- luv-main - luv.asn.au

Making IPV6 routes Persistent in RedHat
by Leigh Sharpe 10 Feb '12

10 Feb '12

Hi All, I'm having trouble getting RHEL to re-instate persistent IPV6 routes on reboot. I have the routes set up in /etc/sysconfig/network-scripts/route6-eth0 but they aren't getting applied on startup. But when I do an ifdown followed by an ifup, or a 'service network restart' I get the following message: RTNETLINK answers: No route to host ..Which is fair enough, as it seems that the IPV6 network isn't actually ready when the route is added. Has anybody got this working before? Is there somewhere else I should be adding the routes? Cheers, Leigh.

4 5

LUV hardware library
by Russell Coker 07 Feb '12

07 Feb '12

I'm starting the LUV hardware library tonight. I will bring a briefcase full of computer stuff which has no monetary value to me (it's stuff that either I got for free or which I can't use or expect to sell) but which would be really useful to anyone who has systems which need such parts. It's probably too late to expect anyone to bring any hardware to donate tonight, but one thing I could use that many people have is a bunch of anti- static bags. Please bring a pocket full if you have some lying around. For donations, any hardware which could be useful for someone and which fits in your pocket will do. Some items of particular interest are: Small USB storage devices. Probably most people are like me and have no use for anything less than 1GB. But some people can really use them (like the relative who took all my 128MB USB devices for MP3s to play in their car). PCI cards. As PCIe is in all new systems there isn't a lot of use for PCI apart from all the old systems that are in production. PCI Ethernet cards are really good for making old systems into routers. IDE disks and small SATA disks. Probably most people use nothing less than 300G nowadays, but there are lots of uses for 40G disks (such as making old systems into small servers with RAID-1). Please don't bring me more than 2 disks at one time without prior arrangement, due to the mass it would be inconvenient to unexpectedly receive many disks. USB keyboards and mice. USB cables and extension cables. RAM, anything faster than PC-133 (I think I've got a lifetime supply of that) and bigger than 128MB. ECC RAM is useful, but please make sure it's labelled as such - otherwise ECC RAM can look like broken RAM... Also if you attend tonight and find that some type of small and old hardware that you need isn't in the collection then feel free to make requests. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

1 0

mysql connection problems
by Russell Coker 06 Feb '12

06 Feb '12

4 4

Re: RAID-1 synchronisation
by Craig Sanders 06 Feb '12

06 Feb '12

On Sat, Feb 04, 2012 at 10:44:51AM -0800, Daniel Pittman wrote: > I suppose it is something more akin to the fsck that ext* still have, > but XFS omits: you can totally trust the journal to recover for you, are you sure you don't mean btrfs there instead of xfs? xfs_check and xfs_repair definitely exist. btw, when btrfs is stable/production-system-ready then the authors are correct - fsck won't be necessary. same as it isn't necessary on ZFS. they both do (or will do in the case of btrfs) continuous consistency checking and file-level checksumming. and the btrfs authors are right to focus on improving btrfs itself rather than getting too side-tracked by an fsck tool. craig -- craig sanders <cas(a)taz.net.au> BOFH excuse #374: It's the InterNIC's fault.

1 0

one free desktop pc for people who need it
by Rock 26 Jan '12

26 Jan '12

hi all I have one free desktop pc for people who need it. it is in good condition and works smoothly. I tested Windows XP and several Linux distributions on it and all of them working perfectly. it can be used for checking email, document processing and small games in daily work. now I would like to give it to someone need one desktop for daily using or Linux study. the details are: CPU:Intel Pentium4 2.80GHz Motherboard:Sapphire RS300-SB200 RAM:768 MB Harddisk:WD 80GB+40GB Graphics:Nvidia GeForce FX 5500 ( 256 MB ) ATAPI DVD R/W Sound:ATI IXP150 AC'97 Audio Controller Network:realtek RTL8139. it does not contain the monitor, keyboard and mouse. if interested you can send email to me. yours Rock

