June 2016 - Talk - lists.gtalug.org

Re: [GTALUG] fuel [was: mysterious restarts]
by James Knott 19 Jun '16

19 Jun '16

On 06/18/2016 07:34 PM, Loui Chang wrote: > Not a great incentive. Propane which is cheaper and burns cleaner than gas, is > already widespread and easy to obtain and is already being used in vehicles. > > Why hasn't it overtaken gasoline as the fuel of choice? > > It's still tricky to handle and it's also a fossil fuel, which adds CO2 to the atmosphere, no matter how clean it is otherwise. We have to get away from fossil fuels.

3 2

RJ45 crimper -- good ones?
by William Park 17 Jun '16

17 Jun '16

I want to buy RJ45 (8c8p) crimper, so that I can make ethernet cable of exact length I need. Which brand is good quality? Sent from Yahoo Mail on Android

6 5

Re: [GTALUG] Thinkpad [3]: Aligning size of dm-crypt and partition/filesystem
by Matt Price 16 Jun '16

16 Jun '16

On Wed, Jun 15, 2016 at 11:59 PM, Allan Zhang <allan(a)azic.ca> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA512 > > Matt, > > just curious, is there any reason you chose a MBR +UEFI boot? > > Surely your laptop came with Windows, which mandates that GPT accompanies > UEFI. > I think because this was (is!) my first UEFI laptop and I was unfamiliar with GPT when I first partitioned the disk. I wiped windows right away so tha wasn't an issue for me. It was also my first Arch install, so there were a number of unfamiliar elements, and I stuck to what I knew where I could. > > Could you have used your M.2 for swap? > I certainly expected to. On boot, the kernel seemed to have trouble finding the M.2 device, and I couldn't immediately tell what the problem was, so it seemed easiest to make space on the old drive and put the swap there. > > As for suspend to RAM, aka regular suspend, I find my Thinkpad does not > live for too long (a few hours) in this state, evident by the "i" LED not > blinking anymore. It's very strange that you have to remove the battery > though. > I don't think those two issues are the same. The battery doesn't drain out on my laptop; instead, I think something goes wrong in the electronics of the power circuitry. I'm not sure what causes it but there are reports of similar behaviour online. > > Also, wouldn't suspending to RAM defeat the purpose of your encrypted disk? > ... yeeeeesssss... though since I'm not encrypting the suspend2disk image either, that is at least as big an issue as s2ram, I suppose. In general it seems to me that best practices are still hard to implement fully unless you're willing to devote a fair amount of time to develoing the necessary expertise. - -------- Original Message -------- > From: Matt Price via talk <talk(a)gtalug.org> > Sent: June 15, 2016 9:22:13 PM EDT > Cc: GTALUG Talk <talk(a)gtalug.org> > Subject: Re: [GTALUG] Thinkpad [3]: Aligning size of dm-crypt and > partition/filesystem > > On Wed, Jun 15, 2016 at 5:59 PM, o1bigtenor <o1bigtenor(a)gmail.com> wrote: > > > On Wed, Jun 15, 2016 at 9:43 AM, Matt Price via talk <talk(a)gtalug.org> > > wrote: > > > Final question! In order to enable suspend to disk, I ended up > shrinking > > my > > > root partition (I also added an M.2 SSD, so the whole thing was a bit > > > complex). In the end I got it to work, but I made some kind of an error > > and > > > currently my on-disk “crypt” is a little larger than the partition and > > file > > > system that I am using for root, so I have some wasted space on the > disk: > > > > > > ➜ ~ sudo fdisk /dev/sda > > > > > > Command (m for help): p > > > Disk /dev/sda: 238.5 GiB, 256060514304 bytes, 500118192 sectors > > > Units: sectors of 1 * 512 = 512 bytes > > > Sector size (logical/physical): 512 bytes / 512 bytes > > > I/O size (minimum/optimal): 512 bytes / 512 bytes > > > Disklabel type: gpt > > > Disk identifier: 9B4C2E92-6C7D-4C69-B67B-1823E4DB9BA6 > > > > > > Device Start End Sectors Size Type > > > /dev/sda1 2048 1026047 1024000 500M EFI System > > > /dev/sda2 1026048 106907647 105881600 50.5G Linux filesystem > > > /dev/sda3 106907648 483340287 376432640 179.5G Linux LVM > > > /dev/sda4 483340288 500117503 16777216 8G Linux swap > > > > > > ➜ ~ sudo cryptsetup status cryptsetup > > > /dev/mapper/cryptsetup is active and is in use. > > > type: LUKS1 > > > cipher: aes-xts-plain64 > > > keysize: 256 bits > > > device: /dev/sda2 > > > offset: 4096 sectors > > > size: 105877504 sectors > > > mode: read/write > > > > > > Anyone out there have advice abot how to safely and reliably expand the > > > partition and the file system so that they both end precisely where the > > > crypt does? > > > > > > Thanks again guys! as always, it’s much appreciated. > > > > > > > No ideas on the other two problems but on this one I would recommend > > using something like: > > > > boot your system using a system rescue disc > > > Yes! I use the Ubuntu livecd just because it's easy to get > > > adjust your partition(s) (carefully please) using gparted > > > is that adequate? Can I be sure that the crypt and the partition match > exactly? I feel like it's somewhat difficult to measure the size & extent > of a crypt in units that gparted also understands > > > close > > > > reboot as regular > > > > Should be real easy (assuming that the empty space is right beside > > where you want it. > > Recommend a backup before but then you already know that - - yes? > > > yes... though I don't always follow recommendations. > > > > > Let us know what you use > > > > > - ------------------------------------------------------------------------ > > - --- > Talk Mailing List > talk(a)gtalug.org > https://gtalug.org/mailman/listinfo/talk > -----BEGIN PGP SIGNATURE----- > > iQI5BAEBCgAjHBxBbGxhbiBaaGFuZyA8YWxsYW5AYXppYy5jYT4FAldiJB4ACgkQ > lQhQ+uFN8kNrPw/+OKK3a5w8oAK3ElJ5CBEyy9Hk+1BSvd+24t04v0yZGejlIBMB > QzJCnpG2FTvQ/Knjr4glTjCGNSXKkMZPdlxD3Jv+laFmMscg9Ohq8aoKO23gGtvQ > yqu9wxXVW1p94RaS5qhkO67QZyPnOB31vZJusnI8uFD4QGa7hASVA1FMnly70Uk6 > finTa3YoQJ0fuLwS+DuVjvD1RHpc8qnM2+LpnlRw6djUilYqMYcRw8aXrApJrhKO > NsImN45XHIRQL7tbaDtGfJIlbr5CB5OvG8Fyi7baL2qDZCiGjKAWy4STFikDM/W2 > 7ulDIgthYrWMZkqJg+6sYtcP6GJ7IUrRMpDmEaf8RRqiEhBryCRt8ECh9LkvPIG7 > O+WytW0PIXMLyHWMbGqu3d1PknkDtk2OXcyG2TFqgQqcQxK7x0DtMTBddOwu4OPB > Dfq8YHEriTA1K2y3GgewC3OJcrnmwCYqeu0vnfQSawf9bScLh3PgG4M9tFhE34lk > nNBXaa1i/H2tkQNPf9pA1uvrCdDLBKoR+IwWXTzCFzWXPgb21FwJ0nCNoWfreKBE > sGvIliCuo81L4RRIjDQd/UZendyyDNrzP9bnzxPu+Upq2eIFSd995IymwepEiA41 > v/Vn27qzEdPKTwiixDpF1q/LOFs2u4yYUhVlcJ85qZyoJcqVhfQGq6SJS4U= > =ScN/ > -----END PGP SIGNATURE----- > >

