http://www.nightclub.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Size of portage tree
no, but I noticed, that reiserfs needs much less space with small files (like portage tree) than ext2/3. The only problem with this solution is you are then stuck using reiserfs... /fsflamewar :D -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Size of portage tree
I've lost 5 filesystems on reiser... So I won't touch it anymore. :P :( I'm all XFS and so far have not many problems (with over 100 machines in production). My favorite part about XFS is snapshotting and a working dump command for mounted filesystems... On Thu, 29 Sep 2005, Hemmann, Volker Armin wrote: On Thursday 29 September 2005 00:32, Bryan Whitehead wrote: no, but I noticed, that reiserfs needs much less space with small files (like portage tree) than ext2/3. The only problem with this solution is you are then stuck using reiserfs... /fsflamewar :D better than stuck with ext3 ;) http://bugzilla.kernel.org/buglist.cgi?bug_status=UNCONFIRMEDbug_status=NEWbug_status=OPENbug_status=ASSIGNEDbug_status=REOPENEDfield0-0-0=producttype0-0-0=substringvalue0-0-0=ext3field0-0-1=componenttype0-0-1=substringvalue0-0-1=ext3field0-0-2=short_desctype0-0-2=substringvalue0-0-2=ext3field0-0-3=status_whiteboardtype0-0-3=substringvalue0-0-3=ext3 they are ALL buggy - choose your poison ;) I have choosen reiser, because space is important for me - and I have a nice tape-drive, which makes backup/restore very easy... but to be honest, I never had reiserfs-bugs.. only hardware errors... -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] specifying DNS servers
This might work, but the correct way to do this is still a split DNS. Since this machine is on the border of private/public networks it should combine the two. Your DNS on this border box should be a slave to the internal master. These slave records should be restricted to being queries from the private addresses. The rest of the DNS records can be pulled from the internet/ISP's DNS. On Fri, 23 Sep 2005, Neil Bothwick wrote: On Fri, 23 Sep 2005 09:38:48 -0400, Mark wrote: I receive internet DNS information from the DSL connection, so no problem there. But I want my internal connection to know about the internal DNS servers as well. Assuming your internal DNS servers give an almost instant response (they should do on a small LAN), you can put all the servers in /etc/resolve.conf, with the internal ones at the top. Then set your DHCP options in /etc/conf.d/net to not overwrite /etc/resolv.conf. Now all you queries should go to the internal servers first. If they don't have (and won't forward and cache) external DNS information, the resolver will try the next server, the ISP one. If your internal servers are also connected to the Internet, you may not even need an external server. Given all of that, which of the above suggestions is my best bet? There's nothing above, you top-posted :( -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] nfs warning: mount version older than kernel
first, make sure portmap is running on both machines. I'm sure it is but check anyhow. Also make sure they are not wildly different versions. Do the same for util-linux package. run 'rpcinfo -p' on each machine and copy/paste the output. I have a feeling one kernel has NFS v3 and another has NFS v4... On Thu, 22 Sep 2005, Mark Knecht wrote: Hi, After months of working really well I've just recently started having trouble with our MythTV setup. I'm looking into what's gone wrong. The symptom is that recordings periodically look like you're hitting fast forward. Both audio and video are messed up in the same way. All machines are Gentoo 32-bit. The setup is like this: - dragonfly - the Myth backend machine has two PVR capture cards in it. Running on the system is mythbackend, mysql and the ivtv driver for the cards. This machine is my wife's desktop box and is updated regularly. I'm seeing various ivtv messages all of a sudden, along with a specific NFS message: nfs warning: mount version older than kernel The kernel is getting a bit old now. Emerge world happens often on this box. dragonfly ~ # uname -a Linux dragonfly 2.6.12-gentoo-r6 #3 Thu Aug 4 06:43:20 PDT 2005 i686 Intel(R) Pentium(R) 4 CPU 3.00GHz GenuineIntel GNU/Linux dragonfly ~ # - myth14 - this is actually a Myth frontend only machine, but it has a large disk and serves as the main storage for our recordings. This machine hasn't been updated in at least 45 days. I see no messages or problems on this machine. myth14 ~ # uname -a Linux myth14 2.6.12-gentoo-r6 #2 Tue Aug 2 16:31:31 PDT 2005 i686 Intel(R) Celeron(R) CPU 2.26GHz GenuineIntel GNU/Linux myth14 ~ # The two machines are hooked together via wired Ethernet going to a DLink switch. These machines have been in this configuration for roughly 2-3 months and have worked great until recently when this problem started. Does anyone know what the message above mean and is it telling me there's a real problem here, or is it truly just a warning and I should look somewhere else for a solution to this Myth problem? Thanks, Mark -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] XFCE4 problem when using package CD installation
Yea, revdep-rebuild is my friend... :wub: On Fri, 23 Sep 2005, HG wrote: Thanks, Bryan. After emerge gentoolkit and revdep-rebuild, Problem Solved! --- Bryan Whitehead [EMAIL PROTECTED] wrote: run revdep-rebuild to find broken libraries/programs. On Thu, 22 Sep 2005, HG wrote: Hi, All, I am a pretty new user to Gentoo and I like it! I am trying to install XFce4 by gentoo 2005.1 package CD. But each time when I do startxfce4, it always gives me a error message error when loading shared libraries. Did I miss anything during the installation? Here is what I did to install Xfce4 put the package CD in mount /dev/cdrom /mnt/cdrom export PKGDIR=/mnt/cdrom emerge --usepkgonly xfce4 then I update rc.conf(using xdm and XSESSION=Xfce-4 and also make a .xsession(exec startxfce4) in my home directory) My machine is kind of old(PII 400). that is why I am trying to use package CD to install everything. Thanks. Haijiang -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] nfs warning: mount version older than kernel
