- luv-main - luv.asn.au

Adding wireless to my network - OpenWRT?
by Jason White 28 Aug '13

28 Aug '13

I've discussed my wireless needs on the list before - I'm thinking that the best solution would just be to buy a device that will take OpenWRT and serve as an 802.11n access point. My network configuration: Desktop Linux workstation with ADSL PCI card installed, also running DHCP, Bind, Postfix and radvd for IPv6 - native IPv4/IPv6 dual-stack connection to Internode and a static Ipv4 address. FreeSWITCH (for VoIP) is running directly on this machine so it isn't behind NAT. Gigabit switch -> printer, SIP phone, laptop on local LAN. Now I want to add 802.11 wireless for phones, tablets, etc., both IPv4 and IPv6. Given previous discussions and my research, I'm considering just buying a device that will take OpenWRT, attaching it to the Gigabit switch and either bridging the wired and wireless networks or routing the packets between separate networks. Hardware suggestions would be welcome, as would any alternative ideas on how best to proceed with this. One option would be the devices that the Buffer Bloat project is recommending - all of the drivers are open.

4 9

Re: Copy files to an audio player and playing order
by Petros 28 Aug '13

28 Aug '13

From: "David" <bouncingcats(a)gmail.com> > On 27 August 2013 22:48, Peter <Petros.Listig(a)fdrive.com.au> wrote: >> >> I have a digital recorder / player, a Sony PCM-M10. >> >> It can take FAT formated Micro SD cards. >> >> I can access them via USB, mount them and copy folders and files. >> >> However, there is one glitch: the order of the files. I call them >> 01_dada.wav, 02_ased.wav etc. ad hope it plays it in that order. > > [..] > >> Any ideas or experiences how to fix it? > > Hi Peter > > I typed "fat directory order" into a search engine and with 5 seconds effort > found this apparently related info: > http://www.murraymoffatt.com/software-problem-0010.html Thanks. I found fatsort there too (see other mail) > I have no additional knowledge to offer you but I thought I might as well > send it to you as you don't mention having done any searching yourself: I searched but did not use the same words. I was not even sure yet what the reason was (just guessed it) > I assume you sent this to luv-main because you are using a linux machine to > copy the files? That page does emphasises other offtopic operating systems > and non-linux filesystems, but it does include a couple of > references to linux utilities and hopefully might provide you other > useful insight into > your problem. I am using Linux (u Ubuntu desktop). The recorder/player came only with a toaster with some scribbling on it ("Microsoft Windows" etc.) It is under my coffee mug now. It was an utility + understanding problem. I was not sure what caused the random playlist order - creation time? Directory order? It seems to be directory order + a MSGLIST.MSF file (created after scanning a SD card using the directory order) The MSGLIST.MSF was created after my first copy in random order (tar -cf - . | tar -xf -C /mnt) and most likely not updated after I copied second time, this time in sorted order. Thanks Peter

1 0

Re: Copy files to an audio player and playing order
by Petros 28 Aug '13

28 Aug '13

From: "Glenn McIntosh" <neonsignal(a)memepress.org> > On 27/08/13 22:48, Peter wrote: >> for f in `find . -type f`; do cp $f /mnt/$f; done > > Apart from not catering for embedded spaces in filenames, the 'find' > here won't sort the files into a collation order. Sorry, I wanted to pipe it to sort (find . -type f | sort) but forgot to write it down. The white space problem was exactly the reason while I did not start it last night. Too late for thinking;-) > Also, there's a program 'fatsort' in the package repositories which will > do recursive directory sorting on FAT16 and FAT32, though I haven't > tried it myself. It worked but it does not fix it. It looks as my problem is another file found on the SD card, MSGLIST.MSF. http://www.sony.com.au/support/faq/1187263 "MSGLIST.MSF is a file that contains information such as the order of folders, playback order of audio files and recording date and time. If user connects the recorder to the computer, the MSGLIST.MSF files are displayed but do not delete or change these files. If these files are operated, the saved files may be broken or the recorder may not work correctly." Oh well, I guess because my first copy was random it remembers the unsorted list. I deleted the MSGLIST.MSF file and it was recreated. And now, most likely because the FAT entries are sorted, the order is right. So next time: 1. Copy 2. fatsort before disconnecting. Thanks! Peter

