--- /dev/null
+[[!meta title="af9015 bulk message failed: Switch USB Ports?"]]
+[[!meta date="2009-08-24 01:00"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Yesterday, I suddenly wasn’t able to use my AF9015-based DVB-T USB stick
+anymore. It only worked for about a minute, then the image froze and
+`dmesg` said the following:
+
+ [10633.166371] usb 1-3: USB disconnect, address 5
+ [10633.186247] af9015: bulk message failed:-22 (8/-339951920)
+ [10633.186254] af9013: I2C read failed reg:d417
+ [10633.186258] af9015: bulk message failed:-22 (8/-1067461161)
+ [10633.186265] af9013: I2C read failed reg:d417
+ [10633.186279] af9015: bulk message failed:-22 (9/-339951284)
+ [10633.186282] mt2060 I2C write failed
+
+I was not even able to `rmmod` the dvb-usb-af9015 kernel module, the
+rmmod process just hung itself up.
+
+After two hours of compiling a new driver version, experimenting with
+kernel parameters and getting a new [firmware
+version](http://www.otit.fi/~crope/v4l-dvb/af9015/af9015_firmware_cutter/firmware_files/4.95.0/),
+I finally booted into my rarely used Windows system. The first thing I
+saw was the “USB 2.0 device on USB 1.1 bus”, and as I also could not get
+a movie stream, I tried to plug the DVB-T stick into a different USB
+port. And voilà, suddenly everything worked again, as well on Windows as
+later on Linux. No more error messages in the kernel log, everything
+just fine as before.
+
+But I wonder if I have been using this USB port before — as far as I can
+remember, the DVB-T stick had always been on the other port… Yet another
+proof for me to never trust yourself :D
+
+**Update:** Unfortunately, this solution did not help for long. After
+about two weeks, the error is still there, and also occurs more
+frequently, near to once an evening (until I reboot). So maybe, it
+hasn’t even to do with switching USB ports, and my hardware is beginning
+to fail… :(
+
+**Update:** Seems it is fixed now, using revision dca0fed8d68b (Sat Sep
+26 19:27:52 2009 +0200) of
+[anttip’s](http://linuxtv.org/hg/~anttip/af9015) repository. At least it
+worked for over 4 hours now :-)
+
+[[!tag AF9015 driver DVB-T USB Linux Ubuntu]]
--- /dev/null
+[[!meta title="Bluetooth tethering via PAN with Windows Mobile and Ubuntu"]]
+[[!meta date="2010-11-08"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+[[!img defaults size=x200]]
+
+I was on the train today, needed some of the [Boost][0] manuals, and had
+no internet connection. So I wanted to use my phone (an old HTC Prophet
+with Windows Mobile 6.1) as a network access point to browse over
+GPRS/EDGE. As I found out, it is fairly simple with [Blueman][1] and it
+even provides [NetworkManager][2] integration, so all
+NetworkManager-capable applications can be notified about the
+connectivity. Windows Mobile 6.1 allows tethering over a Bluetooth PAN
+([Personal Area Network][3]); but there is another method called DUN
+([Dial Up Networking][4]), which I will not describe here. So here is a
+step-by-step tutorial what I did for my PAN approach, with a few
+(german) screenshots, tested on Ubuntu 10.04 Lucid:
+
+* Since my laptop was running on Ubuntu Lucid, there was already a
+ recent Blueman version in the Ubuntu repos available. On older
+ systems, you may want to add the [Blueman Launchpad PPA][5].
+
+ $ sudo apt-add-repository ppa:blueman/ppa # only necessary on pre-lucid systems
+ $ sudo aptitude update
+ $ sudo aptitude install blueman
+
+* Note that this also removes possibly installed `gnome-bluetooth`
+ packages since Blueman is an adequate replacement for the GNOME
+ Bluetooth UI.
+
+* After the installation has finished, I had to enable the NMPANSupport
+ plugin for NetworkManager 0.8 by right-clicking on the Blueman icon in
+ the GNOME notification area and selecting “Plugins”. For older
+ NetworkManager versions, there is also a plugin for NetworkManager
+ 0.7, called NMIntegration.
+
+ <div class="gallery">
+ [[!img blueman-menu.png alt="Blueman context menu"]]
+ [[!img blueman-plugins.png alt="Blueman plugin page"]]
+ </div>
+
+* Then I activated tethering on my phone (“Programs” → “Internet
+ Sharing” on my Windows Mobile 6.1, but YMMV). Apparently this was
+ neccesary with my model, because without tethering enabled I could not
+ get a Bluetooth PAN connection in the next step.
+
+ <div class="gallery">
+ [[!img winmo-program-menu.png alt="Windows Mobile Program screen"]]
+ [[!img winmo-connection-sharing.png alt="Windows Mobile Internet
+ Sharing application"]]
+ </div >
+
+* I paired the phone and my laptop via Bluetooth, and created a PAN
+ (Personal Area Network) by connecting to the “Network Access Point”
+ service on the phone. In Blueman, all you have to do after pairing is
+ right-click on the device and select “Connect To: Network Access
+ Point”. This creates a new network device `bnep0` which is
+ automagically configured through NetworkManager (using [stateless
+ address autoconfiguration][6]).
+
+ <div class="gallery">
+ [[!img blueman-pan.png size=x100 alt="Blueman: Context menu for device
+ “Leia”, menu entry “Network Access Point” is selected"]]
+ </div>
+
+ (Yes, my phone is called [Leia][7]… I also have a yet another HTC
+ Prophet for testing purposes, which is called [Luke][8] :-))
+
+* However, in my setup, though I was able to ping certain IP adresses on
+ the internet, DNS lookups timed out for some reason. It got better
+ when I explicitly set an IP address for the Bluetooth PAN driver on my
+ phone, and did the tethering process all over again.
