I came to the problem that I needed to test native linux/unix virtualisation recently. First, I tried to think of the new test box I have to procure, but then I got the idea that I'm not using my current resources to the full: my MacBook Pro i5 8GB RAM. Found this (see a copy in my Google Drive). It was really easy!
All you need is to do reEFInd bot manager and install Ubuntu afterwards. Worked like a charm! And it is very nice performance of Ubuntu on this metal :)
The only annoyance was that my computer stopped to do sleep normally (from OS X perspective). It needed hard reboot every time after a few hours of sleep. Fortunately, upgrade to Yosemite solved this, probably just a fresh reinstall of anything would have helped.
Enjoy!
Friday, 24 October 2014
Saturday, 27 September 2014
shellshock on my systems - no real worries
To determine if your Linux or Unix system is vulnerable, from a command line, type something like this:
env x=’() { :;}; echo vulnerable’ bash -c “echo this is a test”
If the system is vulnerable, the output will be:
vulnerable
this is a test
An unaffected (or patched) system will output:
bash: warning: x: ignoring function definition attempt
bash: error importing function definition for `x’
this is a test
or it will say that bash is not present or something else, you figure it out :)
So what about my systems? Here's the list:
env x=’() { :;}; echo vulnerable’ bash -c “echo this is a test”
If the system is vulnerable, the output will be:
vulnerable
this is a test
An unaffected (or patched) system will output:
bash: warning: x: ignoring function definition attempt
bash: error importing function definition for `x’
this is a test
or it will say that bash is not present or something else, you figure it out :)
So what about my systems? Here's the list:
- pfSense firewall 2.1.4-RELEASE (amd64), built on Fri Jun 20 12:59:50 EDT 2014, FreeBSD 8.3-RELEASE-p16 : not vulnerable, bash is not there by default :) check /etc/shells if you have it installed
- FreeNAS home server FreeNAS-9.2.1.7-RELEASE-x64 (fdbe9a0), FreeBSD 9.2-RELEASE-p10 #0 r262572+4fb5adc: Wed Aug 6 17:07:16 PDT 2014 : bash present by defeult, system vulnerable, checked jails, bash not present, will wait for the fix in next version, there's good comment on this at the end of this discussion :)
- Mac Darwin Kernel Version 13.3.0: Tue Jun 3 21:27:35 PDT 2014; root:xnu-2422.110.17~1/RELEASE_X86_64 x86_64 : vulnerable, (I think) I haven't played with enabling/opening things, so will be waiting on Apple :)
- Hackintosh Darwin Kernel Version 13.3.0: Tue Jun 3 21:27:35 PDT 2014; root:xnu-2422.110.17~1/RELEASE_X86_64 x86_64 : vulnerable, (I know) I haven't played with enabling/opening things, so will be waiting on Apple
- XBMC based HTPC on Ubuntu 14.04.1 LTS, Linux ___ 3.11.0-13-generic #20-Ubuntu SMP Wed Oct 23 07:38:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux : not vulnerable
- I haven't hacked into my TV and I won't :)
Good luck to you all!
Labels:
apple,
bash,
chellshock,
firewall,
freebsd,
freenas,
hackintosh,
htpc,
linux,
mac,
pfsense,
ubuntu,
unix,
vulnerability
Location:
Salaspils, Salaspils pilsēta, Latvia
Thursday, 12 June 2014
moved my htpc to raspberry pi... and back to "normal" pc
After using Raspberry Pi based HTPC with Raspbian for a while (see my previous posts) it was obvious that it hasn't made it for me. A few reasons:
I definitely could try to tune the thing or try other distros. Maybe I hoped that it will be more out-of-the-box initially. Maybe I'm not that much enthusiast or just too lazy :)
It definitely has the future, it just isn't here for me yet.
So, I'm back on my old good ASUS AT5IONT-I motherboard with pico-psu, 4GB of RAM, SSD for system disk. Good Ubuntu plus XBMC.
Some pics for the dessert.
- Trying to watch browser-based internet TV or just videos in the browser is too much hustle. It was much easier to connect my Mac and watch it.