you don't have rpcinfo? that is not right! epm -qf /usr/sbin/rpcinfo glibc-2.3.5-r1 it should be in /usr/sbin/ maybe you need to rebuild glibc? rpcinfo will ask your portmapper what versions/protocols your mountd, nfs, etc are running as. I don't understand why you don't have this command... that is wierd. On Fri, 23 Sep 2005, Mark Knecht wrote: Will do. portmap is running on both machines but neither machine has rcpinfo. What package do I emerge? dragonfly ~ # rcpinfo -p -bash: rcpinfo: command not found dragonfly ~ # slocate rcpinfo dragonfly ~ # Thanks much, Mark On 9/23/05, Bryan Whitehead [EMAIL PROTECTED] wrote: first, make sure portmap is running on both machines. I'm sure it is but check anyhow. Also make sure they are not wildly different versions. Do the same for util-linux package. run 'rpcinfo -p' on each machine and copy/paste the output. I have a feeling one kernel has NFS v3 and another has NFS v4... On Thu, 22 Sep 2005, Mark Knecht wrote: Hi, After months of working really well I've just recently started having trouble with our MythTV setup. I'm looking into what's gone wrong. The symptom is that recordings periodically look like you're hitting fast forward. Both audio and video are messed up in the same way. All machines are Gentoo 32-bit. The setup is like this: - dragonfly - the Myth backend machine has two PVR capture cards in it. Running on the system is mythbackend, mysql and the ivtv driver for the cards. This machine is my wife's desktop box and is updated regularly. I'm seeing various ivtv messages all of a sudden, along with a specific NFS message: nfs warning: mount version older than kernel The kernel is getting a bit old now. Emerge world happens often on this box. dragonfly ~ # uname -a Linux dragonfly 2.6.12-gentoo-r6 #3 Thu Aug 4 06:43:20 PDT 2005 i686 Intel(R) Pentium(R) 4 CPU 3.00GHz GenuineIntel GNU/Linux dragonfly ~ # - myth14 - this is actually a Myth frontend only machine, but it has a large disk and serves as the main storage for our recordings. This machine hasn't been updated in at least 45 days. I see no messages or problems on this machine. myth14 ~ # uname -a Linux myth14 2.6.12-gentoo-r6 #2 Tue Aug 2 16:31:31 PDT 2005 i686 Intel(R) Celeron(R) CPU 2.26GHz GenuineIntel GNU/Linux myth14 ~ # The two machines are hooked together via wired Ethernet going to a DLink switch. These machines have been in this configuration for roughly 2-3 months and have worked great until recently when this problem started. Does anyone know what the message above mean and is it telling me there's a real problem here, or is it truly just a warning and I should look somewhere else for a solution to this Myth problem? Thanks, Mark -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] transfer large files via samba cause system freeze
Can't you also use the iocharset=utf8 and codepage=utf8? something like: mount -t smbfs -o user=user,iocharset=utf8,codepage=utf8 \\server\path /mnt/samba This doesn't work? On Fri, 23 Sep 2005, Qiangning Hong wrote: Heinz Sporn wrote: Am Donnerstag, den 22.09.2005, 16:26 +0800 schrieb Qiangning Hong: Isn't CIFS considered as a better SMBFS? It's been said. Still - do you have both SMBFS and CIFS support built into your kernel, or just CIFS? If the later is true I would include SMBFS as well and try the same thing with good old SMB as well (you may use both simultanously anyway). I add smbfs support in my kernel and tried (after change the filename to an English-only one). It works. However, as my previous mail said, using smbfs doesn't address my problem. I have plenty (~100) large files to transfer, and maybe you've guessed, they are movie files, with Chinese movie title in the filenames. And the movies directories on my winbox have also Chinese characters. If I stick in the smbfs solution, I'll have to do too many manual filename change things. So, I still prefer to find out what's wrong with my cifs and fix it. Interessting details missing: your kernel and windoze version? Linux r52 2.6.13-suspend2-r4 #4 Wed Sep 21 17:35:55 CST 2005 i686 Intel(R) Pentium(R) M processor 1.73GHz GenuineIntel GNU/Linux Windows XP Professional with SP2. -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] nfs warning: mount version older than kernel
Just to clarify, you can mount right? if so cat /proc/mounts on both machines. can you show me the /etc/exports of the server? If you are able to mount you should be able to ignore the error. It looks like everything is setup ok... unless you have firewall/iptables in the way of any packets... On Fri, 23 Sep 2005, Mark Knecht wrote: On 9/23/05, Dave Nebinger [EMAIL PROTECTED] wrote: you don't have rpcinfo? that is not right! dragonfly ~ # rcpinfo -p -bash: rcpinfo: command not found dragonfly ~ # slocate rcpinfo dragonfly ~ # Hold on guys, notice the spelling difference - rpcinfo which is what you need, but rcpinfo is what was searched for. My bad eyes. Thanks for seeing that Dave! OK, on the Myth backend client side: dragonfly ~ # rpcinfo -p program vers proto port 102 tcp111 portmapper 102 udp111 portmapper 1000241 udp 32768 status 1000241 tcp 32769 status 132 udp 2049 nfs 133 udp 2049 nfs 134 udp 2049 nfs 132 tcp 2049 nfs 133 tcp 2049 nfs 134 tcp 2049 nfs 1000211 udp 32770 nlockmgr 1000213 udp 32770 nlockmgr 1000214 udp 32770 nlockmgr 1000211 tcp 32770 nlockmgr 1000213 tcp 32770 nlockmgr 1000214 tcp 32770 nlockmgr 151 udp786 mountd 151 tcp789 mountd 152 udp786 mountd 152 tcp789 mountd 153 udp786 mountd 153 tcp789 mountd dragonfly ~ # myth14 ~ # rpcinfo -p program vers proto port 102 tcp111 portmapper 102 udp111 portmapper 1000241 udp 1024 status 1000241 tcp 1025 status 132 udp 2049 nfs 133 udp 2049 nfs 134 udp 2049 nfs 132 tcp 2049 nfs 133 tcp 2049 nfs 134 tcp 2049 nfs 1000211 udp 1026 nlockmgr 1000213 udp 1026 nlockmgr 1000214 udp 1026 nlockmgr 1000211 tcp 1026 nlockmgr 1000213 tcp 1026 nlockmgr 1000214 tcp 1026 nlockmgr 151 udp889 mountd 151 tcp892 mountd 152 udp889 mountd 152 tcp892 mountd 153 udp889 mountd 153 tcp892 mountd myth14 ~ # To my UNtrained eye the two look the same. Thanks, Mark -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Postfix receives OK, but won't send to internet
Many ISP's route all outgoing port 25 traffic to a black hole (as in they drop the packets not going to their smtp servers). You just need to configure postfix to push all your mail to your ISP's mail server. On Thu, 22 Sep 2005, Sean Lester wrote: Greetings, I have a partially working Postfix installation. It delivers messages on the localhost. It'll receive messages from the internet. But, it will not send messages to any hosts on the internet. I've checked my iptables and port 25 is open both ways. Also, each error I've gotten has been that the target smtp server has Timed Out. Can anyone point me in the right direction? Postfix? Firewall? Host or Domain Name resolution? Other? Thank you for your time. Sean -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] XFCE4 problem when using package CD installation
run revdep-rebuild to find broken libraries/programs. On Thu, 22 Sep 2005, HG wrote: Hi, All, I am a pretty new user to Gentoo and I like it! I am trying to install XFce4 by gentoo 2005.1 package CD. But each time when I do startxfce4, it always gives me a error message error when loading shared libraries. Did I miss anything during the installation? Here is what I did to install Xfce4 put the package CD in mount /dev/cdrom /mnt/cdrom export PKGDIR=/mnt/cdrom emerge --usepkgonly xfce4 then I update rc.conf(using xdm and XSESSION=Xfce-4 and also make a .xsession(exec startxfce4) in my home directory) My machine is kind of old(PII 400). that is why I am trying to use package CD to install everything. Thanks. Haijiang -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Can I do this with GNUPLOT?