+
+ <div class="gallery">
+ [[!img winmo-connections.png alt="Windows Mobile System Settings
+ Screen, with item “Wi-Fi” selected"]]
+ [[!img winmo-networkdrivers.png alt="Windows Mobile Network Driver
+ settings screen, with menu item “Bluetooth PAN Driver” selected"]]
+ [[!img winmo-bluetooth-ip.png alt="Windows Mobile Bluetooth PAN Driver
+ settings screen"]]
+ </div>
+
+* And off I went with mobile internet access. Woo-hoo! \o/
+
+I may also add that the NetUsage plugin in Blueman is very reasonable to
+use ;-) After activated, the network usage can be viewed by
+right-clicking on the Blueman icon and selecting “Network Usage”.
+
+[0]: http://www.boost.org/
+[1]: https://launchpad.net/blueman
+[2]: http://projects.gnome.org/NetworkManager/
+[3]: http://en.wikipedia.org/wiki/Personal_area_network
+[4]: http://en.wikipedia.org/wiki/Bluetooth_profile#Dial-up_Networking_Profile_.28DUN.29
+[5]: https://launchpad.net/~blueman/+archive/ppa
+[6]: http://en.wikipedia.org/wiki/Link-local%20address
+[7]: http://en.wikipedia.org/wiki/Leia_Skywalker
+[8]: http://en.wikipedia.org/wiki/Luke_Skywalker
+
+[[!tag Blueman bluetooth mobile_internet_access network NetworkManager
+ Ubuntu Ubuntu_Lucid Windows_Mobile]]
--- /dev/null
+[[!meta title="Change partition type without reformatting"]]
+[[!meta date="2011-01-02 17:29"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Note to myself: it is possible to change the partition type of a already
+formatted (and used) partition. For example, if you have already
+formatted the partition with NTFS, but accidentally had created it with
+partition type `0x83` (Linux), so Windows can’t read it, since it expects
+`0x07` (HPFS/NTFS). On Linux, you can use sfdisk for that purpose:
+
+ # Be root
+ # dd if=/dev/sdb of=sdb-bootsector count=1 # backup boot sector
+ # sfdisk -d /dev/sdb | sed -e 's/Id=83/Id=07/' > /tmp/sdb.txt
+ # sfdisk /dev/sdb < /tmp/sdb.txt
+
+(fill in the right values for your case)
+
+Of course, good old fdisk works also, use the `t` command.
+
+[(Source)](http://serverfault.com/questions/46758/can-you-change-the-partition-type-on-a-linux-server-without-starting-up-fdisk/46840#46840)
+
+[[!tag hacking howto useless_bits_of_information fdisk Linux
+ Master_Boot_Record NTFS partition partition_table partition_type
+ sfdisk]]
--- /dev/null
+[[!meta title="Dell Latitude E5500 and Magic SysRq"]]
+[[!meta date="2009-11-17"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+At work I’m using the Dell Latidude E5500 notebook, running on Debian
+testing. Today, I had some issues with Xorg which could not detect my
+keyboard and mouse, so I tried to do the Magic SysRq tricks (you can
+read about it at [Wikipedia][1]). Unfortunatley, to press `SysRq` (on
+`F10`), I had to use the `Fn` key, so if I pressed e. g.
+`Alt`+`Fn`+`SysRq`+`U`, the `U` was detected as `keypad 4` because of
+the `Fn` key. Luckily, it works as intended if you release the `Fn` key
+after having pressed `Fn`+`SysRq`, so to remount all mounted filesystems
+in read-only mode, you would actually hold `Alt`, hold `Fn`, hold
+`SysRq`, release `Fn`, press `U`.
+
+Never thought notebook keyboards were so smart :-)
+
+[1]: (http://en.wikipedia.org/wiki/Magic_SysRq)
+
+[[!tag Debian howto Debian_testing Dell_Latitude Linux SysRq]]
--- /dev/null
+[[!meta title="Flattr me!"]]
+[[!meta date="2010-11-14 15:55"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+[[!flattrthing id=83947]]
+
+Some of you may have noticed that I have put Flattr buttons below my
+posts. For those who are not familiar with Flattr: It is a micropayment
+service which allows you to show your appreciation for (free) content by
+making small donations to its author. Basically it works by spending
+only a fixed (but adjustable) amount of money per month, so you don’t
+have to worry about how much you can afford; and clicking Flattr buttons
+for things you like. At the end of the month, the money you decided to
+spend is divided by the number of Flattr buttons you clicked in this
+month, and the fractions are given to the respective authors. So if
+you decide to spend $4 each month, and you click four Flattr buttons
+of four different authors, every author gets $1. There is also a
+nice video on the [Flattr homepage][0] that explains the idea:
+
+[0]: http://flattr.com/
+
+<iframe class="youtube" width="560" height="315"
+src="https://www.youtube.com/embed/9zrMlEEWBgY" frameborder="0"
+allowfullscreen></iframe>
+
+So, from now on, if you like, you can flattr me. Just sign up on
+flattr.com and click the buttons below each post :-)
+
+<commercial />
+
+**Note:** The standard
+click-on-the-fly-and-show-how-many-users-have-flattrd-this button needs
+JavaScript, which is bad and takes away your privacy, so I only use the
+static button which directs you to the thing on the flattr site.
+
+[[!tag Flattr micropayment video]]
--- /dev/null
+[[!meta title="GNU screen: start with multiple windows and commands"]]
+[[!meta date="2010-07-30"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+I wanted to restart my IRC bot on reboot, but I also wanted to have
+control over it and see its log output, so I wanted to start it inside a
+`screen` session. This is nearly trivial, `screen -dmS yoursessionname
+yourcommand` is your friend, and you can later on reattach using `screen
+-r yoursessionname`.