- AirPlay is lagging on videos.
- Small one - for some reason it kept changing IP address on every reboot / power loss. There were more than a few. Btw, if you have your Raspberry Pi unstable, the first move definitely is to take a look at your PSU. It got much better for me when I changed PSU from a noname to iPhone charger. Google it before you buy.
- For me personally the combination with my receiver was not good: Pi supports HDMI starting version 2, but my receiver's "old" one from year 2008 and supports just 1.1 :D Sorry, no 5.1 sound with those beautiful Dolby Digitals, DTS's etc for me. I was thinking of buying external sound card, but, you know, why should I when I have some PC boards laying around the house having sound built-in with more "usual" connectors, like optical S/PDIF maybe?
- And, the most annoying one which was the final killer: user interface is too slow / lagging. I can't imagine why the distro can decode 1080p normally and can't do smooth user interface :| Especially, if you have a list of multimedia longer than one screen you wait for each next one, and next, and next...
I definitely could try to tune the thing or try other distros. Maybe I hoped that it will be more out-of-the-box initially. Maybe I'm not that much enthusiast or just too lazy :)
It definitely has the future, it just isn't here for me yet.
So, I'm back on my old good ASUS AT5IONT-I motherboard with pico-psu, 4GB of RAM, SSD for system disk. Good Ubuntu plus XBMC.
Some pics for the dessert.
Top view while still open |
Open side view, wires |
Top-front, open |
Before attaching front grille |
Assembled, with "monster" fan |
Charger |
Whole thing |
Installed and running :) |
Labels:
1080p,
airplay,
asus,
browser,
decode,
dolby digital,
dts,
hdmi,
htpc,
internet tv,
lagging,
psu,
raspberry pi,
raspbmc,
receiver,
s/pdif,
slow,
ubuntu,
user interface,
xbmc
Location:
Salaspils, Salaspils pilsēta, Latvia
Wednesday, 26 March 2014
moved my htpc to raspberry pi - couple of my answers to my own questions
Basically, there is no _practical_ WoL for the Raspberry Pi. It seem technically possible, at least they say that the NIC+USB chip HW supports it, but nobody f...ing needs it, so no project cares to make it. Everybody seem so happy about having a device consuming 5W max that leaving it on all the time seem more than normal even if it does productive work just a fraction of the time. Ok, let the time solve this, I'm happy too :)
Btw, there is WoL functionality for XBMC and their mobile remote. So, guys using it on the HW consuming tens of watts of energy, I encourage to try and to use it!
Cool feature of XBMC, Apple's AirPlay protocol, doesn't seem to work stable on Raspbmc. First, I was surprised by the functionality, you can't put the screen on the screen, you can just share photos, videos, music. Second, video is _very_ unstable and there are screen stretching issues which doesn't seem trivial to solve, you need to really dig into configuring the thing. I do not recommend it for now.
Btw, I had 1 Raspbmc crash since start, it just stopped to play internet radio once without obvious reason.
Otherwise, I'm still very happy with it.
Btw, there is WoL functionality for XBMC and their mobile remote. So, guys using it on the HW consuming tens of watts of energy, I encourage to try and to use it!
Cool feature of XBMC, Apple's AirPlay protocol, doesn't seem to work stable on Raspbmc. First, I was surprised by the functionality, you can't put the screen on the screen, you can just share photos, videos, music. Second, video is _very_ unstable and there are screen stretching issues which doesn't seem trivial to solve, you need to really dig into configuring the thing. I do not recommend it for now.
Btw, I had 1 Raspbmc crash since start, it just stopped to play internet radio once without obvious reason.
Otherwise, I'm still very happy with it.
Monday, 24 March 2014
moved my htpc to raspberry pi
Today I unwillingly decided to change my HTPC. It happened by the combination of two events.
First, I was thinking to add a Raspberry Pi to my multi-room audio system for some time already in order to stream saved music and internet radio and manage it by the mobile devices.