1 0

Copy files to an audio player and playing order
by Peter 27 Aug '13

27 Aug '13

Hi, I have a digital recorder / player, a Sony PCM-M10. It can take FAT formated Micro SD cards. I can access them via USB, mount them and copy folders and files. However, there is one glitch: the order of the files. I call them 01_dada.wav, 02_ased.wav etc. ad hope it plays it in that order. The manual states: "If you have copied music files using your computer, those files may not be copied in order of copying because of the system limitations. If you copy a music file one by one to the PCM recorder, you can display and play back tracks in order of copying." So I copied the files after sorting them: find . -depth -print0 | sort -z | pax -wd0 | tar -xf - -C /mnt Still, the order is random. I wonder why.. My memory of FAT is blurry. During studies I "visited" floppy sectors to get deleted data back. If i am not mistaken, if a directory is new, it puts filename by filename in a directory list and references, one by one, although I am not familiar whether that changed with VFAT and long filenames. I may experiment with "real" copy one by one - directories first and then a kind of for f in `find . -type f`; do cp $f /mnt/$f; done or trying to figure out what changing the modification time does.. at the moment I am not sure why it's not working. Any ideas or experiences how to fix it? BTW: Besides of that I am very happy with the Sony PCM-M10. I have it on my HiFi at the moment and it has a crystal clear and full sound. Thanks Peter

3 3

Re: The End of LUV, Inc? An Example of Federalism and Unitary Organisational Structures
by zlinw＠mcmedia.com.au 26 Aug '13

26 Aug '13

Daniel Cross said, >On Sun, August 25, 2013 1:49 pm, Daniel Cross wrote: >> Lev, you say that this will be discussed at the AGM, yet the rest of your >> email (beginning "what will change") reads as though disincorporation is a >> decided matter. >"What will change" is always with the caveat of "if" the motion is passed. >The question mark in the title was also meant to indicate the uncertain >nature of the proposition. A comment on the above statements on this issue, now I have worked on the committees of two volunteer orginisations and __DO__ apreciate what is trying to be done to simplify matters, but on reading the whole series of messages I must say I have same ___impression___ as Daniel, that is the merger is being presented as a done deal and one simply cannot do this in a democratic group such as this. On the whole I am not a bit surprised on the groups reaction. When presenting any information that may be headed for a hostile reception it is wise to be carefull of the wording, rereading and correcting to make sure it says what you wish it to see without giving a false impression. It has taken me usually 2 to 3 __HOURS___ to formulate a 400 word post under these conditions (an undisclosed news thread) and I am proud to say, I never had a single post missinterprited. Lindsay

6 6

Re: Kernel BUG in Debian 7.1
by zlinw＠mcmedia.com.au 26 Aug '13

26 Aug '13

Chris Samuel said, >Out of interest, why build such an old kernel for this? >If you're after the latest ACPI code you'd want to try a 3.10 kernel (or from Linus's tree if you want to test what will be 3.11). Several reasons, when doing any work in a new area, ie research work, in this case 64bit, one should __ALWAYS__ work from a known starting point. In this case 3.7.9 was the last kernel I had compiled that I knew worked. Second point is this particular machines main purpose is a development platform for a real time 3D terrain simulation and uses the NVidia closed source driver. One needs to tread I little carefully on new versions of this (same with the AMD driver). I had trouble getting the version I am using working with a later kernel (3.8 series). I am not really interested in ACPI execpt it gives trouble on this particular machine, given the machines primary use low power consumption is a non issue. Sitting right next to it is another machine of good deal lower specs that is used for normal every day work. This message is being writen on it. It would probably be a good idea to try a latter kernel from the 3.10 series (or what ever) to see if the ACPI issue has been sorted, the MB chips set (the X79) being fairly new. I will probably do this in time. Using Linux for 20 years, kernel 0.96d being the first one, Lindsay

3 2

How do I get out of here?
by Shane Deering 25 Aug '13

25 Aug '13

After reading the emails about the end of LUV, I've decided I don't want to be a part of this any more. I logged onto the LUV site, but I can't find a button that says I don't want to be a member any more. How do I get out of being a LUV member? Shane.