+
+But what if I want to start multiple commands, each in its own screen
+window? My first solution used `screen -dmS` followed by something like
+`screen -r sessionname -X screen; screen -r sessionname -X next; screen
+-r sessionname -X title "my window title"; screen -r sessionname -X exec
+"my command line"`, but it seems that the `next` command fails in this
+context, and I ended up with all the mess in one single window.
+
+My next approach (okay, it took me half an hour of reading [the
+manual][0] until here ;-)) was more succesful: I created a session
+command file which contained screen commands like this:
+
+ screen
+ select 1
+ title "my window title"
+ exec mycommand arguments ...
+
+[0]: http://www.gnu.org/software/screen/manual/
+
+And, voilà, I could paste the single command line into my crontab:
+`screen -dmS sessionname && screen -r sessionname -X source
+sessioncommandfile`
+
+[[!tag hacking howto cron GNU screen]]
--- /dev/null
+[[!meta title="Google Earth and IPv6 DNS lookups"]]
+[[!meta date="2011-01-22 13:02"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Apparently the combination of a WLAN router that blocks IPv6 DNS queries
+of type `AAAA` (in my case, it was a Siemens S1621-Z220-A sold as [Alice
+Modem 1121 WLAN][0]) and the current version of Google Earth for Linux (I am
+using 5.1.3533.1731 from [Medibuntu][1]) do not work well together. The
+problem is that the router simply throws away `AAAA` queries (or
+generally, any type it does not know), so the DNS query times out.
+However, Google Earth does not seem to fall back to IPv4 queries (type
+`A`) in this case, and shows a message about network connectivity errors.
+I don’t know if it’s Google Earth’s fault or if the underlying eglibc
+resolver of my Linux system does something wrong, anyhow there is a
+fairly well-commented [bug report][2] on Launchpad for Ubuntu Karmic and
+Lucid which explains the issue.
+
+Anyway, I got rid of the problem by manually configuring a nameserver on
+my local machine (for example the nameserver(s) of your internet
+provider, or the ones of OpenDNS), and not using the WLAN router as a
+resolver. NetworkManager allows you to do this by editing a connection
+and choosing “Automatic DHCP (Addresses only)” on the IPv4 register tab;
+or you can write the settings directly to your `/etc/resolv.conf` (here
+for the OpenDNS servers):
+
+ nameserver 208.67.222.222
+ nameserver 208.67.220.220
+
+[0]: http://www.alice-wiki.de/Alice_Modem_1121_WLAN
+[1]: http://www.medibuntu.org/
+[2]: https://bugs.launchpad.net/ubuntu/+source/eglibc/+bug/417757
+
+[[!tag Ubuntu workaround Alice_DSL DNS eglibc Google_Earth HanseNet IPv4
+ IPv6 Ubuntu_Karmic Ubuntu_Lucid]]
--- /dev/null
+[[!meta title="Greetings from GNOME"]]
+[[!meta date="2009-12-27"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Now this is really useful: there is a [GNOME panel applet][1] for
+writing blog posts (for the Debian/Ubuntu folks: it’s in the package
+`gnome-blog`). Just gotta try it out :-)
+
+**Edit:** Unfortunately, there is no possibility to assign tags or
+categories this way :-(
+
+[1]: http://www.gnome.org/~seth/gnome-blog/
+
+[[!tag Debian Ubuntu useless_bits_of_information applet blog desktop
+ GNOME Linux]]
--- /dev/null
+[[!meta title="Hello World!"]]
+[[!meta date="2009-08-24"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Yeah. So this is my new blog. I thought, it would be a nice place to
+collect various bits of information that I noticed during my work that
+could be useful to other people too (as well as me, having a head like a
+sieve). So enjoy it, if you like!
--- /dev/null
+[[!meta title="Kaffeine 0.8.7 in Ubuntu Karmic"]]
+[[!meta date="2010-03-07"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+As I [[recently mentioned|ubuntu-karmic-upgrade-issues]], Kaffeine in
+Ubuntu Karmic is the (nearly) unusable version 1.0~pre2. I have
+backported the Jaunty version 0.8.7 to Karmic and put it into my
+[Launchpad PPA][1].
+
+[1]: https://launchpad.net/~rohieb/+archive/kaffeine-0.8
+
+[[!tag backport Kaffeine Ubuntu Ubuntu_Karmic]]
--- /dev/null
+[[!meta title="libdvdread and ISO 9660 file systems"]]
+[[!meta date="2011-03-03 22:33"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Apparently libdvdread only works with UDF file systems. I tried to point
+VLC to an ISO 9660 image file, but libdvdread only complained:
+
+ $ vlc dvd://foo.iso
+ libdvdread: Using libdvdcss version 1.2.10 for DVD access
+ libdvdnav:DVDOpenFileUDF:UDFFindFile /VIDEO_TS/VIDEO_TS.IFO failed
+ libdvdnav:DVDOpenFileUDF:UDFFindFile /VIDEO_TS/VIDEO_TS.BUP failed
+ libdvdread: Can't open file VIDEO_TS.IFO.
+ $ file foo.iso
+ foo.iso: # ISO 9660 CD-ROM filesystem data 'CDROM
+
+However, after I extracted the image file to a folder, everything went
+as expected. (Of course also with an image file containing an UDF file
+system :-))
+
+[[!tag useless_bits_of_information debugging DVD ISO9660 libdvdread UDF
+ vlc]]
--- /dev/null
+[[!meta date=2009-09-24]]
+[[!meta title="netcat, your friendly network sniffer"]]
+[[!meta license="CC-BY-SA 3.0"]]
+[[!meta author=rohieb]]
+
+I was bored today and started to play around with this Windows Mobile
+Messaging application (I think it’s called Outlook Mobile or sort of
+thing), and I found out that I was not able to connect to my IMAP
+mailbox on my root server, though it worked with my Freemail account. So
+I wanted to see what makes Outlook Mobile bother about my IMAP server.