Second, I mixed up PSU's between above mentioned multi-room audio system and current HTPC after dusting them. Bad thing, multi-room has higher voltage (24V instead of 12-17V) than HTPC, so HTPC decided to get to be offended. I don't know what exactly broke, still need to find what got burned there. Good thing, I had urgent need for a new HTPC on my hands, thus motivation to act immediately.
NB It's a pity that you actually can't rely on all those different PSU and charger producers that plugs doesn't match...
So, XBMC on Raspberry Pi it is. Quickly googled and decided on Raspbmc. Very simple setup, just download it, dd it to flash card, plug it in and let it setup itself. After that you get your usual XBMC interface.
XBMC official iPhone app is even easier. And it's interface is better than XBMC's default one. Just easiness of the search function tells.
Btw, if you connect your Raspbmc to the network server for the media, be aware that XBMC may not be very efficient in the networking by default. My config stuttered HD videos by default via Samba and NFS. This helped my thinking, optimised it to flawless playback :)
Now, some observations and, unfortunately, pitfalls, even thou hopefully short-term:
So, todo for me now is to:
Guys, if anybody wants to buy used medium-class AVR in a good condition, just let me know. It's a Marantz SR4001, just looked inside, there is minimal dust and capacitors look good.
First, I was thinking to add a Raspberry Pi to my multi-room audio system for some time already in order to stream saved music and internet radio and manage it by the mobile devices.
Second, I mixed up PSU's between above mentioned multi-room audio system and current HTPC after dusting them. Bad thing, multi-room has higher voltage (24V instead of 12-17V) than HTPC, so HTPC decided to get to be offended. I don't know what exactly broke, still need to find what got burned there. Good thing, I had urgent need for a new HTPC on my hands, thus motivation to act immediately.
NB It's a pity that you actually can't rely on all those different PSU and charger producers that plugs doesn't match...
So, XBMC on Raspberry Pi it is. Quickly googled and decided on Raspbmc. Very simple setup, just download it, dd it to flash card, plug it in and let it setup itself. After that you get your usual XBMC interface.
XBMC official iPhone app is even easier. And it's interface is better than XBMC's default one. Just easiness of the search function tells.
Btw, if you connect your Raspbmc to the network server for the media, be aware that XBMC may not be very efficient in the networking by default. My config stuttered HD videos by default via Samba and NFS. This helped my thinking, optimised it to flawless playback :)
Now, some observations and, unfortunately, pitfalls, even thou hopefully short-term:
- Major one for me for now. If you take a look at the Raspberry Pi specs you will NOT notice that it supports HDMI revisions 1.3 and 1.4. My audio-video receiver (AVR) which I bought in 2008 supports 1.1 and back. Sorry guys, no backwards compatibility, it won't work, at least it doesn't for me. I noticed that HDMI is up to 2.0 now. So, now I'm thinking about changing the AVR, but those versions...
- Btw, talking HDMI revisions, my TV, bought the same year, is using HDMI revision 1.3. Conclusion: Philips was ahead of Marantz back in 2008 :D
- Annoyances. So, I can't put HDMI through AVR. It IS a pity because of the sound. Video I can put directly to TV. Of course, I could use just TV sound, but my Marantz's sound is sooooo much better. Fortunately, Raspberry Pi has a simple analog audio 3.5mm jack, so I can have a good sound. And you can switch on audio output to both HDMI and analog ports in XBMC. The annoyance: not all those digital wonders of nowadays like Dolby Digital or DTS, just a plain simple stereo. So, it works. The annoyance: I need one more analog audio output for the multi-room, which I haven't. And one more for the headphones... And I couldn't use my AVR for switching even if HDMI revisions were OK, because my AVR doesn't convert digital sound to analog (ask for this when buying an AVR).
- One of the best things happened. I'm going to save some electricity definitely: Raspberry Pi can't go much over 5W :) Previous HTPC's PSU was at 110W max and I'm quite sure it used 50W to max.
- There is no power-on button on Raspberry Pi :D You can switch it OFF from software. What about ON? WoL you say? WoL is much, much more sophisticated... And my children are so used to use a power button. For switching on, at least, because switching something off is what older people do, isn't it?