7 7

Kernel BUG in Debian 7.1
by zlinw＠mcmedia.com.au 25 Aug '13

25 Aug '13

Now this is just for information. Yesterday I an into a problem with a new install of Debian, motherboard being a Gigabyte X79-ud5 (this is significant). Debain 7.0 was installed from DVD's then upgraded from the net to 7.1. This worked OK for a number of days then the follwing error came up in the kernel log (it did not appear in dmesg). "kernel BUG at /build/linux-s5x2oE/linux-3.2.46/mm/slab.c:514! invalid opcode: 0000 [#1] SMP CPU 4" This was followed by a register dump and a call trace and another message.. "BUG: unable to handle kernel paging request at fffffffffffffff8" an another register dump and call trace follwed by the message.... "Fixing recursive fault but reboot is needed" Follwed by........ "INFO rcu_sched detected stalls on CPUs/tasks: { 4} (detected by 1, t=15002 jiffies)" This was followed by an NMI backtrace....... After reboot system started giving regular bug reports stating a CPU had hard locked and the system was for all purposes unusable. After anothe reboot I mamaged to get it in a lucid state for enough time to compile my own kernel (3.7.9). System has worked OK since the custom kernel was installed. A glance through the logs showed the debian kernel was having problems with the APIC. The previous version of debian testing obtained around 2 months before the releae of 7.0 would not boot at all on theis motherboard due to APIC problems. It seems the current Debian kernel does not like a Gigabyte GA-X79 ud5 mother board. System is working no problems on the custom kernel. Its probably worth sending Debian a bug on this................... Using linux for 20 years, Lindsay

4 3

Fwd: Re: [Linux-aus] LUV considering joining LA
by Russell Coker 25 Aug '13

25 Aug '13

---------- Forwarded Message ---------- Subject: Re: [Linux-aus] LUV considering joining LA Date: Sun, 25 Aug 2013, 19:23:40 From: "Patrick Elliott-Brennan" <mail(a)elliott-brennan.id.au> To: linux-aus(a)lists.linux.org.au Regarding James Polley's comments in relation to SLUG and LA. As someone who was involved in the transition and is still on the committee, I can confirm that there has been no downside for SLUG. We've experienced no unforseen difficulties and I'm unable to identify any on the horizon. I'd be interested to hear from any of the LUGs who have made the change to see if they've experienced any difficulties as a result. Patrick Elliott-Brennan SLUG ----------------------------------------- -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

1 0

Re: Kernel BUG in Debian 7.1
by zlinw＠mcmedia.com.au 25 Aug '13

25 Aug '13

Oops, I made a blunder....... APIC, ACPI Trent said......... >> A glance through the logs showed the debian kernel was having problems >> with the APIC. >Perhaps verify this experimentally, by booting with noapic. Its the ACPI thats the issue, this being the same in the version of testing. I had a look around the net a couple of months ago when the previous version gave problems, everyone said the ACPI errors should not stop the kernel from working, it appeared though to be so (Note 1). In the custom kernel its disabled in the config, and it works like a dream. Note 1: Its difficult to say though, the custom kernels config is much simpler than the debian one and I have found in the past some options can have some real wierd effects. I have as yet not reported a bug as I like to verify there really is a problem and I cannot find anyone else on the net having an ACPI issue on Linux with the GA-X79-ud5 mb. Lindsay

1 0