+
+If you are familiar with Linux (which I suppose you are ;-)), you
+certainly know `netcat`. With this little tool, you can talk directly to
+servers on a byte-oriented basis, and this can be very useful if you
+have to debug programs which use character-oriented protocols like IMAP,
+SMTP, IRC and so on.
+
+But I’ve realised that I can not only use `netcat` to talk to a server
+myself, but even to build a transparent proxy server that displays all
+the data that comes over it. After a while — okay, it was about 2 hours
+— I got the following nice command:
+
+ $ mkfifo pipe
+ $ tty=`tty`; netcat -l 1234 < pipe | tee $tty |
+ netcat myserver.com 143 | tee pipe
+
+I could now set up Outlook Mobile to talk to port 1234 on my home
+computer and the bytes went straight to my console and also to port 143
+(the IMAP port) on my server.
+
+The first direction was straightforward: the first `netcat` process
+listens to the local port and pipes its output first to the console
+(using `tee`) and then to a second `netcat` instance that does the
+communication with the remote server. Now, the commands from the server
+have to get back to the client, so I created a named pipe using `mkfifo`
+(of course, your filesystem has to support it, so you better not do this
+on FAT) and used this as the input to the first `netcat` process that
+sends it back to the original client.
+
+Of course, I could have used Wireshark, but I hate that it does not
+allow to copy&paste the contents of a packet so I have only the bytes of
+the protocol that I need — which can be quite useful if you want to
+reuse parts of the content, especially in character-oriented protocols.
+Also, the filter settings in Wireshark can be annoying, there is no
+simple way to only have packets from one network connection (or I
+haven’t found it yet).
+
+So, finally I found out that is has something to do with the IMAP
+capabilities that Outlook Mobile bothers about. I suppose I will write
+something about it if I have traced the problem back.
+
+**Update:** Note: You can also rewrite the server and/or client messages
+using `sed`, but be sure to use unbuffered output with `-u` like that:
+
+ $ tty=`tty`; netcat -l 143 < pipe | tee $tty | netcat myserver.com 143 |
+ sed -u 's/^\* CAPABILITY.*/* CAPABILITY IMAP4 STARTTLS/' | tee pipe
+
+[[!tag hacking howto debugging IMAP Linux netcat network_protocols
+ proxy_server shell]]
--- /dev/null
+[[!meta title="Qt: (Nearly) synchronous QNetworkAccessManager calls"]]
+[[!meta date="2010-07-08"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+The [QNetworkAccessManager][0] class is very user-friendly, but it makes
+asynchronous calls. I was in the need for synchronous calls to handle my
+HTTP communication, but I did not want the overhead of another thread,
+so I googled a bit and finally came up with a short call to an event
+loop that processed the request. Like this:
+
+[[!format c """
+QNetworkAccessManager * pnam = new QNetworkAccessManager(this);
+// the slot was declared at another place
+connect(pnam, SIGNAL(finished(QNetworkReply *)), this,
+ SLOT(loginFinished(QNetworkReply*)));
+QNetworkRequest req(QUrl("http://foo.bar"));
+pnam->post(req, postData);
+
+// execute an event loop to process the request (nearly-synchronous)
+QEventLoop eventLoop;
+// also dispose the event loop after the reply has arrived
+connect(pnam, SIGNAL(finished(QNetworkReply *)), &eventLoop, SLOT(quit()));
+eventLoop.exec();
+"""]]
+
+This way my user-defined slot for the `pnam->finished()` signal was
+called immediately, and I could be sure to have the HTTP reply at the
+end of this code snippet.
+
+Found here: [Qt-Interest Mailing List: QNetworkAccessManager and
+QNetworkReply, synchronous][1]
+
+[0]: http://doc.qt.nokia.com/4.6/qnetworkaccessmanager.html
+[1]: http://lists.qt.nokia.com/public/qt-interest/2010-April/022031.html
+
+[[!tag howto programming multithreading network Qt signals slots]]
--- /dev/null
+[[!meta title="Qt: Throw exceptions from signals and slots"]]
+[[!meta date="2010-07-08"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+By default, you can not throw exceptions from signals and slots, and if
+you try it, you get the following message:
+
+> Qt has caught an exception thrown from an event handler. Throwing
+> exceptions from an event handler is not supported in Qt. You must
+> reimplement QApplication::notify() and catch all exceptions there.
+
+So, what to do? The answer is simple: Overwrite the function `bool
+QApplication::notify(QObject * receiver, QEvent * event)` so that it
+catches all thrown exceptions. Here is some sample code:
+
+[[!format c """
+#include <QtGui>
+#include <QApplication>
+class MyApplication : public QApplication {
+public:
+ MyApplication(int& argc, char ** argv) :
+ QApplication(argc, argv) { }
+ virtual ~MyApplication() { }
+
+ // reimplemented from QApplication so we can throw exceptions in slots
+ virtual bool notify(QObject * receiver, QEvent * event) {
+ try {
+ return QApplication::notify(receiver, event);
+ } catch(std::exception& e) {
+ qCritical() << "Exception thrown:" << e.what();
+ }
+ return false;
+ }
+};
+int main(int argc, char* argv[]) {
+ MyApplication app(argc, argv);
+ // ...