So, todo for me now is to:
- find a soundcard for my Raspberry Pi, that would solve my DD-and-DTS-like "problems", plus offload the processor of decoding those, I hope
- buy 3.5mm audio jack splitter
- change my AVR
- invent a power button for Raspberry Pi or configure WoL (from iPhone app should it work)
- or all of those options...
Guys, if anybody wants to buy used medium-class AVR in a good condition, just let me know. It's a Marantz SR4001, just looked inside, there is minimal dust and capacitors look good.
Labels:
audio,
avr,
electricity,
hdmi,
htpc,
iphone,
multi-room audio,
music,
nfs,
power,
psu,
raspberry pi,
raspbmc,
receiver,
video,
wol,
xbmc
Location:
Salaspils, Salaspils pilsēta, Latvia
Monday, 17 March 2014
retired my cisco linksys e3000 as router/firewall, moved to pfSense
For some time I felt like my home Cisco Linksys e3000 wireless router was not able to perform according to requirements of today's traffic at home. There are four of us, everybody use Internet, we have 20+ devices with IP address, so it could be quite normal that signs of overload appeared more and more often. Slowness sometimes, sometimes router even hung.
Historically, I tried out dd-wrt project, which has great functionality, but stability of the router brought me back to default Cisco firmware. Recently even default firmware was not good enough.
One of the steps I've taken already was switching DHCP and DNS servers off and giving those functions to a Raspberry Pi. But that was more because I wanted to try out Pi :)
I've been thinking about a change to more powerful router, but somehow took a look at my "server room" wall, where one of two Intel Atom D525MW boards hung without a proper job and decided to take a look at what opensource is doing in router / firewall space.
I found very nice review by David Pavlina which led me to choose between pfSense and Sophos UTM Home Edition (scroll down a bit). pfSense looked better to me just for one reason: it's not limited version of commercial SW like Sophos. So let's try it.
First of all I was quite excited because it was first time I was going to try something that is based on FreeBSD project. And I have to say that first impression - install - was very good. I'll see later what it takes to do basic OS things there.
Install of pfSense is quite straightforward and easy.
Initially I tried to put it on USB stick, but I gave it up after several iterations of failed formatting and partitioning... Or should I say: slicing? :D As I had spare SSD on shelf I decided that USB I will try some other time.
Initial config is trivial. Basically, you just need to notice NIC names and feed those back to pfSense as you decide on WAN and LAN side. Autodiscovery didn't work for me, but who cares when manual one is so easy, right?
Connected to webConfigurator interface, nice one. I took a look at menu and saw there all I could imagine I would need from home firewall/router. Except for QoS, but then googled that this is named Traffic Shaping here and and decided to go further with it.
I put in basic things like hostname, domain, etc, switched off DHCP and DNS for now, checked that firewall is on (I have to say "once more", because during install pfSense told that it will be on) and get to putting actual wires into NIC's.
NB It is always interesting to come up with a new hostname. I will not publish actual one, but my thinking came out of this terrible animal, watch out for:
It took me some time to redo wiring, now it looks even more terrible than previously, but this absolutely is a matter of the another project :) Plus, switching my e3000 to "just wireless access point" also took a few hard resets because it kept becoming unavailable after I assigned it a "normal" IP to it. Obviously, didn't want to give up being router and firewall :D
So, here I am, typing this blog in the new network :)
Now, just switch DHCP and DNS servers from Raspberry Pi to new server. DHCP works like out-of-the-box :) DNS forwarder (the same dnsmasq actually) doesn't want to resolve FDQN for some reason :( But this shouldn't be a big problem, I just need to get to be used to the new one. And DNS is one of the most weird-built things in IT as you all know - already or just now.
Historically, I tried out dd-wrt project, which has great functionality, but stability of the router brought me back to default Cisco firmware. Recently even default firmware was not good enough.
One of the steps I've taken already was switching DHCP and DNS servers off and giving those functions to a Raspberry Pi. But that was more because I wanted to try out Pi :)
I've been thinking about a change to more powerful router, but somehow took a look at my "server room" wall, where one of two Intel Atom D525MW boards hung without a proper job and decided to take a look at what opensource is doing in router / firewall space.
