- luv-main - luv.asn.au

apache2 and ssl accelerator
by James Harper 30 Jun '12

30 Jun '12

I have squid functioning nicely as a https accelerator for tomcat6 and IIS7 (Outlook web access and terminal server gateway), and now I want to publish a mantis server that way too. The problem is that squid takes the https request and proxies it as a http request to the server running mantis. apache2 gets the http request and so any hard links in there are prepended with http, which doesn't work (eg browser tries to get css via http which isn't reachable). Squid is adding the front-end-https header to the requests which should tell apache that the front end request was https:// but it doesn't seem to be working. What do I need to do to make apache understand that the user request was https not http? Thanks James

2 1

ipsec.
by Andrew Spiers 29 Jun '12

29 Jun '12

Thought I might play around with IPSEC. It seems there are a few different linux implementations, Strongswan, openswan, racoon (which I think is based on a project called KAME) and I think you can do it with Shorewall. Have I missed any notable implementations? Which is your favourite, and why?

8 11

partx in modern kernels
by James Harper 23 Jun '12

23 Jun '12

I get the following when I try and partx an lvm volume containing a Windows 7 machine: root@hostname:~# partx /dev/vg-fast/virt-capture1-0 # 1: 2048-125827071 (125825024 sectors, 64422 MB) The above is the correct dimensions of the partition. root@hostname:~# partx -a /dev/vg-fast/virt-capture1-0 partx: /dev/vg-fast/virt-capture1-0: error adding partition 1 The interesting bits from an strace: open("/sys/dev/block/254:26", O_RDONLY) = 3 newfstatat(3, "partition", 0x7ffff554a660, 0) = -1 ENOENT (No such file or directory) openat(3, "dm/uuid", O_RDONLY) = 4 fcntl(4, F_GETFL) = 0x8000 (flags O_RDONLY|O_LARGEFILE) fstat(4, {st_mode=S_IFREG|0444, st_size=4096, ...}) = 0 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x7f83f10b0000 lseek(4, 0, SEEK_CUR) = 0 read(4, "LVM-QUSNncQ5s6iHG1gMD7g9dNFQGV6Y"..., 4096) = 69 close(4) = 0 munmap(0x7f83f10b0000, 4096) = 0 close(3) = 0 ... stat("/dev/vg-fast/virt-capture1-0", {st_mode=S_IFBLK|S_ISVTX|0660, st_rdev=makedev(254, 26), ...}) = 0 open("/dev/vg-fast/virt-capture1-0", O_RDONLY) = 3 fadvise64(3, 0, 0, POSIX_FADV_RANDOM) = 0 fstat(3, {st_mode=S_IFBLK|S_ISVTX|0660, st_rdev=makedev(254, 26), ...}) = 0 uname({sys="Linux", node="keechvs3", ...}) = 0 ioctl(3, BLKGETSIZE64, 0x12ef0e0) = 0 ioctl(3, CDROM_GET_CAPABILITY, 0) = -1 ENOTTY (Inappropriate ioctl for device) lseek(3, 0, SEEK_SET) = 0 read(3, "3\300\216\320\274\0|\216\300\216\330\276\0|\277\0\6\271\0\2\374\363\244Ph\34\6\313\373\271\4\0"..., 1024) = 1024 ioctl(3, BLKSSZGET, 0x12ef0f8) = 0 ioctl(3, BLKPG, 0x7ffff554b880) = -1 EINVAL (Invalid argument) ... close(3) Any suggestions? Kernel is debian 3.2.0-2-amd64 Thanks James

1 1

Re: [luv-announce] Speakers, please?
by Russell Coker 22 Jun '12

22 Jun '12

On Thu, 21 Jun 2012, lev(a)levlafayette.com wrote: > This is a general pitch to all members of LUV on whether they would > consider stepping up to the speaker's plate for the community. So what happened to all the people who answered that survey about speaking at a LUV meeting? Maybe we should have such polls ask people for their email address... -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