+}
+"""]]
+
+Of course, you can also inherit from `QCoreApplication` to get rid of
+the `QtGui` dependency, or display a nice dialog box instead of printing
+the messages to the console, or…
+
+Found at: [Stack Overflow: Qt and error handling strategy][1]
+
+[1]: http://stackoverflow.com/questions/1578331/qt-and-error-handling-strategy
+
+[[!tag howto programming exceptions Qt signals slots]]
--- /dev/null
+[[!meta title="SSH key authentication with encrypted home directories"]]
+[[!meta date="2010-10-09"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+[[!flattrthing id=80700]]
+
+Yesterday, I ran into an interesting problem: I tried to set up [SSH
+public key authentication][0] between two of my machines, `c3po` and
+`r2d2`, so I could log in from `rohieb@r2d2` to `rohieb@c3po` without a
+passphrase. However, everytime I tried to login to `c3po`, I was
+prompted to enter the passwort for `rohieb@c3po`, and the debug output
+mentioned something that the key could not be verified. More
+astonishing, when I established a second SSH connection while the first
+was still running, I was *not* prompted for a password, and debug output
+said that key authentication had been sucessful. I googled a bit, and
+after a while got to [this comment][1] on Launchpad, mentioning problems
+when the user on the remote machine had its home directory encrypted
+through ecryptfs – which was the case for me. Of course, since ecryptfs
+only encrypts the user’s home *after* he has been authenticated, the SSH
+daemon cannot read his `~/.ssh/authorized_keys` at the first time, and
+falls back to password authentication.
+
+[0]: http://www.openbsd.org/cgi-bin/man.cgi?query=ssh&sektion=1#AUTHENTICATION
+[1]: https://bugs.launchpad.net/ubuntu/+source/openssh/+bug/362427/comments/12
+
+The Launchpad comment proposes to first unmount the ecryptfs filesystem,
+then store `~/.ssh/authorized_keys` unencrypted, and then mount the
+encrypted home again (**note** that no program should be running that
+could try to access your home directory):
+
+ $ ecryptfs-umount-private
+ $ cd $HOME
+ $ chmod 700 .
+ $ mkdir -m 700 .ssh
+ $ chmod 500 .
+ $ echo $YOUR_REAL_PUBLIC_KEY > .ssh/authorized_keys
+ $ ecryptfs-mount-private
+
+This works indeed, but has the drawback that key authentication only
+works for the *first* login, because ecryptfs hides the unencrypted
+files when it mounts the encrypted directory on login; and you had to
+synchronize the encrypted and the unencrypted version of
+`authorized_keys` everytime you add a new key. To circumvent that, I
+simply moved the file to `/etc/ssh/authorized_keys/rohieb` (with the
+file only readable and writable by me, and `/etc/ssh/authorized_keys`
+writeable for all users) and adjusting `/etc/ssh/sshd_config`
+appropriately:
+
+ $ sudo vi /etc/ssh/sshd_config # or use your favorite editor instead of vi
+ [... some lines ...]
+ AuthorizedKeysFile /etc/ssh/authorized_keys/%u
+ [... some more lines ...]
+ $ sudo /etc/init.d/ssh restart
+
+## Update
+
+There is yet a better approach instead, which doesn’t need the SSHd
+config to be edited at all:
+
+1. login to the user on the remote machine
+2. create `/home/.ecryptfs/$USER/.ssh` and put your `authorized_hosts` there
+3. symlink your encrypted version there:
+
+ $ ln -s /home/.ecryptfs/$USER/.ssh/authorized_hosts ~/.ssh/authorized_hosts
+
+4. symlink your unencrypted version there (as above, **make sure** no
+ process wants to write to your home directory in the meantime):
+
+ $ ecryptf-umount-private
+ $ mkdir ~/.ssh
+ $ ln -s /home/.ecryptfs/$USER/.ssh/authorized_hosts ~/.ssh/authorized_hosts
+ $ ecryptfs-mount-private
+
+The paths are for Ubuntu 9.10 (Karmic Koala) and later. On other
+systems, you might want to replace `/home/.ecryptfs` with
+`/var/lib/ecryptfs`.
+
+[[!tag Debian howto Ubuntu workaround eCryptfs encrypted_home Linux
+ OpenSSH]]
--- /dev/null
+[[!meta title="Standby with Lenovo Thinkpad SL510 on Ubuntu Lucid"]]
+[[!meta date="2010-07-30"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+…note to myself: Remove a potentially mounted SD card before suspending
+your SL510, otherwise the kernel gets stuck…
+
+**Update:** Turned out the SD card was a bit buggy, the driver mostly
+got a timeout when trying to speak with it.
+
+[[!tag Ubuntu broken_hardware driver failure laptop Linux standby
+ Thinkpad_SL510 Ubuntu_Lucid]]
--- /dev/null
+[[!meta title="Sync your Windows Mobile 5/6 PDA with your Linux PC"]]
+[[!meta date="2009-10-30"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+I recently bought a new (ok, a rather old ;-)) PDA, and I got used to
+the calendar and task features in Windows Mobile. Of course, I also
+wanted to synchronise all the tasks, appointments, contacts and files
+with my PC, but whatever I tried, it didn’t work somehow on my Ubuntu
+9.04 machine… So, I’ve got it now: I had do blacklist the `ipaq` kernel
+module which wrongly handled the PDA when I plugged it in (i. e.
+edit `/etc/modprobe.d/blacklist.conf` and insert the line `blacklist
+ipaq`), and then the [manual from ubuntuusers.de][1] suddenly worked
+like a charm :-)
+
+[1]: http://wiki.ubuntuusers.de/Archiv/Synchronisation_mit_Windows_Mobile
+
+[[!tag howto Ubuntu blacklist Evolution iPAQ kernel_module Linux PDA
+ synchronisation Ubuntu Windows_Mobile]]
--- /dev/null
+[[!meta title="Tell Xorg to re-grab the keyboard"]]
+[[!meta date="2011-12-29 23:45"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+OK, I was doing some debugging with Xorg, and thought I had to use the
+[Magic SysRq][0] key to kill it. But when I had pressed Alt-SysRq-R to
+give the keyboard control from Xorg back to the kernel, it turned out
+that I not longer needed to do another SysRq because my Xorg magically
+worked again… ;-) Unfortunately now, everytime I pressed Alt-F4 to close
+a window, I found myself on tty4… rather poor. So I needed some way to
+tell Xorg to grab the keyboard again, and [there it is][1]: Just open an
+xterm and execute
+
+ sudo kbd_mode -s
+
+[0]: http://en.wikipedia.org/wiki/Magic_SysRq_key
+[1]: https://learninginlinux.wordpress.com/2010/06/16/debugging-notes-to-self/
+
+[[!tag fix howto debugging Linux shell SysRq Xorg]]
--- /dev/null
+[[!meta title="Ubuntu Karmic upgrade issues"]]
+[[!meta date="2010-01-17"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Thinking of the phrase “never change a running system” I am usually not
+very keen on updating my Ubuntu system to a new distribution. However,
+not every new software version is available from the backports, so
+finally, I saw myself forced to upgrade to Karmic Koala.