1 0

Thinkpad [3]: Aligning size of dm-crypt and partition/filesystem
by Matt Price 16 Jun '16

16 Jun '16

Final question! In order to enable suspend to disk, I ended up shrinking my root partition (I also added an M.2 SSD, so the whole thing was a bit complex). In the end I got it to work, but I made some kind of an error and currently my on-disk “crypt” is a little larger than the partition and file system that I am using for root, so I have some wasted space on the disk: ➜ ~ sudo fdisk /dev/sda Command (m for help): p Disk /dev/sda: 238.5 GiB, 256060514304 bytes, 500118192 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: gpt Disk identifier: 9B4C2E92-6C7D-4C69-B67B-1823E4DB9BA6 Device Start End Sectors Size Type /dev/sda1 2048 1026047 1024000 500M EFI System /dev/sda2 1026048 106907647 105881600 50.5G Linux filesystem /dev/sda3 106907648 483340287 376432640 179.5G Linux LVM /dev/sda4 483340288 500117503 16777216 8G Linux swap ➜ ~ sudo cryptsetup status cryptsetup /dev/mapper/cryptsetup is active and is in use. type: LUKS1 cipher: aes-xts-plain64 keysize: 256 bits device: /dev/sda2 offset: 4096 sectors size: 105877504 sectors mode: read/write Anyone out there have advice abot how to safely and reliably expand the partition and the file system so that they both end precisely where the crypt does? Thanks again guys! as always, it’s much appreciated.