1 0

reverse proxy with authentication
by James Harper 25 Jan '12

25 Jan '12

I need to secure a bunch of web apps on separate internal servers for accessing from the internet, and it seems that a reverse proxy is probably a decent way of doing this. The problem is going to be authentication as I want to put front end security on the proxy (so unauthenticated users can't touch my apps with their own dubious security) but some of the apps require their own username and password, potentially different to the proxy. This won't work using conventional web security obviously, so I'll need to get a little creative. What I was thinking is something that goes like this: . User goes to apps.company.com.au/auth and enters username + password . Proxy validates credentials and sets a cookie . The Proxy validates the cookie for each request then 'eats' it before passing the request to the internal server . Any access without the set cookie gets redirected to the /auth page for authentication . Apps are accessed via apps.company.com.au/appname My google-fu is weak just lately though and I can't find anything that might do what I want... the reverse proxy part is fine (squid or apache) but not the front end authentication layer. The docs for mod_proxy (which I've used before) don't seem to suggest that I can do it that way. Any suggestions? Thanks James

2 2

reading raw disk of Dell perc RAID adapter
by James Harper 19 Jan '12

19 Jan '12

I have been handed a raw disk of a Dell perc (I think) RAID adapter. It was a RAID1 array but the machine is broken and can't be repaired in the short term. I was expecting to be able to track down the partition table at some offset into the disk and then identify the partitions etc but of course it's not that easy. I've found a few instances of the 0x55 0xaa partition signature at the end of a sector but only one actually appears to be a partition table (about 64MB from the start of the disk), but even then it doesn't look right. I then tracked down an NTFS signature (the disk is out of a Windows machine but I'm trying to get the data off on a Linux machine) but ntfs-3g and other tools says it's corrupt. So I'm guessing that one of the following is happening: 1. The RAID controller crapped all over the disk before the system failed and it is actually corrupt 2. The RAID is actually RAID0 (I was assured it was RAID1 but you know how these things go) 3. Dell have some non-linear mapping of partitions dmraid doesn't know anything about it, which kind of makes sense as the Dell raid controllers are normally not fakeraid. Any suggestions? Thanks James

2 1

Linux and music
by Kathy Reid 13 Jan '12

13 Jan '12

Hi everyone, Happy new year! If anyone has any advice for me on the following, it would be very much appreciated. Problem statement: I'm reliant on Windows for music. I own an iPod shuffle, and am reliant on iTunes to download music to this device. What I would like to do is buy another, non-Apple music player, and fulfill the following requirements; 1. Run a music program within Linux (Ubuntu / Unity) 2. Download music legally within Linux 3. Load my music player from within the music program 4. Convert the music from iTunes/Apple format and have it available to the Linux music program. What are my options? Regards, Kathy

7 9

Sympa MySQL changes mean upgrades are difficult
by Jesse Stevens 12 Jan '12

12 Jan '12

Heya again I've found that my previous question RE moving Sympa can be partially fixed. I've found that on the new server running the sympa --upgrade will set things ready to go but still have no subscriber data. If I then manually dump the entries from the MySQL database (subscribers_table) on the old Sympa server to the new in the right place, and then go to the column created which didn't exist on the old server (robot_subscriber) and add the proper robot details relating to the new server, the subscriber will attach as it's supposed to! I've been using phpmyadmin to browse the tables and check out the structure as I'm really not that much of a MySQL admin, so does anyone know what I could use to change the contents of every single entry for "robot_subscriber" in the table "subscriber_table" in the database "sympa" ? Also this raises another question - why is grabbing a subscriber list so difficult? To me it would be a vital thing to be able to dump this list in whatever format you'd like for easy backups and fail safe restoration. Sometimes these lists can be thousands of entries! Any help is very much appreciated Jesse

2 2

Migrating from an ancient Sympa to modern Sympa setup
by Jesse Stevens 12 Jan '12

12 Jan '12

Hey there list - long time no speak! I've been migrating everything off an ancient Debian 3 Server (kernel 2.2.20!!) which was until just now still in use. Mostly things are going well except for trying to get Sympa off it. Basically it's running Sympa 4.1.5 using MySQL for data storage. According to what I've ready I really just need to copy the MySQL database, mail list home configs (/var/lib/sympa/*) and once copied, run the sympa --upgrade tool. Doing it this way simply upgrades the database structure and dumps all of the subscription information. The mailing lists survive but with subscibers which isn't good. I see that I can run sympa --dump to dump the subscribers into files located at /var/lib/sympa/expl/maillinglistname/subscribers.db.dump But the format of this file doesn't allow importing into sympa via sympa --import=listname ./filename The formats are different. Does anyone have any idea of what I can do here? My preferred path was to create the lists nice and fresh on the new server from scratch, and simply import subscribers but I can't find a path to do that. Is there perhaps a script that will perform the conversion around that someone else has used? I can't seem to find more info in the googling I've done.. Jesse

1 0