Does this wiring look ok? No way... |
I found very nice review by David Pavlina which led me to choose between pfSense and Sophos UTM Home Edition (scroll down a bit). pfSense looked better to me just for one reason: it's not limited version of commercial SW like Sophos. So let's try it.
First of all I was quite excited because it was first time I was going to try something that is based on FreeBSD project. And I have to say that first impression - install - was very good. I'll see later what it takes to do basic OS things there.
Install of pfSense is quite straightforward and easy.
Initially I tried to put it on USB stick, but I gave it up after several iterations of failed formatting and partitioning... Or should I say: slicing? :D As I had spare SSD on shelf I decided that USB I will try some other time.
Initial config is trivial. Basically, you just need to notice NIC names and feed those back to pfSense as you decide on WAN and LAN side. Autodiscovery didn't work for me, but who cares when manual one is so easy, right?
Connected to webConfigurator interface, nice one. I took a look at menu and saw there all I could imagine I would need from home firewall/router. Except for QoS, but then googled that this is named Traffic Shaping here and and decided to go further with it.
I put in basic things like hostname, domain, etc, switched off DHCP and DNS for now, checked that firewall is on (I have to say "once more", because during install pfSense told that it will be on) and get to putting actual wires into NIC's.
NB It is always interesting to come up with a new hostname. I will not publish actual one, but my thinking came out of this terrible animal, watch out for:
Regal horned lizard (Phrynosoma solare) |
It took me some time to redo wiring, now it looks even more terrible than previously, but this absolutely is a matter of the another project :) Plus, switching my e3000 to "just wireless access point" also took a few hard resets because it kept becoming unavailable after I assigned it a "normal" IP to it. Obviously, didn't want to give up being router and firewall :D
So, here I am, typing this blog in the new network :)
Now, just switch DHCP and DNS servers from Raspberry Pi to new server. DHCP works like out-of-the-box :) DNS forwarder (the same dnsmasq actually) doesn't want to resolve FDQN for some reason :( But this shouldn't be a big problem, I just need to get to be used to the new one. And DNS is one of the most weird-built things in IT as you all know - already or just now.
Labels:
atom,
d525mw,
dd-wrt,
dhcp,
dns,
dnsmasq,
e3000,
firewall,
freebsd,
hostname,
linksys,
opensource,
performance,
pfsense,
raspberry pi,
router,
slow,
sophos,
usb,
wifi
Location:
Salaspils, Salaspils pilsēta, Latvia
Thursday, 6 March 2014
cybersecurity day today for me #security #linux #truecrypt #opensource #keychain #apple #keychain2go
All of the following in one day (+night...)
Twitter told me that Critical Linux Flaw Threatens More Systems Than You Think today.
Updated my Linux-es tonight (Ubuntu server 13.10 and raspbian), noticed that simple update doesn't affect package in question :D Need to be persistent and upgrade package from source...
And left my XBMC box for the other day, it also runs on Ubuntu.
As mounting Apple's encrypted disk failed me last week plus it makes duplicate copies in my Dropbox, decided to look for alternatives. Moved to open software, namely TrueCrypt.
Interesting and convincing. Plus, now I can view it on my mobile iOS devices too :)
I used to store some passwords in some files... Finally it got me, too hard to manage and search. Need some password manager. For me as Apple user the obvious first choice was keychain plus Keychain2Go app to get it on iOS devices. Set up sync, nice! :) Now need to move passwords there gradually. The question remains, what about being dependent on Apple? Noticed 2 nice things about it: keychain is in Apple's opensource domain actually (!) and there are tools for export available. So, some day there will be integration.
Twitter told me that Critical Linux Flaw Threatens More Systems Than You Think today.
Updated my Linux-es tonight (Ubuntu server 13.10 and raspbian), noticed that simple update doesn't affect package in question :D Need to be persistent and upgrade package from source...
And left my XBMC box for the other day, it also runs on Ubuntu.