2 2

Re: [GTALUG] mysterious restarts
by Russell Reiter 15 Jun '16

15 Jun '16

Sent from mobile. On Jun 15, 2016 12:29 PM, "Bob Jonkman via talk" <talk(a)gtalug.org> wrote: > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Russell writes: > > Take a look at a trolly pole then a pantograph array and tell me > > the pantograph needs less maintainence till EOL > > It's not the pantograph that requires less maintenance, it's the > overhead wires. A single So perhaps the increased array maintenance balances out against the reduced wire maintenance. This leaves the main feature of the system as stated to be, as its increased voltage capacity. trolley pole requires the overhead wire to > have switches (like track switches, not electrical switches) at > intersections and any location where the streetcar can choose a > direction. Very often there are disconnections as the trolley pole > moves from one wire to another. That can't be good for the circuitry. > And how often does the trolley pole come disconnected from the > overhead wire? > > OTOH, the wires for the pantograph can be continuous, there can be > multiple wires at intersections, and there are far fewer > disconnections, intentional or accidental. But still there is the cost of the maintenance stations for the array. I've seen trolly pole's disconnect and be reconnected by the driver in a minutes time. I don't know how the pole contacts are maintained but the graphite bar maintenance of the pantographic grid is a stated weakness of the system. > > - --Bob. > > > On 2016-06-15 11:01 AM, Russell Reiter via talk wrote: > > On 6/15/16, Lennart Sorensen <lsorense(a)csclub.uwaterloo.ca> wrote: > >> On Tue, Jun 14, 2016 at 05:57:44PM -0400, Russell Reiter wrote: > >>> I don't doubt it at all. I don't understand not taking > >>> advantage of the pantographic ability to carry higher voltage; > >>> that is its primary feature. > >> > >> Actually the primary feature is less maintenance, simpler wiring, > >> and higher speed supported (not that you need that in Toronto > >> streets). > > > > Take a look at a trolly pole then a pantograph array and tell me > > the pantograph needs less maintainence till EOL. You've never heard > > of a trolly pole monitoring station, there is a pantograph > > monitoring station though. It's higher voltage capacity that is the > > reason this form factor takes dominance in overhead wiring. > > > > As you said higher vehicle speeds are a moot point in Toronto. > > > >> > >>> That would be like buying a Porsche and detuning the turbo, > >>> just doesn't follow common sense. > >> > >> Well all that really matters is how much power they can deliver, > >> so voltage * current. Higher voltage means less current, but > >> also bigger insulators needed. It's a tradeoff. Given they > >> already have 600V equipment, sticking with that makes sense. > >> After all increasing the insulator length might not be practical > >> based on the height of the mounting point for the current wires. > >> > >>> On the other hand, for future capacity planning while load > >>> balancing the polyphase AC grid we draw that energy from, the > >>> advantage of inverting to higher DC voltage under higher > >>> demand, does make sense. > >> > >> Well I am quite sure they are running AC motors, so they would be > >> taking the DC and inverting it to AC and making it whatever > >> voltage they need at the time for the motors. > > > > So doesn't economy of scale account for the sensibility of > > starting with a higher DC voltage when part of the load is inverted > > back to AC for motors? I'm assuming the onboard solid state keeps > > the input at DC, rectifying it again would go beyond the pale. > > > >> > >> -- Len Sorensen > >> > > --- Talk Mailing List talk(a)gtalug.org > > https://gtalug.org/mailman/listinfo/talk > > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v2 > Comment: Ensure confidentiality, authenticity, non-repudiability > > iEYEARECAAYFAldhglMACgkQuRKJsNLM5ervBACgjVBKk8woXmXYcQrMAaAB+Yq2 > yXEAninGZStjLXcxcm51FYw8t480jEeN > =DEgv > -----END PGP SIGNATURE----- > --- > Talk Mailing List > talk(a)gtalug.org > https://gtalug.org/mailman/listinfo/talk

1 0

Speaker for July 2016 Meeting
by Myles Braithwaite 15 Jun '16

15 Jun '16

We are looking for a speaker for 12 July 2016 meeting. Anyone interested in giving a talk?

1 0

Last Night's Meeting Annoucments
by Myles Braithwaite 15 Jun '16

15 Jun '16

<https://wiki.gtalug.org/meeting:2016-06> # Linux in the Park 2016, **23 July** All planning is being done in the wiki @ <https://wiki.gtalug.org/event:linux_in_the_park_2016> # Upcoming Events ## June 27 June 2016 at 7pm - Damian Conway -- Perl 6: More Concurrency, Less Contusion Even if you have no interest in Perl, Damian is always interesting. <http://www.meetup.com/Toronto-Perl-Mongers/events/231418224/> 28 June 2016 at 7pm - Damian Conway -- On the Shoulders of Giants: 400 Years of Perl 6 <http://www.meetup.com/Toronto-Perl-Mongers/events/231418278/> ## July 27 July 2016 at 6pm - BOSH 101 with David Ing Our August 2014 meeting was on BOSH so if you are looking for a refresh you should definitely check this out. <http://www.meetup.com/Toronto-Pivotal-User-Group/events/227459952/> ## August 6 August 2016 at 6pm - Dan Harrison on his battles and winnings creating a SalesForce client for Pebble Watch <http://www.meetup.com/PebbleTO/events/224057584/>