3 2

Linux Journal digital edition and wget
by Mark Trickett 20 Jun '12

20 Jun '12

Hello All, As are others, I am a Linux Journal subscriber, and they have discontinued the print edition, providing a PDF from the web instead. I get a monthly email, with a link, and clicking on that brings up a page in FireFox, and an auto start download with the FireFox download manager. Sometimes it "completes" short of the whole file, useless, but usually I can get with multiple tries. This month, no such luck. I have tried copying and pasting the link into wget, but without success, there is a stub document, and wget then exits instead of listening for the following file. I have looked at the wget man page, and there is a provision for getting, and saving, session cookies, as well as ways to do login credentials, but I am not certain of quite what is required. I do have login credentials available, but appear not needed. The URL is, without all the final string as I am required to not pass that on, but of 6 alphanumeric characters, a hyphen, then fourteen characters, another hyphen, then ten characters. There is a shortened version follows, which should fail in a browser, but provide the first portion for reference. http://linuxjournalservices.com/portal/wts/cgmciy- I am wondering whether I need recursion, or some other option to get wget to wait for the download to start on the opened connection. Regards, Mark Trickett

5 7

RAID (md) bug in some 3.2.x and 3.3.x kernels
by Chris Samuel 19 Jun '12

19 Jun '12

Hi all, For those who've not yet seen this yet, Neil Brown has reported a nasty bug in the MD software RAID driver introduced into 3.4-rc1 and fixed in 3.4-rc5, but unfortunately the bad commit was copied to the stable@ address and so got backported into the 3.2.14 release (and fixed in 3.2.17) and the 3.3.1 release (fixed in 3.3.4). http://neil.brown.name/blog/20120615073245 # The bug will only fire if, while the machine is shutting down, # and array is partially assembled but not yet started. If the # array is started and running it is safe ... as long as it stays # that way. # # The typical way that an array can get into this "partially # assembled by not started" start is for "mdadm --incremental" # to have been run on some, but not all, of the devices in the # array. If it had been run on all it would have started the # array, but until then it is assembled but not started and this # is the dangerous situation. It won't toast your data, but will toast the MD metadata and break your array. Your best way to recover is to have an already existing copy of the output of "mdadm -Evvvvs" for your arrays. See Neils blog for more info. cheers, Chris -- Chris Samuel : http://www.csamuel.org/ : Melbourne, VIC

2 1

Linux friendly banks
by Glenn McGrath 18 Jun '12

18 Jun '12

Im having trouble use NAB's online bank under firefox (under debian testing), the default gnome browser works but im redirected to a website ofr mobile browsing (im on my PC). Ive had other issues with NAB so im looking to change banks. Can anyone recommend an Australian bank that has a goon linux friendly online banking ? Thanks

12 17

Oracle Linux
by Russell Coker 17 Jun '12

17 Jun '12

Avi, you said that Oracle Linux is now free with all patches and updates. How do I get it? http://www.oracle.com/au/technologies/linux/index.html The above page only seems to have a link to how to pay for it. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

2 1

Re: Linux friendly banks
by Alex Hutton 17 Jun '12

17 Jun '12

Hi Glenn, You could try running firefox with new profile, so that it doesn't load any of your plugins. iceweasel -no-remote -ProfileManager at the command line should do the job. Cheers, Alex On 17/06/2012 5:00 PM, "Glenn McGrath" <glenn.l.mcgrath(a)gmail.com> wrote: I did some digging, it looks like like its the torbutton addon installed (not even activated) that stops NAB online banking from working. Any of you have it installed by any chance ? _______________________________________________ luv-main mailing list luv-main(a)luv.asn.au http://lis...

1 0

VIM command for editing patches
by Russell Coker 12 Jun '12

12 Jun '12

Is there a VIM command that can delete the current line and all following lines until it meets a line starting with "---" (or "diff" depending on the type of diff file)? I'm still working on merging patches and looking for ways to make it less painful. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

5 7