1. Add “deb http://us.archive.ubuntu.com/ubuntu/ lucid universe” to the end of “/etc/apt/sources.list” (or use the Software Sources applet)
2. sudo apt-get update
3. sudo apt-get install netatalk (This will install netatalk 2.0.5 from the Lucid universe repository)
4. Make sure that your “/etc/netatalk/afpd.conf” doesn’t override the default, so remove the last uncommented line if you followed this guide previously, or otherwise ensure that “uams_dhx2.so” is listed.
5. sudo /etc/init.d/netatalk restart
6. Finally, you need to remove any .AppleDB folders from your shares. These folders might be hidden in Nautilus, so you may need to do this from the CLI.

If your interested, I made a quick note of how it’s done, http://davidgoldberg.me/tutorial/using-wordpress-post-thumbnail-with-fancybox/.

I added:
‘add_image_size( ‘postliste-thumbnail’, 90, 90, true );’
to the functions.php

Then I added this in the loop:
‘php the_post_thumbnail(( ‘postliste-thumbnail’ )’…

meaning that now I have a thumb on the home.php in 200 x 140 px and a 90 x 90 px for every post where I put the code.

nice …

My problem is that I have defined a thumb-size in WP admin at 200 x 140 px. This works fine on my home.php.

I have used your suggestions so when using eg. 90 x 90 px in a loop, the images will display a proportionals size like 65 x 90 px and NOT 90 x 90 px.

Any suggestions, please?

Now I want to use the thumbs in a loop of a specific category on a page:

As you can see, the the_post_thumbnail triggers 90 x 90 px thumb, but on my page, it will display the proportions, so 90 x 90 will become something like 65 x 90 px. It will not display a square image!

Can anyone figure this out?

Any help would be great!

thx

Do you perhaps have a fix for -op-?

thanks – Val

src=http://www.gravatar.com/avatar/6e2c869ce80cd3fe413e001f1b26b1ec?
s=65&
d=http%3A%2F%2Flocalhost%2F1gs%2Fwp-content%2Fthemes%2F1green%2Fgr-pg%2Favat06.gif
%3Fs%3D65&r=G' class='avatar avatar-65 photo' height='65' width='65'

”

http://wpengineer.com/add-avatar-to-wordpress-default/#comment-3662

adds this to the first line:
if ( !function_exists(‘fb_addgravatar’) ) {
***
}

but either way I can’t get either php to work. It just shows up as a blank image in settings, and the path in browser “view source” is:

which doesn’t make sense. Why is it bringing gravatar into it? I have “Show Avatars” checked.

The path in php is:
get_bloginfo(‘template_directory’) . ‘/gr-pg/avat06.gif’;

which should read in browser as:
<img src="http://localhost/1gs/wp-content/themes/1green/gr-pg/avat06.gif"

I tried one solution on that other link, to add (although it shouldn’t be needed) but that didn’t work.

The other suggestion “In the wp-settings.php plugins.php needs to be required before functions.php because plugins.php contains the add filter function” linking to:
http://wordpress.org/support/topic/102858?replies=4

I don’t know how to implement because “function add_filter” is in \mytheme\wp-includes\plugin.php

but my 2.9.1 wp-settings.php looks like this:
require (ABSPATH . WPINC . ‘/plugin.php’);
require (ABSPATH . WPINC . ‘/default-filters.php’);
include_once(ABSPATH . WPINC . ‘/pomo/mo.php’);
require_once (ABSPATH . WPINC . ‘/l10n.php’);

there’s no:
require (ABSPATH . WPINC . ‘/functions.php’);

so I can’t reverse the order.

I also tried the suggestion above:
get_stylesheet_directory_uri()

but that also makes no difference.

Do you perhaps know why I get a blank image in Discussion Settings and that funny browser path of src=’http://www.gravatar.com/avatar/etc?

thanks, Val

great tut! I have followed it for 2 machines (+1 mac to view them from) and gotten mixed results.

machine 1 = Ubu 9.10, ext3 fs shared out + the resumable vnc thing, all working perfectly, except that spotlight can’t index it because it’s not hfs+. That’s a bad thing for my Work Projects folder…

machine 2 = Deb 5 Squeeze, with a dedicated hfs+ drive just for this (in the hopes of getting spotlight to work). can’t seem to write to it, only read, can’t figure out why and don’t know where to start looking.

But here’s the even bigger problem, assuming I can figure out machine 2’s permissions issue… if both machines are running netatalk, the mac confuses one for the other and shows the same shares for both machines (it shows machine 1’s shares), even though the 2 lin boxes have different shares entirely, different static IPs, different mac addresses, you name it…

so why would the Finder be showing the shares from machine1 when browsing machine2? Where do I even begin to look for clues?

1) using gui installer or yum install netatalk
2) configure the xml afp service for avahi as per this article instructions. can’t remember if avahi was already configured to start but you can do chkconfig avahi on
3) yum install nss-mdns.
4) edit /etc/nsswitch.conf
#hosts: db files nisplus nis dns
hosts: files mdns4_minimal [NOTFOUND=return] dns

To test, use terminal to ping each other. e.g. ping .local and from fedora ping .local

Using Fedora remote desktop (vino), Avahi, I was able to use Apple Remote Desktop screen sharing.

hope this helps someone with fedora 12.