As mounting Apple's encrypted disk failed me last week plus it makes duplicate copies in my Dropbox, decided to look for alternatives. Moved to open software, namely TrueCrypt.
Interesting and convincing. Plus, now I can view it on my mobile iOS devices too :)
I used to store some passwords in some files... Finally it got me, too hard to manage and search. Need some password manager. For me as Apple user the obvious first choice was keychain plus Keychain2Go app to get it on iOS devices. Set up sync, nice! :) Now need to move passwords there gradually. The question remains, what about being dependent on Apple? Noticed 2 nice things about it: keychain is in Apple's opensource domain actually (!) and there are tools for export available. So, some day there will be integration.
Labels:
cloud,
cybersecurity,
dropbox,
encryption,
gnutls,
ios,
keychain,
keychain2go,
linux,
opensource,
password manager,
passwords,
raspberry pi,
raspbian,
security,
ssh,
truecrypt,
ubuntu,
vulnerability,
xbmc
Location:
Salaspils, Salaspils pilsēta, Latvia
Thursday, 6 February 2014
my favorite and most important tech fact of year 2013 : bug #1 solved
It took some time for me to level all my tech interests, preferences, readings, writings, happenings of the year 2013. But now I'm done: my favorite and most significant tech fact of the year is this: Mark Shuttleworth marked bug #1 as fixed!
If you didn't know, the bug #1 was "Microsoft has a majority market share".
Read it here https://bugs.launchpad.net/ubuntu/+bug/1/comments/1834 or copy just for my recods:
----
Mark Shuttleworth (sabdfl) wrote :
----
If you didn't know, the bug #1 was "Microsoft has a majority market share".
Read it here https://bugs.launchpad.net/ubuntu/+bug/1/comments/1834 or copy just for my recods:
----
Mark Shuttleworth (sabdfl) wrote :
----
Personal computing today is a broader proposition than it was in 2004: phones, tablets, wearables and other devices are all part of the mix for our digital lives. From a competitive perspective, that broader market has healthy competition, with IOS and Android representing a meaningful share (see http:// www.zdnet. com/windows- has-fallen- behind- apple-ios- and-google- android- 7000008699/ and in particular http:// cdn-static. zdnet.com/ i/r/story/ 70/00/008699/ meeker620- 620x466- 620x466. jpg?hash= ZQxmZmDjAz& upscale= 1).
Android may not be my or your first choice of Linux, but it is without doubt an open source platform that offers both practical and economic benefits to users and industry. So we have both competition, and good representation for open source, in personal computing.
Even though we have only played a small part in that shift, I think it's important for us to recognize that the shift has taken place. So from Ubuntu's perspective, this bug is now closed.
There is a social element to this bug report as well, of course. It served for many as a sort of declaration of intent. But it's better for us to focus our intent on excellence in our own right, rather than our impact on someone else's product. In the (many) years since this bug was filed, we've figured out how to be amazing on the cloud, and I hope soon also how to be amazing for developers on their desktops, and perhaps even for everyday users across that full range of devices. I would rather we find a rallying call that celebrates those insights, and leadership.
It's worth noting that today, if you're into cloud computing, the Microsoft IAAS team are both technically excellent and very focused on having ALL OS's including Linux guests like Ubuntu run extremely well on Azure, making them a pleasure to work with. Perhaps the market shift has played a role in that. Circumstances have changed, institutions have adapted, so should we.
Along those lines, it's good to reflect on how much has changed since 2004, and how fast it's changed. For Ubuntu, our goal remains to deliver fantastic experiences: for developers, for people building out production infrastructure, and for end-users on a range of devices. We are doing all of that in an environment that changes completely every decade. So we have to be willing to make big changes ourselves - in our processes, our practices, our tools, and our relationships. Change this bug status is but a tiny example.
Location:
Salaspils, Salaspils pilsēta, Latvia
Sunday, 2 February 2014
spinning down / standby HDD's on Ubuntu server 13.10 (to be continued)
Ubuntu server 13.10 with 7 disks, 1 sdd for system, 4 x 2 TB in raid10 and 2 x 1 TB mirrored.