This is a bit offtopic... maybe the gnuplot mailinglist? http://www.lns.cornell.edu/public/COMP/info/gnuplot/gpltinfo_1.html On Thu, 22 Sep 2005, Hareesh Nagarajan wrote: Hi All: Suppose I have 2 files, the first file's contents are (0, 100) and the second file's contents are (0, 50). Now, can I make GNUPLOT plot a graph with the average, i.e. (0, 100+50/2)? Sure, I could create a third file which stores the average (0, 75) and I could just plot the third file using GNUPLOT. The solution needs to be generic because each of these files could have any number of lines and I could have any number of files. But what I need to plot will always be the average. Thanks! Hareesh PS: Is there an IRC channel for GNUPLOT? -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] specifying DNS servers
No. Look into a split DNS setup: http://www.networkworld.com/columnists/2001/00288013.html On Thu, 22 Sep 2005, Mark wrote: Can I specify different DNS servers for each of my two physical interfaces to use? One nic is configured for DHCP, and the other is static. The DHCP enabled NIC gets its DNS server list automatically and updates (overwrites) /etc/resolv.conf. How can I point my static IP NIC to a different DNS server since it's on a different network? Or is there a different/better solution? -- Mark [unwieldy legal disclaimer would go here - feel free to type your own] -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Cups does not see devices
http://www.freebsd.org/support.html#mailing-list On Thu, 22 Sep 2005, Paweł Madej wrote: Hello, I got problem with my FreeBSD box. I installed cups cups-base and other stuff for it and when i try to add printer via webinterface it don't see any device (like parallel, usb, http and so on). Anybody got idea what could cause it? And how to resolve that problem? Thanks for any help Pawel -- Bryan Whitehead Email:[EMAIL PROTECTED]
Re: [gentoo-user] reiserfs file size issue
I would gamble your shell doesn't have support for files larger than 2GB. You'll run into this with tcsh. :( The pipes created by the shell for child processes will inherit this problem. On Mon, 12 Sep 2005, Stuart Howard wrote: Hi people This question follows on from a post I made regarding a hard drive death and backup problems, I made a little script that tar and gzip's each of the root directories in order. At the point of /usr which is a very large one I get the following error /root/fullsysbackup.sh: line 30: 18492 File size limit exceededcp /tmp/fullsysbackupusr.tgz When checking the file size I see that it is approx 2.1Gb and to be exact = 2^31 -1 After a fair bit of checking around into possible causes of the error [samba, cp, backup box runs XP, and finally reiserfs] I discovered from the reiser site that it has a file size limit of ... 2^31 -1. However this seemed strange but I accepted it. I asked a friend who is of the beleif that reiserfs 3.5 had the limit but 3.6 does not. Now for the silly question of the day, how do I know what version of reiserfs I have installed? OK well it is possible that I have got the wrong end of the stick at this point and indeed the file size limit is 2.1Gb, the solution would be to split the tgz on backup but I also have some rather large movie type files and Iwould rather have a fs that works for me rather than the other way around. I dont want to turn this thread into everyones favorite which fs for me? but any simplistic advice on the status of reiserfs file size limits and such would be helpful. stu ps. Comon England and up the Aussies :) we won the ashes! pps. Its a cricket thing ^^ ;) -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
[gentoo-user] ebuild - .src.rpm
Does anyone know how to do this? There are some simple pachakes (like netselect) that I'd love to use at work - but don't want to go thru the hastle of making a rpm .spec file. ebuild file rpm only poops out a binary RPM... and using an updated gentoo system doesn't move over well with old versions of redhat... -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] ebuild - .src.rpm
Yea... I'm not that much of n00b! I remember that newsletter when it was first posted! :) On Tue, 13 Sep 2005, Nick Rout wrote: Of course you might like to consider this post again :-) http://www.gentoo.org/news/en/gwn/20030401-newsletter.xml Be careful to note the date of the article ! On Mon, 12 Sep 2005 14:54:55 -0700 (PDT) Bryan Whitehead wrote: Does anyone know how to do this? There are some simple pachakes (like netselect) that I'd love to use at work - but don't want to go thru the hastle of making a rpm .spec file. ebuild file rpm only poops out a binary RPM... and using an updated gentoo system doesn't move over well with old versions of redhat... -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Encrypted NFS via ssh tunelling
