I will work on adapting this theme to better suit my needs, and especially allow easy reading of “long” articles. I hope my readers will enjoy it.

At first sight, everything was working well “out of the box”, but when I have checked the log files, I could see the following lines repeated very frequently:

Jun  1 08:22:08 athyr kernel: [ 1412.134045] usb 6-2: reset high speed USB device using ehci_hcd and address 3

I repeat everything seems to work OK, but I can’t stand this kind of messages, especially when it deals with storage (by experience, everything works well during the tests, you simply ignore error messages, and you loose your data the first time you really use your hardware!).

After some googling, I have found this issue is hardware related. A lot of people seem to get hit by this. Some have even reverted to USB 1 (UHCI)!

Further googling pointed out this article.

I have tried to disable USB auto-suspend, and changing the idle-delay time (default is 2 seconds). But it doesn’t help.

Just for reference, this can be done like this:
echo -1 > /sys/module/usbcore/parameters/autosuspend

Note the following part of the documentation:
The idle-delay values for already existing devices will not be affected

To make the change permanent, you need to append usbcore.autosuspend=-1 to the boot parameters (USB support is compiled into Debian kernels).
More information on: http://www.mjmwired.net/kernel/Documentation/usb/power-management.txt

I can also see the following message at boot time:

unable to enumerate USB device on port 4

According to this recent discussion, it seems it is harmless.

I have however noticed that unplugging the card reader and re-plugging it after the system has booted make these log entries disappear. Does someone have an idea?

Note that I use the latest Debian 2.6.25 kernel:

$ uname -a
Linux athyr 2.6.25-2-amd64 #1 SMP Tue May 27 12:45:24 UTC 2008 x86_64 GNU/Linux

I have made a typing error when incrementing a zone serial number this morning (I though we were in year 2009!). I realized this error after the transfer was made to my secondary server, and I couldn’t figure out how I could fix this: if the serial number on the slave server is lower than the serial number on the master, the slave server will attempt to update its copy of the zone.

I have read some websites explaining one should add 2147483647 (2^31-1) to the number, reload the zone, let the slave get updated and revert to the appropriate serial number.

Unfortunately, this solution hasn’t worked for me.

I have read in RFC 1912 (section 2.2 - SOA records), that the system was planned until year 4294. I have henced updated the number of my master server to year 4294123199 (ie. the last possible serial number), reloaded the zone. After the transfer was propagated to the slave, I could simply revert to a standard serial number.

The DNS transition is not completely over, but should be ok tomorrow.

It appears everything is working well.
Please report any issue you might notice.

The aim is to ease maintenance, reduce power usage, and avoid any risk of fire when not at home.

Most of the websites and the mail server were moved to the new virtual server at the beginning of the week.

I had to postpone my backup configuration to deal with the OpenSSL vulnerability.

Unfortunately, one of the filer of my new hosting company crashed on Thursday and all the data of my virtual host have been lost.

The server is still unreachable, I had to move back to my home server on Friday.

This lead to some outages on all my websites in the meantime. Sorry for any inconvenience.

I have also lost a few e-mail (which I have received on the new host, but not yet read, ie. not yet in my MUA cache). If you sent something I haven’t yet answered to, please send it again.

I hope I will be able to work on the virtual server again next week-end.

• ufraw was updated to support my Sony DSLR-A700 camera - packaged based on the latest release available in sid.
• hugin package in unstable is outdated - I have based my package on Debian PhotoTools member Cyril Brulebois’s work. It seems that some legal questions have been raised, preventing the package to be updated in the official archive.
• autopano-sift fixes FTBS and other bugs in Christian Marillat’s package and proves working well with hugin. I will e-mail him a patch with the amendments I have worked on.
• enblend needs to be built from VCS to work with current hugin snapshots. The package is fully based on the package available in sid.

Though I use these packages on a regular basis, they haven’t been thoroughly tested as official packages would be, and might still contain bugs or errors. Please report any problem you might have.

I will try and update these packages when time allows it. VCS snapshots might be unstable, or even non-usable. As usual, use these non-official packages on your own risk.

Also note that these packages are currently only available for amd64. I will eventually build i386 packages if requested.

As for music, it works fine with rhythmbox from experimental rebuilt against latest libgpod library.

But I can’t find any Linux software which would allow us to transfer pictures. I have tried gpixpod, but it seems the iPod is too new as it doesn’t even detect it. iTunes does not run with Wine.

I think I will try to set up iPod Linux some day…

I have also lost some time due to the fact I had forgotten to install the headers of the upgraded kernel, preventing me from building the module of my wireless card.

Again, many thanks to Micah Anderson, Christoph Haas and Loïc Minier for their advocacies and, in a more general way, for the help they bring on a day-to-day basis.