Tried out pure hdparm thing first. Installed hdparm, edit /etc/hdparm.conf adding lines for each disk, switched ACHI in BIOS, reboot. Didn't work for me. Tried manually, didn't work either.
Found this post, made some progress. Saw at least one disk being standby. As the intention was to try the thing "as is", meaning I didn't disconnect server from network and didn't try to figure what can be using the disks, fine. Will take a look later / sometime what's the status of disks.
Made a shell script for displaying the status of all disks. Thought about emailing myself status from time to time, started to read about emailing from command line, but there was a lot to read, remembered that this is lazy weekend and dropped it this time.
----
Just to document the post in case the thing go away:
1. Install sg3-utils
2. Add the following to crontab (sda being the disk to be managed this way):
0-59/15 * * * * ( if [ ! -f /dev/shm/1 ] ; then touch /dev/shm/1 /dev/shm/2; fi ; mv /dev/shm/1 /dev/shm/2; cat /proc/diskstats > /dev/shm/1 ) >/dev/null 2>&1
Tried out pure hdparm thing first. Installed hdparm, edit /etc/hdparm.conf adding lines for each disk, switched ACHI in BIOS, reboot. Didn't work for me. Tried manually, didn't work either.
Found this post, made some progress. Saw at least one disk being standby. As the intention was to try the thing "as is", meaning I didn't disconnect server from network and didn't try to figure what can be using the disks, fine. Will take a look later / sometime what's the status of disks.
Made a shell script for displaying the status of all disks. Thought about emailing myself status from time to time, started to read about emailing from command line, but there was a lot to read, remembered that this is lazy weekend and dropped it this time.
----
Just to document the post in case the thing go away:
1. Install sg3-utils
2. Add the following to crontab (sda being the disk to be managed this way):
0-59/15 * * * * ( if [ ! -f /dev/shm/1 ] ; then touch /dev/shm/1 /dev/shm/2; fi ; mv /dev/shm/1 /dev/shm/2; cat /proc/diskstats > /dev/shm/1 ) >/dev/null 2>&1
0-59/15 * * * * ( export HD="sda "; if [ "$(diff /dev/shm/1 /dev/shm/2 | grep $HD )" = "" ] ; then /usr/bin/sg_start --stop /dev/$HD; fi ) >/dev/null 2>&1
Location:
Salaspils, Salaspils pilsēta, Latvia
Wednesday, 22 January 2014
my first experience in programming
I don't remember exactly, but it has to be year 1982 when I got my hands on Elektronika B3-21 (Электроника Б3-21). Later on it was more advanced Elektronika B3-34 (Электроника Б3-34), both programmable by assembler-like coding. I was enchanted by those devices for months I spent trying out examples of code written by others and messing around with it trying to do something myself. I think the concept of "lets restart/erase everything and start over" came to me naturally back then :)
Labels:
calculator,
first experience,
programming,
restart
Location:
Salaspils, Salaspils pilsēta, Latvia
quote: it's time for mobile to grow up, publishers need to feed people's appetites for mobile [or go away]
Source
Tieu: "It's time for mobile to grow up. In 2014 publishers will discover that they need to go all in for mobile. It is not enough to just have an Android app or an iOS app or to be optimized for the mobile web. Designing for a single device is not going to fly. Publishers need to offer mobile to consumers on every device so that readers can engage when and where they please. People today are digital omnivores, and publishers need to feed their appetites across the varying mobile outlets."
Tieu: "It's time for mobile to grow up. In 2014 publishers will discover that they need to go all in for mobile. It is not enough to just have an Android app or an iOS app or to be optimized for the mobile web. Designing for a single device is not going to fly. Publishers need to offer mobile to consumers on every device so that readers can engage when and where they please. People today are digital omnivores, and publishers need to feed their appetites across the varying mobile outlets."