On Fri, 9 Sep 2005, Hans-Werner Hilse wrote: Hi, On Fri, 9 Sep 2005 09:29:18 +0200 (CEST) Patrick Marquetecken [EMAIL PROTECTED] wrote: I always get this error: mount: localhost:/usr/portage failed, reason given by server: Permission denied Attach NFS port of Server (2049) to local port 2818 ssh -f -L 2818:10.32.3.172:2049 -l root 10.32.3.172 sleep 86400 Attach mountD port of Server (675) to local port 3818 ssh -f -L 3818:10.32.3.172:675 -l root 10.32.3.172 sleep 86400 so the SSH server will make a connection to its own external IP. It will also probably use its own external IP (not 127.0.0.1) as originating address. What IPs are allowed access by its /etc/exports ? -hwh -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Encrypted NFS via ssh tunelling
On Fri, 9 Sep 2005, Patrick Marquetecken wrote: Hi, I can do a nfs mount, but for security i would like to do it over ssh. I always get this error: mount: localhost:/usr/portage failed, reason given by server: Permission denied without the ssh tunnel i have no problems. There are no firewall between the two machines, ssh between both goes fine. My setup: Attach NFS port of Server (2049) to local port 2818 ssh -f -L 2818:10.32.3.172:2049 -l root 10.32.3.172 sleep 86400 Attach mountD port of Server (675) to local port 3818 ssh -f -L 3818:10.32.3.172:675 -l root 10.32.3.172 sleep 86400 Mount mount -t nfs -o tcp,port=2818,mountport=3818 localhost:/usr/portage /usr/portage ps -ef root 9165 1 0 10:22 ?00:00:00 ssh -f -L 2818:10.32.3.172:2049 -l root 10.32.3.172 root 9173 1 0 10:23 ?00:00:00 ssh -f -L 3818:10.32.3.172:675 -l root 10.32.3.172 whats wrong here ? TIA Patrick -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
RE: [gentoo-user] iptables example on Gentoo
Wow, that is news to me... I've always just banged out iptables rules and then saved them... On Tue, 6 Sep 2005, Dave Nebinger wrote: I've been trying to build a simple firewall with a DMZ for a web server. Dude, trying to use iptables directly was your first mistake. Take a spin out and look at shorewall (I'm sure others have different recommendations). Shorewall will get you up and running in no time and will easily handle the configuration stuff from your original post. Trying to manage such a complex config using iptables directly is doomed to failure; any mistake in ordering of rules, etc., will break your connectivity. Sticking with a tool like shorewall will simplify rules maintenance and pose less of a problem when performing updates later on. Dave -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
[gentoo-user] what happened to php and mod_php?
I currently run php 5.0.4 as build thru portage For some reason, portage only has up to 4.4.0 now. Was php 5.x abandoned? What's going on? http://packages.gentoo.org/search/?sstring=mod_php looks like only 4.4.0-r1 is around now... what the heck? -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] 4gb mpeg to 100mb xvid?
check out this webpage: http://www.transcoding.org/cgi-bin/transcode?back=Examples you can start the encode with crazy options to make it small... and then view it while encoding to see if the quality is good enough. On Wed, 7 Sep 2005, Arturo 'Buanzo' Busleiman wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi people, I was wondering, do you have any pointers to transcode an mpeg file to an xvid? The original is a +4gb file, and I need to create a good quality of aprox. 100mb xvid file. What are your suggestions? Any interesting sites to read about this? scripts? ebuilds? I've done some research in the area for the last two days, but I am definitely not a video guru, and many things I do not understand. Yours, - -- Arturo Buanzo Busleiman - www.buanzo.com.ar Consultor en Seguridad Informatica KTP Consultores - info AT ktpconsultores.com.ar -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org iD8DBQFDHy8dAlpOsGhXcE0RAtG1AJ4hKwTIoDebDzICqNTJv3ZHe4fXawCfWLeh J2IsjTmur5KhaN5MUb3ty00= =dgAi -END PGP SIGNATURE- -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] rsync problems
Sounds like bad memory. Run memtest86? On Tue, 30 Aug 2005, Joshua Armstrong wrote: They are syncing from the same mirror. I haven't tried changing mirrors though. If it helps, when I read the kernel logs I notice that during the time it's syncing, I see a lot of readlink() failed: I/O error for files in /usr/portage. I've run fsck on the disk and it detects no errors. All other apps can read all the files in /usr/portage without a problem. Thanks for all your help! On Tue, 2005-08-30 at 21:02 +0200, Holly Bostick wrote: Joshua Armstrong schreef: Hello, I'm having a problem with rsync on one of my servers. Every time I rsync from one of the gentoo portage mirrors, it tells me rsync error: some files could not be transferred (code 23) at main.c(1064). I've tried re-emerging rsync and re-emerging portage but to no avail. I know it isn't a firewall/routing issue because all my other Gentoo boxen can sync without a problem. Thanks! Are the other boxes that aren't having the problem syncing with the same server as the box that is having the problem? Have you tried changing the mirror that the box with the problem is syncing with? Holly -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] VPN?