+
+I instantly regretted it. Of course, I appreciate all the new software,
+but as in every upgrade I ever made (and I probably should know by now
+-.-), there are a few things that were annoying me or are still doing
+it:
+
+* kaffeine-1.0~pre is not even as useful as 0.8.7 from jaunty: While
+ staying in touch with the main KDE (4) line, the developers seemed to
+ completely remove the Playback -> Video menu, so it is not possible
+ anymore to toggle deinterlacing, or any other video filter, which is
+ very annoying for DVB-T. I filed a [bug report][1] on that.
+ Futhermore, my DVB-T channels were lost, so I had to rescan them.
+
+* The notifications displayed by notify-osd somehow have wandered from
+ the upper right edge towards the middle right. This seems to be fixed
+ now, as pointed out in the [bug report][2], but somehow this fix never
+ got onto my system, though the changelog of notify-osd says so…
+ Additionally, the notifications for received messages in Pidgin do not
+ hide anymore if I actually read the messages, they persist until their
+ default timeout has elapsed. And they even show up now when my IM
+ status is on “Do not Disturb” – this was not the case (just as I
+ expected it) in jaunty.
+
+* Icons in GTK menus are now hidden by default, which seemed very
+ unfamiliar to me, since I always used them as an orientation guide,
+ especcially in the Firefox search plugin menu. You can show them again
+ in the System -> Preferences -> Appearance applet on the Interface tab
+ by selecting “Show icons in menus”.
+
+* GDM is no longer custumizable through themes. It just doesn’t support
+ it, as it seems to be a [complete rewrite][3]. IMHO just another
+ Unmature Software Thing.
+
+ **Edit:** I just found out that it is also not able to start GDM in
+ Xnest, as I usually did for testing:
+
+ $ gdmflexiserver --xnest
+ ** (gdmflexiserver:5916): WARNING **: Not yet implemented
+
+* And finally, my customized GNOME window theme (based on Clearlooks)
+ was broken :-( I am very confident that the color of selected text was
+ not the same color in the title bar, but now both seem to be the same
+ color. This is really bad, as for the title bar, I used to have a
+ darker shade of orange than for selected text. OK, changing to another
+ theme may be simple, but until now, I haven’t found anything I like
+ best.
+
+[1]: https://bugs.launchpad.net/ubuntu/+source/kaffeine/+bug/499938
+[2]: https://bugs.launchpad.net/ubuntu/+source/notify-osd/+bug/419894
+[3]: http://ubuntuforums.org/showthread.php?p=7854130
+
+[[!tag Ubuntu fix rant Ubuntu Ubuntu_Jaunty Ubuntu_Karmic upgrade]]
--- /dev/null
+[[!meta title="Use Ghostscript to convert PDF files"]]
+[[!meta date="2012-06-09 19:06"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+If you have a PDF file and want it to be in a specific PDF version (for
+example, the print shop where you just ordered some adhesive labels
+wants the print master in PDF 1.3, but your Inkscape only exports PDF
+1.4), Ghostscript can help:
+
+ gs -sDEVICE=pdfwrite -dCompatibilityLevel=1.5 -dNOPAUSE -dQUIET \
+ -dBATCH -sOutputFile=new-pdf1.5.pdf original.pdf
+
+(this converts the file `original.pdf` to PDF 1.5 and writes it to
+`new-pdf1.5.pdf`)
+
+Also, if you have a huge PDF of several megabyte because there are many
+high-resolution pictures in it, Ghostscript can minify it (and shrink
+the pictures to 96 dpi) if you use the parameter
+`-dPDFSETTINGS=/screen`.
+
+[0]: http://www.ghostscript.com/
+
+[[!tag howto desktop_publishing Ghostscript PDF]]
--- /dev/null
+[[!meta title="Windows Device Manager: Code 39 with CDROM drive"]]
+[[!meta date="2010-11-14 14:29"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+My sister asked me to have a look at her notebook (a Medion Akoya P6612
+with Windows Vista) because the CDROM drive wouldn’t work, and it was
+not even displayed in the Windows Explorer. I looked into the Device
+Manager and noticed that the CDROM device (TSSTCorp SN-S083a) was
+displayed with a small yellow exclamation mark besides its icon, and it
+said on the Properties page that the device could not be started and
+referred to Code 39. Reinstalling the drivers had no effect, but after I
+had a little chat with [Big Blue G][0], I found a [howto entry][1] which
+suggested the following:
+
+1. Be logged in with an administrator account
+2. Open the Registry Editor (choose it from the Start Menu or press
+ Win+R and type `regedit`)
+3. Navigate to
+ `HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Class\{4D36E965-E325-11CE-BFC1-08002BE10318}`
+4. Then, in the right pane, delete all of the following keys:
+ * UpperFilters
+ * LowerFilters
+ * UpperFilters.bak
+ * LowerFilters.bak
+5. Restart your computer
+
+This worked fine for us.