Jag

1. Experimented with ddb, but had to change it back to the initial cdb in /etc/netatalk/AppleVolumes.default
cause it was giving:

“The operation could not be completed because the original item for ‘Time Machine’ could not be found”

2. It’s important you figure out en0’s mac address and Don’t look at the Console’s: disk image path, cause putting an image in that location doesn’t work. You really need the _.sparsebundle, and nothing else, or you’ll get an:

“Error 45″

Deleting all of the metadata files that OS X stores on the drive fixed the problem (basically delete everything in the root of the share that starts with a “.” – be careful)

Just posting in case others are having this issue.

http://verytogether.com/lifestyle/play/javascript-feelings-intimacy-with-strangers.html

I had to shift to another server, and moved all my storage over to another box. I used the same IP and other settings. Despite repeated netatalk setups, I kept getting “Connection Failed.” The folders would even show in a finder popup window.

Turns out that when I moved the files over, I brought over the .AppleDB directories with them. AFP doesn’t like this, per this post: http://www.gentoo-wiki.info/HOWTO_Share_Directories_via_AFP.

If you simply go in and delete those .AppleDB directories from each share, everything will be hunky dory again on your new server.

However, after reading the comments on Debian bug #191790 and a little investigation, I discovered that the stock version distributed in the repository is now good to go without the need to re-compile. In other words, step 1 could be much simpler:

sudo aptitude install netatalk

Evidently, the maintainers upstream have enabled DHX2 authentication support, but not activated it by default. Based on step 2 above, I edited /etc/netatalk/afpd.conf to use uams_dhx2.so instead of uams_dhx.so and now my 10.6 client connects just fine. (I haven’t tested with any other versions, but DHX2 support for encrypted passwords has been built-in since OS X 10.2.)

My /etc/netatalk/afpd.conf:

- -transall -uamlist uams_dhx2.so -nosavepassword -advertise_ssh

Perhaps in the future the package maintainers will activate DHX2 by default (rendering this one-time step unnecessary), but until then, this is a lot simpler than recompiling. Plus, this means you can now safely upgrade to new releases of netatalk without breaking your setup.

Thanks again for the excellent post.

Howerver, TimeMachine continuted to fail to create the first backup with the error 45.

Any ideas what I’m missing?? Thanks in advance. Below are the details.

ubuntu server 9.10
OSX 10.6.2

I used hdiutil to manually create a volume on my desktop. I then copied the com.apple.TimeMachine.MachineID xml file with my guiid inside into that new sparsebundle. The I copied that sparsebundle to the server.

In watching the mac’s syslog it looks like Time Machine is not using the sparsebundle I created.

For example.
1) When I create a sparsebundle named “hostname.sparsebundle” then TimeMachine fails trying to create “hostname 1.sparsebundle.”
2) When I create a sparsebundle named “hostname 1.sparsebundle” then TimeMachine fails trying to create “hostname.sparsebundle.”

Here is a snipped from my Mac’s sys log file.

23:11:37 new-host /Applications/System Preferences.app/Contents/MacOS/System Preferences[15755]: TMShowUnsupportedNetworkVolumes is set to TRUE which allows selection of unsupported Time Machine network destination volumes. Use of an unsupported network destination can result in silent data corruption and complete loss of backup data under certain circumstances!
23:11:43 new-host UnmountAssistant[15768]: Volume unmounted successfully
23:12:09 new-host com.apple.backupd[15584]: Starting standard backup
23:12:09 new-host com.apple.backupd[15584]: Network destination already mounted at: /Volumes/backup-2
23:12:09 new-host com.apple.backupd[15584]: Creating disk image /Volumes/backup-2/hostname 1.sparsebundle
23:12:10 new-host hdiejectd[15776]: running
23:12:20 new-host hdiejectd[15776]: quitCheck: calling exit(0)
23:15:57 new-host com.apple.backupd[15584]: Error 45 creating backup disk image
23:15:57 new-host com.apple.backupd[15584]: Failed to create disk image
23:16:02 new-host com.apple.backupd[15584]: Backup failed with error: 20
23:16:02 new-host com.apple.backupd[15584]: Ejected Time Machine network volume.

_rfb._tcp
5901

on the .service file, but I want to remove the second vnc service announce.

Thank you

Great tutorial, and I see it’s still growing! I have 9.10 and by default the VNC service is broadcasted. With that the afp service is broadcasted, but with an other name.

The afpd.service file you use is perfect because I can change the name and the icon, Perfect! But the vnc service is broadcasted as “%u’s remote desktop on %h”. How can I change that? Also, how can I see all the services that avahi is broadcasting, like firefly etc…
Thank you allot.

http://notepad.bobkmertz.com/2010/01/apple-time-machine-backups-to-ubuntu.html

thanks
sharath

Unfortunately someone forgot, that the Squeeze netatalk package needs a newer version of the package libgcrypt11 than Squeeze is offering at the moment.