Get dyndns working on one end... and then use ppp over ssh... :) http://www.csh.rit.edu/~psionic/articles/ppp-over-ssh/ http://www.faqs.org/docs/Linux-mini/ppp-ssh.html On Sat, 27 Aug 2005, Michael W. Holdeman wrote: I have a dyndns account, it keeps the dns updted. I can always get to my router/gateway which is dyndns enabled, how do I get to my ip on the server, I need to mount the servers nfs export throught the router. I setup the correct ports for nfs to foreward, is that it? If I can do that it would be very insecure, then I would disable that and set-up openvpn on the server to the router... As I said I am over my head.. Mike On Saturday 27 August 2005 12:15 am, David Miller wrote: I've been having alot of luck with openvpn it's ssl based rather than ipsec. I have found it to be easier to setup and less confusing and it has clients for various platforms including windows...which is not always the easiest platform to use IPSEC with unless you go with a commercial client. You will need to setup a certificate authority and understand the basics of openssl the rest is pretty simple. It even works behind a NAT router or firewall. If the vpn connection is lost it will re-establish it's connection automatically once it's routable again. This works for both dynamic ip clients and even the server as long as you're using some sort of deamon to update dyndns info. For the most part, atleast in my area, I find comcast IP's to be very stable. My IP hasn't changed in years. My ip least just gets renewed. -- David On 8/26/05, Michael W. Holdeman [EMAIL PROTECTED] wrote: OK I have read the vpn howto, and tunneling from the howto, and to be truthfull i am totally over my head. I have a LAN at the office including a freebsd file server, the server has an ip of 192.168.xx.xx, and several other gentoo desktops have the same +1 each. I have a linksys cable modem, and wireless router serving through the gateway of 192.168.xx.1. dns is fixed with 2 dns servers from comcast. Problem is I have a dynamis ip from comcast. I want to be able to access a desktop machine, and most importantly the bsd file server with my laptop, again with a dynamic assigned ip from remote locations. What is the best combination, and some sort of howto for dummies would be great!! Thanks in advance. Mike -- Michael W. Holdeman Powered by Gentoo Linux www.gentoo.org | Kernel 2.6.11-ck8 | Win4Lin 5-1-20 netraverse.com | Win4LinPro 6.1.1-03 win4lin.com | | -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Problems after reemerging GCC
remove the LS_LIBRARY_PATH hacks and run revdep-rebuild to find the applications that are pointing at the wrong shared libraries - or need to be recompiled. Also make sure gcc-config is pointing to a fully-installed and functional compiler. This is my output: beavis ~ # gcc-config -l [1] i686-pc-linux-gnu-3.4.4 * [2] i686-pc-linux-gnu-3.4.4-hardened [3] i686-pc-linux-gnu-3.4.4-hardenednopie [4] i686-pc-linux-gnu-3.4.4-hardenednopiessp [5] i686-pc-linux-gnu-3.4.4-hardenednossp On Mon, 22 Aug 2005, Bill Six wrote: Hi, I just installed Gentoo. I decided I wanted GCJ and Objective C support in the GCC, so I recompiled it. What do you have to do after reemerging it? Because afterwords, whenever I would try to run most applications, I would get some error like failed to load shared libraries. I read somewhere on the internet to add the new gcc path(/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130) to LD_LIBRARY_PATH and then env-update. While this stopped giving me the error I was seeing, now when I try to emerge programs, the compilation fails. It says i686-pc-linux-gnu-gcc: installation problem, cannot exec `/usr/lib/gcc/i686-pc-linux-gnu/3.3.5-20050130/gcc-lib/i686-pc-linux-gnu/3.3.5-20050130/cc1': No such file or directory I've now looked at http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?part=2chap=5, but that hasn't helped me too much. I don't know if the following helps, but my /etc/env.d/05gcc looks like PATH=/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130 ROOTPATH=/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130 MANPATH=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130/man INFOPATH=/usr/share/gcc-data/i686-pc-linux-gnu/3.3.5-20050130/info LDPATH=/usr/lib/gcc/i686-pc-linux-gnu/3.3.5-20050130 GCC_SPECS= and /etc/env.d/05gcc-i686-pc-linux-gnu PATH=/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130 ROOTPATH=/usr/i686-pc-linux-gnu/gcc-bin/3.3.5-20050130 Start your day with Yahoo! - make it your home page http://www.yahoo.com/r/hs -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT - vsftp 425 bad IP connecting
Are you blocking port 20 outgoing? http://slacksite.com/other/ftp.html On Thu, 18 Aug 2005, Willie Wong wrote: A quick search on google reveals the following as a possibility: Your server might be responding to the PASV connection by giving the lan-ip. Try setting the pasv_promiscuous=YES and/or pasv_addrs=insert wan ip here in the config. The suggestions are completely untested... and I don't use VSFTP myself, just acting as a Google-Proxy (= W On Thu, Aug 18, 2005 at 09:03:23AM -0500, Michael Sullivan wrote: One of my users is having a problem with FTP access to my server. He says that he can connect and get a listing for his home directory, but he can't do anything beyond seeing the listing. He's connecting from outside the network. I can connect and interact with my personal account through FTP just fine from inside the network, but everytime I try to connect like he does (using ftp.espersunited.com) I get a 425 Security Bad IP error. I don't have access to a computer physically outside the network to use to diagnose this problem, so working around this Bad IP error is my only option. The IP address that ftp.espersunited.com points to is the external address of my router, so it might be complaining because the requesting IP is the same as the requested IP. Any help on fixing this? Google and the vsftpd.conf man page were no help... -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Backup-software for small network
