main

main@lists.alug.org.uk

2 participants
10369 discussions

Server permissions
by Chris Walker 25 Oct '16

25 Oct '16

Having bought an HP ProLiant server, I have finally managed to install Ubuntu Server on it and configure all the shares on it to replicate my old Edimax NAS. The problem comes when I try to run get_iplayer. What I used to do was run get_iplayer on this machine but save the files to the Edimax in a Media_Share folder. That same folder exists on the Ubuntu box, but although I can write to it from here, get_iplayer on ubuntu can't write to it. My intention is to have get_iplayer run every few hours but I'm guessing that it will still run as the main user on the box and therefore will still have problems writing to the share. Plainly I'd like a solution to this. My initial thought was to add the user to the same group as the one that owns the share but is this the best way to do it or is there another way? -- __ __| |_ __ __ .----------------------------------------------. / _/ _` \ V V / | mailto:alug_cdw@the-walker-household.co.uk | \__\__,_|\_/\_/ |______________________________________________|

1 0

Making sense of ps
by Mark Rogers 24 Oct '16

24 Oct '16

I'm trying to monitor the memory usage of kweb running on a Pi. I have always found ps confusing with its multiple UNIX/BSD/GNU personalities. This is what I started with: $ps -F -C kweb UID PID PPID C SZ RSS PSR STIME TTY TIME CMD pi 936 756 1 67783 111088 1 09:53 ? 00:02:47 kweb -KFJHEP I find that the SZ/RSS values increase over time then drop, presumably due to some garbage collection (I'm really just guessing!). However the %age of memory used by kweb according to "top" keeps increasing, and is closer to the values I get from: $ps uax |grep -E '(kweb|USER)' USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND pi 936 1.1 11.7 271380 111304 ? Sl 09:53 2:47 kweb -KFJHEP What figures should I care about if I'm trying to track down a possible memory leak, and how do i coax them out of ps (or another tool)? Bonus question: What's the best way to get the values I'm interested in as a comma separated string (I'm trying to build a CSV file I can investigate in LibreOffice). A simple search & replace for spaces to commas doesn't work because spaces occur within some fields (and especially if I try to include other fields like start or running times). Mark -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0844 251 1450 Registered in England (0456 0902) 21 Drakes Mews, Milton Keynes, MK8 0ER

1 0

Ubuntu has trashed the RAID
by Chris Walker 18 Oct '16

18 Oct '16

I've installed Ubuntu server on to a recently purchased HP ProLiant server. I did it first without my 2 1TB drives in it as I wanted to make sure that everything worked. It did and I could ssh into it from here. I then installed the 2 1TB drives which came from my old NAS machine and tried again. When the HP booted, it failed to see the SSD on to which I'd installed Ubuntu and as my skills aren't sufficiently good to rescue the system, and as I'd not done anything special to Ubuntu, I re-installed it. But Ubuntu has written the swap partition, not to the SSD but to one of the RAID drives which I'm not impressed by. I still have all the data on another disc so I'm not worried too much about that, I'm more concerned with how I move the swap partition to the SSD. Any suggestions on how I might go about that? -- __ __| |_ __ __ .----------------------------------------------. / _/ _` \ V V / | mailto:alug_cdw@the-walker-household.co.uk | \__\__,_|\_/\_/ |______________________________________________|

1 1

[green.blackhousekeeping@gmail.com: Re: [alug social] Hello from new subscriber]
by Steve Engledow 17 Oct '16