+
+[0]: http://google.com/
+[1]: http://www.pchell.com/hardware/cd_drive_error_code_39.shtml
+
+[[!tag howto useless_bits_of_information Windows CDROM Device_Manager
+ driver Medion_Akoya_6612 registry TSSTCorp_SN-S083a Windows Vista]]
--- /dev/null
+[[!meta title="Wireless USB keyboards and delayed keystrokes"]]
+[[!meta date="2012-04-29 01:09"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+Everytime I was using my wireless USB keyboard with my laptop while the
+power cable was not connected, the keyboard behaviour was lousy, and
+keystrokes tend to be delayed by several seconds. The kernel logs said
+something like
+
+ [23302.802096] uhci_hcd 0000:00:1a.0: setting latency timer to 64
+ [23302.842177] uhci_hcd 0000:00:1a.1: PCI INT B -> GSI 21 (level, low) -> IRQ 21
+ [23302.842190] uhci_hcd 0000:00:1a.1: setting latency timer to 64
+ [23302.882145] uhci_hcd 0000:00:1a.2: PCI INT C -> GSI 19 (level, low) -> IRQ 19
+ [23302.882158] uhci_hcd 0000:00:1a.2: setting latency timer to 64
+ [23302.929065] uhci_hcd 0000:00:1d.1: PCI INT B -> GSI 19 (level, low) -> IRQ 19
+ [23302.929079] uhci_hcd 0000:00:1d.1: setting latency timer to 64
+
+Searching on Google, I found [a forum post][0] which suggested to
+disable USB auto-suspend in the [laptop-mode][1] configuration files.
+The relevant file on my Ubuntu 12.04 is
+`/etc/laptop-mode/conf.d/usb-autosuspend.conf`, which is fairly well
+documented, and has an option `AUTOSUSPEND_USBID_BLACKLIST` which
+allowed me to blacklist my USB keyboard, so the device was no longer put
+in auto-suspend mode. (The USB ID needed for
+`AUTOSUSPEND_USBID_BLACKLIST` can be found in the output of [lsusb][2])
+
+After editing that file, I had to restart the laptop-mode daemon (`sudo
+/etc/init.d/laptop-mode restart`), and keystrokes from my wireless
+keyboard arrived again without any delay.
+
+[0]: https://bbs.archlinux.org/viewtopic.php?pid=898114#p898114
+[1]: http://packages.ubuntu.com/laptop-mode
+[2]: http://packages.ubuntu.com/usbutils
+
+[[!tag fix Ubuntu auto-suspend battery_power laptop-mode Linux Linux_3.2
+ powersave timing Ubuntu_Precise USB wireless_keyboard]]
--- /dev/null
+[[!meta title="X screen shots from the console"]]
+[[!meta date="2011-03-24 19:10"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+I had to debug a machine that was behind a DSL-2000 connection, and had
+about 20 KB/s of upstream. For that, I needed to see what was going on
+on the X screen, but due to the low bandwidth (and a screen resolution
+of 1920×1080), VNC was about as fast as 2 frames per minute.
+
+But I found a [comparable replacement][0]: The [ImageMagick suite][1]
+has a program called `import` that allows you to dump the contents of
+the X screen to a image file. So I took a few screen shots from the
+console via `DISPLAY=:0 import -window root foo.png` and then copied the
+files to my machine.
+
+[0]: http://www.mysql-apache-php.com/website_screenshot.htm
+[1]: http://www.imagemagick.org/
+
+[[!tag howto ImageMagick Linux VNC X X11 Xorg]]
--- /dev/null
+[[!meta title="XULRunner rocks!"]]
+[[!meta date="2011-03-15 22:16"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+
+For [[one of my projects|projects/infopoint-html]], I needed an
+application to display a web page in full screen mode. At first, I used
+Firefox with the [AutoHide extension][0], but this solution was more of
+a hack and not easy to deploy to multiple machines — I worked with a
+pre-configured user profile that was copied every time the application
+started. Furthermore, after each update, Firefox would check for
+compatibility of installed plugins and displayed a nasty dialog in the
+meantime.
+
+So I tried to move away from Firefox and do something on my own,
+something slim which did just what I wanted, nothing more, and do it
+good — according to the [UNIX philosophy][1]. But writing another
+C/C++/Python/whatever application from scratch was not an option
+(implementing an HTML renderer would be a pain, and I didn’t fancy
+reading extensive manuals about WebKit, Gecko or any other rendering
+engine).
+
+After a while of thinking, which included thought fragments of
+[Songbird][2] and [Conkeror][3], I decided to give [XULrunner][4] a shot
+(for those who do not know, XUL is the XML-based user interface language
+used by the Mozilla applications and the Firefox and Thunderbird
+extensions, and XULRunner is an interpreter and run-time environment for
+XUL documents).
+
+So after a while of hacking (there is a good [tutorial on the Mozilla
+Developer Network][5]), I ended up with a few lines of code:
+
+[[!format xml """
+<?xml version="1.0"?>
+<?xml-stylesheet href="chrome://global/skin/" type="text/css"?>
+<?xml-stylesheet href="main.css" type="text/css"?>
+<window
+ xmlns="http://www.mozilla.org/keymaster/gatekeeper/there.is.only.xul"
+ id="viewer" windowtype="viewer" title="Infopoint HTML View"
+ hidechrome="true">
+
+ <hbox flex="1">
+ <iframe id="contentview" flex="1"
+ src="chrome://infopointhtmlviewer/content/default.html" />
+ </hbox>
+ <script>
+ // load URI given on command line
+ var content = document.getElementById("contentview");
+ var cmdLine = window.arguments[0].QueryInterface(
+ Components.interfaces.nsICommandLine);
+ var uri = content.getAttribute("src");
+ alert("Default URL: " + uri);
+ if(cmdLine.length > 0) {
+ uri = cmdLine.getArgument(0);
+ }
+
+ if(content != null) {
+ content.setAttribute("src", uri);
+ }
+
+ // resize to full screen
+ window.resizeTo(screen.width, screen.height);
+ </script>
+</window>
+"""]]
+
+The above code is in the public domain.
+
+And that was basically everything. I was surprised that there was
+nothing more to it.
+
+You can get the source code of the full application [on GitHub][6].