Amanda can't span tapes. So you must split up your directories/partitions so they are smaller than your tape size. Just buy something - your time is worth more than a few thousand for backups. ;) On Sun, 14 Aug 2005, Dan Johansson wrote: Hi, I'm looking for an OpenSource backup solution for a small network. Today I'm using Arkeia-Light to do the job, but it's not OpenSource and can only support one server and two clients and also have some other restrictions. So I'm now asking for suggestions for alternative solutions. These are my requirements: * Secure * OpenSource * One Central Backup Host (with tape-drive and backup-db) * Backup of Clients over Network (even through Firewall) * Support for DDS and DLT tapes * Unlimited number of Clients * Tape Management (the SW keep track of used/unused tapes) * Full/Incremental/Differential/Archive - Backups * Retention time management for Backups * Command Line Interface (for scripting) and GUI (for easy Backup/Restore) * Definition of Backup-sets * Online/Offline backup of Databases (MySQL) * And of course it should be in Portage. At the moment I see two candidates, Amanda and Bacula. Any comments (pro/con) on those two or any suggestion on some other Backup-SW. Regards, -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] NFS configuration (tcp/ip MythTV)
Version 3 should work... the internal filesize is a 64bit value... Do a search for NFS v2/v3 and you can read up on it all. (It is boring and just a simple footnote...) On Tue, 2 Aug 2005, Mark Knecht wrote: On 8/2/05, Bryan Whitehead [EMAIL PROTECTED] wrote: BTW, it could be you are using NFS v2 which is ONLY 32bit so you have the 4gb filesize limit. OK, I've built the kernels on both machines and have support for both V3 and V3 clients and servers built in. Ethereal tells me now that I'm using V3. We'll see if that stays on line longer than the V2 protocol did. thanks! - Mark -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] NFS configuration (tcp/ip MythTV)
The best way is to ask the portmapper (example below): [EMAIL PROTECTED] ~]$ rpcinfo -p program vers proto port 102 tcp111 portmapper 102 udp111 portmapper 1000241 udp921 status 1000241 tcp928 status 172 udp 1000 ypbind 171 udp 1000 ypbind 172 tcp 1003 ypbind 171 tcp 1003 ypbind 1000111 udp647 rquotad 1000112 udp647 rquotad 1000111 tcp669 rquotad 1000112 tcp669 rquotad 132 udp 2049 nfs 133 udp 2049 nfs 134 udp 2049 nfs 132 tcp 2049 nfs 133 tcp 2049 nfs 134 tcp 2049 nfs 1000211 udp 32768 nlockmgr 1000213 udp 32768 nlockmgr 1000214 udp 32768 nlockmgr 1000211 tcp 32768 nlockmgr 1000213 tcp 32768 nlockmgr 1000214 tcp 32768 nlockmgr 151 udp165 mountd 151 tcp165 mountd 152 udp165 mountd 152 tcp165 mountd 153 udp165 mountd 153 tcp165 mountd [EMAIL PROTECTED] ~]$ Look at the nfs service. It has udp and tcp. I also have protocol 2, 3, and 4 available to clients. On Tue, 2 Aug 2005, Mark Knecht wrote: Matthew, Michael and Richard, Thanks for the responses. They seem to outline the options pretty clearly. One question - once I get it converted and I think I'm running NFS using tcp, how do I determine that I actually am? Thanks, Mark On 8/2/05, Richard Fish [EMAIL PROTECTED] wrote: Matthew Cline wrote: On 8/2/05, Mark Knecht [EMAIL PROTECTED] wrote: From the MythTV-Users list I've seen people talking about using NFS devices but recommending that they be set up with TCP instead of UDP. So far I haven't yet found any Gentoo docs on how to do this. IIRC, there is also a kernel config option that enables NFS over TCP, which you need to enable on the server (maybe also client?). You should also add 'tcp' to your mount options in fstab. See 'man mount'. -Richard -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] NFS configuration (tcp/ip MythTV)
cat /proc/mounts | grep -E 'nfs.*tcp' On Tue, 2 Aug 2005, Mark Knecht wrote: Matthew, Michael and Richard, Thanks for the responses. They seem to outline the options pretty clearly. One question - once I get it converted and I think I'm running NFS using tcp, how do I determine that I actually am? Thanks, Mark On 8/2/05, Richard Fish [EMAIL PROTECTED] wrote: Matthew Cline wrote: On 8/2/05, Mark Knecht [EMAIL PROTECTED] wrote: From the MythTV-Users list I've seen people talking about using NFS devices but recommending that they be set up with TCP instead of UDP. So far I haven't yet found any Gentoo docs on how to do this. IIRC, there is also a kernel config option that enables NFS over TCP, which you need to enable on the server (maybe also client?). You should also add 'tcp' to your mount options in fstab. See 'man mount'. -Richard -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] NFS configuration (tcp/ip MythTV)
BTW, it could be you are using NFS v2 which is ONLY 32bit so you have the 4gb filesize limit. run nftstat -s (on the server) and nfsstat -c (on the client) to see what version of NFS you are using (note: what version of NFS you are using is not related to the transport - udp/tcp). I use bigger than 4GB files on Linux server/client all the time to move DVD iso's to machines with better burners... you are running the 2.6 kernel? On Tue, 2 Aug 2005, Bryan Whitehead wrote: What filesystem are you exporting over NFS? On Tue, 2 Aug 2005, Mark Knecht wrote: On 8/2/05, Matthew Cline [EMAIL PROTECTED] wrote: On 8/2/05, Mark Knecht [EMAIL PROTECTED] wrote: but how do I know it's being used? And how do I know that the rsize option is being used? Thanks, Mark Could you watch the traffic between the two using something like ethereal? This should tell you which protocol is being used. Hi Matt, OK, ethereal was pretty easy to use, and it does indeed show that I'm using TCP for packat transfer. I see a proto=NFS packet followed by a number of TCP packets with sizes of 8K bytes so this seems to verify that both options I was looking for ar indeed working. Thanks! Unfortunately this means I'm no closer to the root cause of my real problem which is mythbackend shutting down without warning. It happened again just a few minutes ago. This all started happening after I brought this NFS mount on-line as storage for the mythbackend server. I suppose I'll have to go back to the reduced storage option (15 hours instead of 120 hours) and make sure that it's really this disk/PC/network connection. Thanks again for your help. Cheers, Mark -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] NFS configuration (tcp/ip MythTV)