17 Oct '16

Did anyone else get this bizarre bit of spam? Love it :) (Sent from my Arch linux machine - I don't run Ubuntu ;) ----- Forwarded message from Hung Sing Choy Li Fut Lady虎爪 <green.blackhousekeeping(a)gmail.com> ----- Date: Mon, 17 Oct 2016 19:58:26 +0100 From: Hung Sing Choy Li Fut Lady虎爪 <green.blackhousekeeping(a)gmail.com> To: Steve Engledow <steve(a)offend.me.uk> Subject: Re: [alug social] Hello from new subscriber You are misnamed. You call yourself a Linux user group, but you should call it an Ubuntu User Group, since Ubuntu users are clearly all you are prepared to accept. They're certainly the only people you're prepared to offer help to, even if you could help them convert to Ubuntu from my experience. Once a Fedora user, condemned forever, apparently. Oh and the problem turned out to be with my hardware, not my OS. If you had been able to get over laughing at me for being so utterly stupid as to use Fedora, you might have been able to help me work that out. Can't say the attitude I've experienced has made me want to convert to Ubuntu. If that is what Ubuntu users are like, you can keep your supposedly superior Linux system. On 13 Oct 2016 12:05, "Steve Engledow" <steve(a)offend.me.uk> wrote: > Hi Ian, > > Don't worry, we'll be conspicuous enough ;) > > Steve > > On 10/10, Ian Henderson wrote: > > Hi, > > > > I'm Ian. > > > > I've just braved the new world (for me) of Linux (Ubuntu), having been > > a long-time user both personally and professionally of Windows. Kind > > of scary and exciting to find myself a complete newbie with an OS > > again. But Microsoft's push updates and the critical role that my > > laptop has for me means I need to make the change or risk periods with > > no computer. > > > > I understand that you meet every 2nd Thursday in the Coach & Horses > > pub - and am assuming that means this Thursday (13th) at 8pm? If so, > > I'd like to go along so would ask you to "make yourselves > > conspicuous"(!) if you would, please. :) > > > > I'm happy to exchange home tel no with someone if you think that wise. > > > > Take care, all, > > Ian > > (C) Copyright 2016 by Ian Henderson. All rights reserved. > > Do not pass on my email address without my prior written consent (if > > emailing multiple recipients at the same time, use the BCC field in > > place of the To field). > > > _______________________________________________ > > The Alug Social mailing list > > To post: Social(a)lists.alug.org.uk > > To change your subscription options, visit:- > > https://lists.alug.org.uk/mailman/listinfo/social > > > _______________________________________________ > The Alug Social mailing list > To post: Social(a)lists.alug.org.uk > To change your subscription options, visit:- > https://lists.alug.org.uk/mailman/listinfo/social > ----- End forwarded message -----

1 0

Re: [ALUG] Recovering a NAS
by Chris Walker 14 Oct '16

14 Oct '16

On Thu, 29 Sep 2016 16:58:07 +0100 mick <mbm(a)rlogin.net> wrote: > On Tue, 27 Sep 2016 14:52:28 +0100 > Chris Walker <alug_cdw(a)the-walker-household.co.uk> allegedly wrote: > > [ some snips ] > > > > 2009 is a long time ago. Time to upgrade. :-) > > > > I didn't realise it was so long ago until I started looking for the > > receipt. It's stored as a scanned image on the NAS ;-) > > But of course you have a backup copy... :-) Isn't that what RAID is for? So, as you might well assume from the previous line, no I don't have another copy as I've always trusted RAID to deliver everything for me. I'm beginning to see the error of my ways though. > I'd agree with Steve. If you wish to continue to use a separate NAS, > then build a PC to hold the disks and copy the data to new disks. I've > had success with both debian and ubuntu in both raid 1 and 5. My > current NAS uses 4 * 2 TB disks in a raid 5 array (backed up to a 6 TB > USB disk) built using ubuntu server 16.04. (Incidentally, if you > haven't already seen it, the "Wintel Guy" has a useful site at > http://wintelguy.com/raidcalc.pl giving all sorts of useful info about > raid configurations.) Thanks. I'll take a look at that site. I'm tending to move to the 'build a PC for NAS' idea. I've looked at small(ish) cases and now need to start looking for suitable motherboards and power supplies. > > Noel Galer suggested that I look on ebay for a replacement Edimax > > NAS box. Been there, done that, but all the t-shirts were sold out. > > You don't want an old T shirt anyway. I've already got plenty :-) > > I think the drives hold something like 850GB of files of all sorts. > Given that, I'd go for 2 * 2 TB disks in raid 1. Ok. Assuming that's the route I take, and adopt your maxim of backing that up to a large USB disc, can I automate that? I don't mind copying stuff to a NAS but then don't want to have to go to the trouble of copying stuff to a USB disc as well. The reason I want it to be automatic is that I know things will be overlooked if they're not. P.S. I remembered to copy this to the list as well this time! -- __ __| |_ __ __ .----------------------------------------------. / _/ _` \ V V / | mailto:alug_cdw@the-walker-household.co.uk | \__\__,_|\_/\_/ |______________________________________________|