Labels:
android,
digital omnivores,
ios,
mobile,
mobile web,
optimised,
publishers
quote: protecting network perimeter approach starts to go away
Source
Jouret: “In the coming 1-3 years, we will see a new form of multi-layered security emerge, replacing today’s model of “protecting the perimeter” with a combination of security technologies that includes localized clients embedded within devices or localized connections and centralized cloud-based intelligence which constantly scans to protect. The connected car will happen; we just have to care about the connected car and develop strategies so that we don’t get hacked.”
Jouret: “In the coming 1-3 years, we will see a new form of multi-layered security emerge, replacing today’s model of “protecting the perimeter” with a combination of security technologies that includes localized clients embedded within devices or localized connections and centralized cloud-based intelligence which constantly scans to protect. The connected car will happen; we just have to care about the connected car and develop strategies so that we don’t get hacked.”
quote: Edward Snowden disclosures have heightened awareness of cybersecurity all over the world
Frymier: “Regardless of what you might think of Edward Snowden, there is no denying that his disclosures have heightened awareness of cybersecurity all over the world. Before that, many enterprises were running unencrypted data on their internal networks, which they believed were secure. Now they are beginning to use encryption internally as well, so I expect 2014 to be the year of encryption. The increased use of encryption will both enable and encourage more companies to use infrastructure-as-a-service (IaaS) cloud solutions, where previously they might have had concerns about the safety of their data in the cloud.”
Labels:
cybersecurity,
encryption,
iaas,
increase,
internal network,
snowden,
unencrypted
Location:
Salaspils, Salaspils pilsēta, Latvia
quote: broad implementation of large, long-term storage systems for unstructured data in 2014
Rector: “In 2013, there was a huge shift among enterprises in evaluating emerging methods to store, manage, analyze and access data – looking beyond traditional storage methods and storage vendors and starting to evaluate the benefit of open source options, RESTful/Object Storage and open long-term storage data formats such as LTFS; 2013 was also the "coming out party" for what the industry is labeling big data, which has become clear goes beyond just data analytics. Building on these trends in 2014, I anticipate broad implementation of large, long-term storage systems for unstructured data (Deep Storage, Object Storage, LTFS tape systems and Active Archives). Implementation of these systems will provide the enterprise with the foundational tools needed to store valuable data for longer periods and prepare for future big data/analytic initiatives to help make better predictive business decisions.”
quote: we will see the rise of a new Internet architecture [in 2014]
Source
Ward and Cooney: “We will see the rise of a new Internet architecture characterized by (a) ‘Fog Computing’ – the convergence of networking and compute at the edge of networks to create a more distributed intelligence that balances the need for centralized mega-scale data centers with more locally-useful computing and decision making capabilities, (b) a new type of networking architecture characterized by open APIs and by the embrace of developer communities who will create applications that optimize the integration of networks and management systems and business applications, (c) we will pass the peak of the SDN hype cycle and see real-world applications emerge in 2014, and (d) a new set of IT skillsets which accompany the convergence of computing, networking, storage and applications will emerge. IT professionals will need to become much more comfortable and familiar with IT domains beyond their silos; look to IT as new business enabler – not a burden if you build systems correctly thanks to the ability to deliver data upstream into business.”
Ward and Cooney: “We will see the rise of a new Internet architecture characterized by (a) ‘Fog Computing’ – the convergence of networking and compute at the edge of networks to create a more distributed intelligence that balances the need for centralized mega-scale data centers with more locally-useful computing and decision making capabilities, (b) a new type of networking architecture characterized by open APIs and by the embrace of developer communities who will create applications that optimize the integration of networks and management systems and business applications, (c) we will pass the peak of the SDN hype cycle and see real-world applications emerge in 2014, and (d) a new set of IT skillsets which accompany the convergence of computing, networking, storage and applications will emerge. IT professionals will need to become much more comfortable and familiar with IT domains beyond their silos; look to IT as new business enabler – not a burden if you build systems correctly thanks to the ability to deliver data upstream into business.”
Labels:
community,
enabler,
fog computing,
internet architecture,
it,
open,
sdn,
silos,
skillset
Location:
Salaspils, Salaspils pilsēta, Latvia
Subscribe to:
Posts (Atom)