I think it is much more easy to get NFS working right... ;) Just my 2 cents. On Tue, 2 Aug 2005, Michael Crute wrote: Well if you are a perl or python kinda guy you could write a more sophisticated script to copy the files and update the database so that everything is transparent as far as myth is concerned. -Mike On 8/2/05, Mark Knecht [EMAIL PROTECTED] wrote: Mike, Thanks for the idea. I like the idea of being able to record locally for 15 hours safely and then just using the new NFS storage for playback only, but I think it won't work from a practical standpoint: 1) MythTV runs in conjunction with MySQL which is managing the data files. If I simply move the data files to some other location then MySQL won't know where they are for playback. 2) As far as I know MythTV expects all the data file to be in a single location for playback. I've never heard of anyone having multiple disks for playback, but if they could then your idea would possibly work. I like the idea though and will do some research to see if there's a practical solution. Possibly some sort of logical disk drive? That's a bit beyond my meager skill set. Thanks, Mark On 8/2/05, Michael Crute [EMAIL PROTECTED] wrote: Mark, Here is my suggestion to get the best of both worlds (note my limited knowledge of mythtv). Setup a shell script to copy all your video files from the myth capture directory over to the nfs share and delete the files thus clearing your local space and also allowing you to capture 135 hours. You could even cron it so you don't have to think about it. Pardon me if this is a gross misunderstanding of mythtv but if its not it should work like a charm. -Mike On 8/2/05, Mark Knecht [EMAIL PROTECTED] wrote: On 8/2/05, Matthew Cline [EMAIL PROTECTED] wrote: On 8/2/05, Mark Knecht [EMAIL PROTECTED] wrote: but how do I know it's being used? And how do I know that the rsize option is being used? Thanks, Mark Could you watch the traffic between the two using something like ethereal? This should tell you which protocol is being used. Hi Matt, OK, ethereal was pretty easy to use, and it does indeed show that I'm using TCP for packat transfer. I see a proto=NFS packet followed by a number of TCP packets with sizes of 8K bytes so this seems to verify that both options I was looking for ar indeed working. Thanks! Unfortunately this means I'm no closer to the root cause of my real problem which is mythbackend shutting down without warning. It happened again just a few minutes ago. This all started happening after I brought this NFS mount on-line as storage for the mythbackend server. I suppose I'll have to go back to the reduced storage option (15 hours instead of 120 hours) and make sure that it's really this disk/PC/network connection. Thanks again for your help. Cheers, Mark -- gentoo-user@gentoo.org mailing list -- Michael E. Crute Software Developer SoftGroup Development Corporation In a world without walls and fences, who needs windows and gates? -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] can't unload modules
Sounds like a forgotten make mrproper or make clean before a full build? I'm sure the kernel isn't making this stuff up. :D On Tue, 2 Aug 2005, maxim wexler wrote: Module unloading support is optional. You want to enable it in your kernel config. See the start of this thread. It *is* enabled Zac -- gentoo-user@gentoo.org mailing list __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: Size of a gentoo mirror
A better question would be to ask about how much bandwidth you'd be willing to give away. Storage is cheap... bandwidth can get pricey. On Mon, 1 Aug 2005, Marek Więcek wrote: Pshem Kowalczyk wrote: Hi, Together with my boss we are thinking about mirroring gentoo (both distribution and packages) . Can someone tell me how much space should we reserve for that? Read the document: http://www.gentoo.org/doc/en/source_mirrors.xml Our Mirror's usage: # du -khs /mirror/gentoo 51G /mirror/gentoo -- Bryan Whitehead Email:[EMAIL PROTECTED]
Re: [gentoo-user] screen and Ctrl-S not working any more?
Fromt he screen man page: You can still send these characters to the current program, but you must use the appropriate two-character screen commands (typically C-a q (xon) and C-a s (xoff)). The xon/xoff commands are also useful for typing C-s and C-q past a terminal that intercepts these characters. man screen then search for C-s to get the scoop. ;) On Mon, 4 Jul 2005, Martin Carpella wrote: Hi! When connecting to one of my gentoo servers via SSH, I've got a strange problem: Ctrl+S is not sent to my screen sessions any more, instead handled directly by the shell, causing the sesssion to freeze. This is extremly annoying as many emacs shortcuts require Ctrl+S. Anyone got any idea what could have changed? Best regards, Martin -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Disk Backup From One Machine To Another
Use rsh to just pip data over with rsync? Use iptables to restrict rsh... On Fri, 24 Jun 2005, Raphael Melo de Oliveira Bastos Sales wrote: Hi there, I have one machine (Machine 1) that I need backup its files periodically. I also have another machine (Machine 2) that will hold the backup. Machine 2 can see (make requests to) Machine 1, but the opposite isn't true. The network is covered by a firewall, so I don need a paranoid solution. I was thinking about doing the following: On Machine 1, put it on the crontab to put netcat waiting for requests, and when it did receive a request, dump the files. Like this: tar -jc / | nc -l -p 500 And on the crontab of Machine 2, I'd put something like this, only a few minutes later, to avoid any errors from clock differences: nc machine1 500 backup.tar.bz2 But before doing this, I'd like some suggestions or may be some better solutions you guys might know. Thanks for the attention, Raphael. -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] masking issues.
I changed from: =dev-php/mod_php-5.0.4 =dev-php/php-5.0.4 to: =dev-php/mod_php-5.0 =dev-php/php-5.0 Now I get a downgrade when I emerge. :( beavis ~ # emerge -up world These are the packages that I would merge, in order: Calculating world dependencies ...done! [ebuild U ] app-text/docbook-xsl-stylesheets-1.68.1-r1 [1.68.1] [ebuild NS ] app-text/docbook-xml-dtd-4.2-r1 [ebuild N] app-text/xmlto-0.0.18 [ebuild NS ] sys-kernel/gentoo-sources-2.6.12-r1 [ebuild U ] app-vim/gentoo-syntax-20050618 [20050515] [ebuild U ] dev-java/java-config-1.2.11-r1 [1.2.11] [ebuild NSF ] dev-java/sun-jdk-1.4.2.08 [ebuild UD] dev-php/php-4.4.0_rc1 [5.0.4] [ebuild U ] sys-apps/less-385_p4 [385] [ebuild U ] app-antivirus/clamav-0.86.1 [0.86_rc1] [ebuild U ] net-misc/stunnel-4.10 [4.09] [ebuild U ] dev-php/PEAR-DB-1.7.6 [1.6.8] [ebuild U ] sys-kernel/linux-headers-2.6.11-r2 [2.6.11-r1] [ebuild NS ] dev-php/mod_php-4.4.0_rc1 [ebuild NS ] mail-client/squirrelmail-1.4.4-r1 [ebuild U ] dev-perl/Locale-gettext-1.05 [1.04] [ebuild U ] sys-devel/autoconf-2.59-r7 [2.59-r6] Adding the mask of =dev-php/php-5.1 does not help. :( beavis ~ # grep php /etc/portage/package.unmask =dev-php/mod_php-5.0 =dev-php/php-5.0 beavis ~ # grep php /etc/portage/package.mask =dev-php/php-5.1 =dev-php/mod_php-5.1 On Fri, 24 Jun 2005, Tim Igoe wrote: Bryan Whitehead wrote: I unmasked php and mod_php so I could get php version 5.0.x installed. However, now my emege wants to install php 5.1... I don't php 5.1 series - how can I mask in 5.0 but mask out 5.1 ? unmask =dev-php/php-5.0 or mask =dev-php/php-5.1 repeat for mod_php -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] masking issues.