3 5

diff of binary files with helper program
by Mark Rogers 07 Oct '16

07 Oct '16

I need to perform diff against DBF (dBase database) files. dvview does a decent job of turning them into text files, but I don't want to have to convert all the files to text just to perform a diff. So I want a way to do something like: diff --helper=dbview file1.dbf file2.dbf .. or diff $(dbview file1.dbf) $(dbview file2.dbf) .. so that the files get converted to text on the fly. (Neither of those are remotely valid but hopely they get the idea across.) Is there a good way to do this? One objective is to do this using streams if possible, so that if file1.dbf and file2.dbf are very large there isn't any need to intermediate storage of files, but that isn't a huge problem if that's the only solution. -- Mark Rogers // More Solutions Ltd (Peterborough Office) // 0844 251 1450 Registered in England (0456 0902) 21 Drakes Mews, Milton Keynes, MK8 0ER

1 1

2600 meeting at 18:00 on Friday the 7th October.
by Viperfang 04 Oct '16

04 Oct '16

Quick heads up about this months Norwich 2600 meeting, more info at http://www.norwich2600.co.uk

1 0

Re: [ALUG] Recording Streaming Audio
by Peter Berrie 02 Oct '16

02 Oct '16

Phil, Thanks a lot. Seems to be that if you stream at say 320k mp3, what will happen is that it will be decoded, imported into audacity, then when you save, it will be recoded into FLAC or WAV or whatever you choose. Probably the thing to do is take the FLAC version in the first place, then save to FLAC from Audacity, and there should be no losses. My service has an offline listening service as it happens, so you can use their app and download and then sign in, in some way, not sure whether that is on or off line, and play without re-streaming. If you leave the service all those saved tracks go. And having done this on one album, I cannot for the life of me find where they have stored the tracks. Today I was on the service, which is generally really nice, and got an album up. It turns out that all tracks cannot be either listened to online or downloaded, but have to be bought. This is basically a totally irritating business. I like to use my little Sandisk player sometimes when out, which is cheap and cheerful but has quite extraordinary sound quality for the price -- and more importantly, for the minimal weight and bulk. But the restrictions imposed by the streaming service with the aim of preserving copyright seem utterly ridiculous by comparison to the freedom you have with a bought CD, which you can rip and move around and listen to on whatever you want. I am now looking at mplayer, which lets you send the user agent string of your choice, as well as do user name and password. Probably there will be some ingenious and cunning way they have thought of to prevent that! Peter