+
+[0]: http://web.archive.org/web/20120415204503/http://www.krickelkrackel.de/autohide/autohide.htm
+[1]: http://en.wikipedia.org/wiki/Unix_philosophy
+[2]: http://getsongbird.com/
+[3]: http://conkeror.org/
+[4]: https://developer.mozilla.org/en/XULRunner
+[5]: https://developer.mozilla.org/en/Getting_started_with_XULRunner
+[6]: https://github.com/rohieb/infopoint-html
+
+[[!tag programming GitHub JavaScript Mozilla Mozilla_Firefox XUL
+ XULRunner]]
--- /dev/null
+[[!meta title="ZSNES on AMD64 Ubuntu"]]
+[[!meta date="2010-10-06"]]
+[[!meta author="rohieb"]]
+[[!meta license="CC-BY-SA 3.0"]]
+[[!flattrthing id=80701]]
+
+**[ Update, 2013-10:** This post post is not up to date anymore. On newer
+Debians (since 7.0/wheezy) and Ubuntus (at least since 12.04, Precise Pangolin),
+you should be able to install zsnes out of the box: `sudo apt-get install
+zsnes:i386`. For details see the MultiArch documentation for
+[Debian][MultiArchDebian] and [Ubuntu][MultiArchUbuntu]. **]**
+
+[MultiArchDebian]: https://wiki.debian.org/Multiarch/
+[MultiArchUbuntu]: https://help.ubuntu.com/community/MultiArch
+
+Before I bought my current hardware, I was working on a 32-bit-based
+system, and I really appreciated ZSNES as an SNES emulator. But
+unfortunately, my new hardware was an AMD64 system, and there is
+currently no ZSNES package for 64-bit Ubuntu or Debian :( So I decided
+to google a bit about the issue, but it took me until now (a year later)
+to get ZSNES finally working on my machine. The problem is, if you build
+ZSNES on a 64-bit machine, all the application does is segfault at
+start, and if you [try to compile for 32-bit systems][0], you get errors
+about missing 32-bit libs (in particular, configure does not find a
+suitable `libsdl`). Instead, if you just take the binary which was
+compiled on a 32-bit system, and install the `ia32-libs` package,
+everything seems to work—at least I was able to play a few levels of
+Super Mario World succesfully :-)
+
+[0]: http://board.zsnes.com/phpBB3/viewtopic.php?p=118067&sid=dd9a2a54d9178eb5009c33586aea703c#p118067
+
+So here was my idea: take the 32-bit package from the Ubuntu repository,
+and just change the Architecture control field, and by this fool dpkg :P
+And as it turned out, this idea worked great. You can get the Debian
+package here if you want, it *should* work for Ubuntu Karmic and Lucid,
+as well as for Debian testing (**but** I only tested it on Lucid, so
+there is no warranty here—but I’m happy to hear if it works :-)):
+
+* [zsnes_1.510-2.2ubuntu3~ppa1_amd64.deb][2]
+* SHA1: `716bbd37267b477ef02961a7727212619309b83f`
+* MD5: `452ea5230ad17df1dee649ab4cc6c8c0`
+
+[2]: http://rohieb.name/stuff/zsnes_1.510-2.2ubuntu3~ppa1_amd64.deb
+
+## How to Reproduce It
+For the curious people reading here, here is what I actually did:
+
+1. `wget http://archive.ubuntu.com/ubuntu/pool/universe/z/zsnes/zsnes_1.510-2.2ubuntu3_i386.deb`
+2. `ar x zsnes_1.510-2.2ubuntu3_i386.deb`
+3. `tar xzf data.tar.gz`
+4. Edit `usr/share/applications/zsnes.desktop` and added `-ad sdl` to the
+ `Exec:` field, otherwise it would just segfault on the first run:
+
+ Exec=zsnes -ad sdl
+
+5. Edit `usr/share/doc/zsnes/changelog.Debian.gz` and added a new
+ changelog entry for the version (just copy one of the previous
+ entries and adapt it)
+6. `tar xzf control.tar.gz`
+7. Edit the `control` file, changed the `Version:` and `Architecture:`
+ field to `amd64`, added the `ia32-libs` dependency, and set myself as
+ maintainer:
+
+ Package: zsnes
+ Version: 1.510-2.2ubuntu3~ppa1
+ Architecture: amd64
+ Maintainer: Roland Hieber <foobar@example.org>
+ Installed-Size: 4160
+ Depends: ia32-libs, libao2 (>= 0.8.8), libc6 (>= 2.4), libgcc1 (>= 1:4.1.1),
+ libgl1-mesa-glx | libgl1, libpng12-0 (>= 1.2.13-4),
+ libsdl1.2debian (>= 1.2.10-1), libstdc++6 (>= 4.1.1), zlib1g (>= 1:1.2.2.3)
+ [...]
+
+8. Change the `md5sums` file for the right values for
+ `usr/share/applications/zsnes.desktop` and
+ `usr/share/doc/zsnes/changelog.Debian.gz` (I used the `md5sum`
+ command and copy-pasted it)
+9. `tar czf control.tar.gz control md5sums postrm postinst`
+0. `tar czf data.tar.gz usr/`
+1. `ar r zsnes_1.510-2.2ubuntu3~ppa1_amd64.deb debian-binary
+ control.tar.gz data.tar.gz`
+
+I’m afraid that I can’t put the package into [my PPA][3], Launchpad only
+accepts source packages for uploads, and builds the binary packages
+itself, both for i386 and AMD64. This approach can not be used here,
+since we needed the i386 binary for AMD64.
+
+[3]: https://launchpad.net/~rohieb/+archive/ppa
+
+[[!tag Debian hacking howto Ubuntu workaround Linux packaging
+ Ubuntu_Jaunty Ubuntu_Karmic ZSNES]]
projects / www-rohieb-name.git / commitdiff