That worked! I had tried dev-php/php-5.1 in /etc/portage/packages.unmask but I got the same problem. I didn't realize that I'd have to be so specific about what to unmask... Thanks for the tip. On Fri, 24 Jun 2005, Neil Bothwick wrote: On Fri, 24 Jun 2005 12:08:53 -0700 (PDT), Bryan Whitehead wrote: I unmasked php and mod_php so I could get php version 5.0.x installed. However, now my emege wants to install php 5.1... I don't php 5.1 series - how can I mask in 5.0 but mask out 5.1 ? echo dev-php/php-5.1.0_beta /etc/portage/packages.unmask -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] different times in postfix log
Try restarting postfix. If you changed your timezone, clock, etc at any point without restarting postfix different parts will have different times. Might want to restart cron while your at it... ;) On Sat, 18 Jun 2005, Jan Callewaert wrote: Hi, if I watch the logs of postfix, I have different times in it. Jun 18 19:17:43 [postfix/qmgr] 1E365EDCC3: from=bla, size=4398, nrcpt=1 (queue active) Jun 18 21:17:44 [postfix/local] 1E365EDCC3: to=[EMAIL PROTECTED], orig_to=[EMAIL PROTECTED], relay=local, delay=1, status=sent (delivered to command: procmail) Jun 18 19:17:44 [postfix/qmgr] 1E365EDCC3: removed Jun 18 21:17:48 [postfix/smtpd] disconnect from localhost[127.0.0.1] this is at 21:18. Date displays the correct hour. Any idea what is causing this? Only [postfix/qmgr] is showing the wrong time, the rest is correct. Regards, Jan Callewaert -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Bad NFS performance
sounds like NFS packets are being dropped. Are you using UDP? Try using TCP... On Sun, 19 Jun 2005, Andreas Karlsson wrote: Hello all, I am getting quite sick of my self and NFS. I?ve read many posts at forums.gentoo.org, I?ve googled, I?ve tested and tested and tested... No help could be found. The problem is not unique as it seems. I have an r/w speed against the NFS-server going from ~200kb/sec to 2mb/sec. I can?t predict or find a cause for the strange speed changes. I have three computers set up: Aquarius is my main work station (AMD XP 3200+, 1 gig ram) Leo is my file server (AMD 1700+, 630 meg ram) Aries is for TV-out (AMD 1700+, 512 meg ram) All connected via a switched 100mbit network. All have 100mbit NICs. Netperf is telling me this: aquarius hermes # netperf -H leo TCP STREAM TEST to leo Recv SendSend Socket Socket Message Elapsed Size SizeSize Time Throughput bytes bytes bytessecs.10^6bits/sec 87380 16384 1638410.00 89.61 And when I do the same from/to all others the results are quite the same. So, no problem with the network. Next up NFS. I have now tried so many different combinations at my /etc/fstab that I don?t know what is up and down: It currently looks like this: aquarius hermes # cat /etc/fstab | grep nfs leo:/home/files /home/hermes/philez nfs rw,users,auto,hard,intr,tcp,retrans=5,rsize=1024,wsize=1024 I?ve tried rsize and wsize from 1024 up to 32768 (1 through 32 * 1024). I?ve tried it in combination with sync and async. Here I find async working best. The file server (Leo) has a 120 gig HD with DMA enabled: leo linux # hdparm -t /dev/hdc /dev/hdc: Timing buffered disk reads: 88 MB in 3.02 seconds = 29.15 MB/sec So there should not be a I/O related problem either. I also run Samba on it, and when I am in Windows, filetransfers are as expected on 100mbit network (haven?t even mesured them as they work fine). Can anyone please give me any hints or advices on how to proceed with this problem? Best regards, Andreas Karlsson -- Bryan Whitehead Email:[EMAIL PROTECTED]
Re: [gentoo-user] ssmtp - what wants to install it?
My reason, Disk space is too cheap to waste time keeping a small tiny place-holder MTA inactive... The fact you and others have emailed about this has already not been worth the cost in disk space... ;) On Mon, 13 Jun 2005, Mark Shields wrote: A. Khattri, Ah yes, I remember reading it's not a full-fledged MTA. I was just curious as to why it was trying to install. And to answer the first reply's question: The question I have is why would you want to remove all MTAs?. Why have something on your system when you don't use it? This is a personal server; I have ssh setup, I check the log files myself. On 6/13/05, A. Khattri [EMAIL PROTECTED] wrote: On Mon, 13 Jun 2005, Mark Shields wrote: Trying to remove any MTAs from my system, so I did an emerge -C ssmtp, which unmerged it just fine. Its a virtual dependency. ssmtp is not really an MTA, but a wrapper for a script that emulates sendmail. Its harmless and uses very little space so you should keep it (besides, you might need it if you want output from cron scripts emailed to you). -- -- gentoo-user@gentoo.org mailing list -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Compile error
Maybe you have old kernel stuff lying around? did you do a make clean first? you can try disabling the advanced tcpip kernel options till you can resync with a gentoo mirror and get newer/better sources? On Mon, 13 Jun 2005, Zac Medico wrote: Charles Trois wrote: Hello! While installing Gentoo on my G4 iMac from the Universal 2005.0 disk, I was able to go as far as configuring the kernel. But, at the make step, I got this error (text copied by hand): net/ipv4/netfilter/ip_fw_compat.c: In function fw_in net/ipv4/netfilter/ip_fw_compat.c: 83: error: too few arguments to function 'ip_ct_gather_frags' [/net/ipv4/netfilter/ip_fw_compat.o] Error 1 I can't properly look at the files, because the console is quite lame. There is nothing very special about my configuration, except that I added support for Firewire, ATM, pppoatm, speedtouch. I would be grateful for all suggestions. Charles Hi Charles, That net/ipv4/netfilter/ip_fw_compat.c file doesn't even exist in my 2.6.11 sources. If the problem feature isn't needed then perhaps you can simply disable it in the kernel config. Zac -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
[gentoo-user] proxy to hide redirects?
At my job; we have a new application (j2ee) that is throwing redirects about 5 times for each normal page load. In other words, we have 5 302's for every 1 200. Sadly, there is nothing I can do to fix the application. However, I can put a proxy between the application and the outside. I want to setup a proxy that will resolve all 302's internally so from the user/client point of view they only see 200's returned from our site and zero 302's. Is this possible? or am I just screwed? -- Bryan Whitehead Email:[EMAIL PROTECTED] -- gentoo-user@gentoo.org mailing list