1 0

Re: [ALUG] Recording Streaming Audio
by Phil Ashby 30 Sep '16

30 Sep '16

On 30/09/16 12:00, PB <peter.northerly(a)gmail.com> wrote: > Hi, > > Does anyone understand how Linux audio works, well enough to give me > some pointers on this? Hi Peter, I'll give it go, having spent far too long with my virtual arm up the insides of Linux audio stuff while getting radio stations working :) > ... > Sometimes I want to listen to the material on my tiny little portable > player when out and about. So I record the stream to file and copy > it across. I've discovered two fairly straightforward ways of doing > this. They are a bit manual, but I do not do it often so its OK. > > [recording the outgoing stream using gramofile or Audacity which > both produce WAV files - why?] This is simply because these tools pick up the raw, uncompressed digital audio stream going out of your SPDIF port, /after/ any app or media player has decoded the incoming (compressed) audio stream. They do this as it's guaranteed to work (no messing with decoders, or strange interfaces) and is always available either from the low level drivers (ALSA) or a preceding audio mixer such as PulseAudio. > There is also a package called streamripper, which seems like it > would do this, but I have not yet bothered. It lets you send a user > string corresponding to Firefox and to input user name and password. > And maybe mplayer will do it, and there is probably a way to input > user name and password on VLC from a command line startup. But the > above two methods work so I have not hacked around with alternatives > any further. You may have some success, it depends on how the streaming service protect their content using various Digital Rights Management (DRM) techniques (aka encryption/obfuscation). See below. > I would like to be able to save the stream in the mode in which its > received. If it is, for instance 320k mp3, then I would like to save > it as that, with no recoding which presumably will lower quality. > But something else must be going on with gramofile, because its WAV > that is written to disk. Similarly with Audacity. At what point are > they getting the stream and what are they doing with it? > ... > And if so, how do you get to write the digital stream being received, > with no conversions, just as it comes in, to file? @see above regarding where gramofile and Audacity get their audio from and why it's always WAV (raw) at this point. These raw audio recordings will be a bit-perfect copy of what you are hearing, no loss of quality. If you are really streaming MP3 files (not some weird AAC, WMA or other encrypted/protected format), then you /may/ be able to recover the file URL from your browser and use wget or curl to simply fetch the file. Most streaming services however go to great lengths to stop this as they don't want you taking a copy for yourself. Typical 'protection' mechanisms are: generating the majority of the browser content dynamically using obfuscated Javascript, embedding the media player in Adobe crud ware (Flashplayer), using custom codecs or whole applications that contain decryption keys, using proprietary protocols built on Websockets.. the list is endless and this is why tools like 'streamripper' may not work. I would continue using your existing technique(s), perhaps compressing the raw WAVs with flac or lame if you want to save disk space. Cheers, Phil. PS: It used to be possible to use a similar recording technique on Windows XP (from the 'Stereo Mix' device). Microsoft were encouraged by the recording industry to make that much more difficult in later versions, and some sound card manufacturers removed the capability entirely from their device drivers... think yourself lucky that won't happen on Linux :)

1 0

Recording Streaming Audio
by PB 30 Sep '16

30 Sep '16

Hi, Does anyone understand how Linux audio works, well enough to give me some pointers on this? I subscribe to an audio streaming service. I listen to this by accessing the material on Firefox, and this sends it over SPDIF out to an outboard combined DAC and headphone amp. So far everything is fine and it sounds great. However, this is at my desk....where we are not always sitting when wanting to listen to tunes. Sometimes I want to listen to the material on my tiny little portable player when out and about. So I record the stream to file and copy it across. I've discovered two fairly straightforward ways of doing this. They are a bit manual, but I do not do it often so its OK. One is a neat little command line tool called gramofile. You start it up in record mode, then tune in the stream, and it writes it to a WAV file of your choice. Why WAV, when its coming in as MP3? Dunno. The second is, you start pavucontrol, then Audacity, and after getting the source right, you just do record, and start up the stream. This records it, and you then save it as whatever you want. There is also a package called streamripper, which seems like it would do this, but I have not yet bothered. It lets you send a user string corresponding to Firefox and to input user name and password. And maybe mplayer will do it, and there is probably a way to input user name and password on VLC from a command line startup. But the above two methods work so I have not hacked around with alternatives any further. My puzzle is not about HOW to do it, its about WHAT this is doing. I would like to be able to save the stream in the mode in which its received. If it is, for instance 320k mp3, then I would like to save it as that, with no recoding which presumably will lower quality. But something else must be going on with gramofile, because its WAV that is written to disk. Similarly with Audacity. At what point are they getting the stream and what are they doing with it? When doing the above recording, what exactly is being recorded? Is it a digital stream that is being written to file? Is there any format conversion going on in gramofile or audacity? Are they maybe taking the analogue signal, and if so from where, and encoding it? And if so, how do you get to write the digital stream being received, with no conversions, just as it comes in, to file? PB

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

main