1 0

Thinkpad [2] : Power malfunction on suspend; faster suspend to disk?
by Matt Price 15 Jun '16

15 Jun '16

Another thinkpad issue, this one more serious: when I suspend to ram, my Thinkpad T440s sometimes has a serious power malfunction. While suspend generally succeeds, at some point the pulsing power light on the lid goes out and it’s no longer possible to power on the laptop. I have to open the lid, disconnect the internal bettery, and generally also hit the internal power reset button. I can then restart, but of course lose all my work. At first this happened about once a month, then once a week, and now it happens several times a day. 1. Anyone else having this issue? And do you know what actually causes it, and if it can be fixed? I’ve searched online and there are various threads about it, but I haven’t found any actionable information. Recently updated my BIOS and changed some BIOS settings, but to no apparent effect. 2. As a workarond, I’ve just switched from suspend-to-ram to hybrid suspend, which writes an image to disk before suspending to ram. (Took me a long time to do this, because I had no swap partition & my main partition was encrypted, so I had to resize the encrypted partition, ettc.) This seems to work! Yay! However, suspend-to-disk is really slow. In the old days I used Txonice for hibernaion, but that project has really slowed down in the last few years. Is anyone still using it? How about uswsusp? Is either one compatible with the systemd hibernation commands? My experience with tuxonice was that it was abot 5 times faster than native hibernation, so I’d really like to compress the disk image Thanks again! Matt

1 0

Thinkpad [1] : Twitchy lid close switch
by Matt Price 15 Jun '16

15 Jun '16

Hi Folks, It’s been a long time since I participated i n the group, hope everyone’s well! I have Arch installed on a Thinkpad T440s; it’s mostly worked well for quite a while, but I’m running into some problems; separating them out into individual emails for clarity. The simplest issue is a very sensitive lid close sensor. The sensor is inside the lid hinge, and it often registers a state change when I adjust the screen angle. So, two questions: - any other THinkpad users to there seeing this, and if so, have you figured out a hardware fix? - failing that, does anyone have a suggestion for a software fix – say, a custom script that systemd can run as a hook before the LidSwitch event fires? I don’t know much about the systemd power management features. I noticed the “HoldoffTimeoutSec” variable in /etc/systemd/logind.conf, and currently have it set to 30s (cf. https://www.freedesktop.org/software/systemd/man/logind.conf.html) But it hasn’t yet solved my problem. As usual, thanks so much for your help! Matt

1 0

40" 4k Philips Monitor for $699.95 at BestBuy, Jun. 8 and 9 only
by CLIFFORD ILKAY 14 Jun '16

14 Jun '16

Hello, I had bought a Wasabi Mango 42" monitor from a Korean vendor on eBay last May and when it broke down for the second time within the first 13 weeks, neither the vendor nor Wasabi Mango provided warranty support. Fortunately, I had bought a SquareTrade warranty and they came through and reimbursed me for the full purchase amount. Orchids for SquareTrade, onions for Wasabi Mango and gnestore on eBay, and onions for eBay's useless reputation system that prevents buyers from leaving reviews after 90 days from purchase. All that does is provide an incentive for the seller to be responsive within the first 90 days and not even pretend to care once the buyer can no longer leave a negative review. Just to give you an idea of how ill-equipped Wasabi Mango must be to provide after-sale support and how little they care about their reputation, I left a scathing review of their monitor on their Facebook page and they never responded and the last time I checked some months ago, it was still there for the whole world to see. I bought this same 40" 4k Philips monitor for $699.99 at a Black Friday sale last year at BestBuy and it has been working great(*). The Wasabi Mango was an LG IPS panel. This one is a VA panel. Theoretically, the IPS panel is supposed to be "better" but I really do not have any objections to the video quality on the Philips. I have not noticed any dead pixels and the brightness is uniform throughout. It has an easy-to-use joystick control and a hard power switch. I have an ASUS GT 960 Strix video card running the nVidia binary driver on Fedora 23 driving the monitor through a DisplayPort cable. (*) Once every few months, I have to update the binary driver (it's not automatically updated) when I lose X after a kernel update. Occasionally, I'll see a flickering of a window that is not in the foreground. If I move the mouse, the flickering will disappear, which has me suspect the video driver and not the monitor. It's not frequent enough to be anything more than a minor irritant. Neither of those things are particular to this monitor. With a panel this large, I have not found any need for font-scaling. There is no way I could read text on a 28" 4k monitor without font-scaling. It would be hard to beat the price/performance of this monitor and with the weaker Canadian dollar, monitor prices have been firm. You will not see this monitor at this price very often. In short, it's a great buy if you've been thinking of upgrading to a 4k monitor. -- Regards, Clifford Ilkay + 1 647-778-8696

13 31