So I compiled a new libgcrypt11 package (1.4.5-1):

aptitude build-dep libgcrypt11
cd /usr/local/src
wget http://ftp.de.debian.org/debian/pool/main/libg/libgcrypt11/libgcrypt11_1.4.5.orig.tar.gz
tar xvf libgcrypt11_1.4.5.orig.tar.gz
cd libgcrypt-1.4.5
wget http://ftp.de.debian.org/debian/pool/main/libg/libgcrypt11/libgcrypt11_1.4.5-1.debian.tar.gz
tar xvf libgcrypt11_1.4.5-1.debian.tar.gz
wget http://ftp.de.debian.org/debian/pool/main/libg/libgcrypt11/libgcrypt11_1.4.5-1.dsc
dpkg-buildpackage -b -uc
dpkg -i libgcrypt11_1.4.5-1_amd64.deb (system architecture may vary, insert yours instead of “amd64″)
The other packages build by dpkg-buildpackage I guess are not necessary, but I installed them all too:
dpkg -i libgcrypt11-dev_1.4.5-1_amd64.deb
dpkg -i libgcrypt11-dbg_1.4.5-1_amd64.deb
dpkg -i libgcrypt11-doc_1.4.5-1_all.deb

/etc/init.d/netatalk restart
And I was able to connect to the Server, after as usually deleting the .AppleDB directories in the root of all afp server volumes.

Hope this helps someone. Since the new libgcrypt11 package is already available in Sid (unstable), you could also install it from there if you want to mess around in your sources.list …

Thanks,

maze

Jack

/media/movies “movies” cnidscheme:cdb options:usedots, upriv

After you change the configuration file, restart the netatalk services: sudo /etc/init.d/netatalk restart

I had some trouble at the beginning too, and this worked for me. Good luck.

They get created on every root shared folder that I access from my Mac via AFP. While not visible to the Mac, they are visible to Windows and through Ubuntu when using the command line.
I’ve tried deleting them, but they keep reappearing. I do not want a “Network Trash.”
Searching online hasn’t been much help.
Anyone?

Anyone have any ideas???

I have followed this guide to enable a TM and a shared file server on ubuntu.

I have just encountered one problem :

When i lauch the external disk on my mac, I see my customized icons (for example on movie files I insert the poster as icon).
But when i see it through the file server, i can not see the customized icons but only the default one.

I used usedots option if it can help.

Interesting though that title tag is always correct. So if title tag is displayed then why not ALT tag is displayed without any additional code?

Is there anything else we need to change to make it automatically display ALT tag.

Tx

Thanks,
Axel

I have a question though. Once the backup is finished, I plan to move my iTunes library from the hard drive to the USB drive that used to be my TimeMachine backup and then I want to add this drive as well to the backup. My question is, will I need to create a second sparseimage file for the USB when added?

Thanks,
Leo

Just thought I’d add the some netstat info from the debian box:

tcp 0 0 0.0.0.0:548 0.0.0.0:* LISTEN 2387/afpd
tcp 0 0 127.0.0.1:4700 0.0.0.0:* LISTEN 2385/cnid_metad
udp 0 0 0.0.0.0:39358 0.0.0.0:* 2091/avahi-daemon:
udp 0 0 0.0.0.0:5353 0.0.0.0:* 2091/avahi-daemon:
udp6 0 0 :::41490 :::* 2091/avahi-daemon:
udp6 0 0 :::5353 :::* 2091/avahi-daemon:

Those are all the daemons related to this right? Anything off?

Thanks
/dan

There was an error connecting to the server. Check the server name or IP address and try again.

Any ideas? Thanks, Leo

I’m an idiot – thanks! Can’t believe I didn’t check the zip file first. *sigh*

/vjl/

TIA,

/vjl/

I only had to manually mount the share once to have it show under the list of drives under Time Machine. After that, it’s all automatic.

The only thing that does bother me, and this has been happening for a while now even before changing the location of my TM, is that Time Machine will automatically attempt to back-up every hour – even when I’m not on the same network where the TM drive is. The TM icon “spins” for a few seconds, and then it stops, with the message “Latest Backup: Delayed”.
It will do this for as long as the Mac is not connected to my home network where I have the Ubuntu server and Time Machine backup share, but after connecting to my home network the backup proceeds as usual…

As far as I can remember, the original behavior of Time Machine wasn’t like this. I think (I could be wrong), that it detected that the current network wasn’t where the TM backup was, and it didn’t attempt to do the backup until it was back to its “TM” network.

Does anyone have a solution to this? “How to stop TM from attempting to back-up every hour when the backup disk is not connected.”

It’s not a problem (for now), but a minor nuisance I’d say.

and thaks for sharing

This tutorial is so great in needs to be put up on a wiki. So that more devs can update it as the thread is kinda old.

for example i have

/media/movies “movies” allow:username cnidscheme:cdb options:upriv

doesnt seem to be working.

add_filter( 'post_thumbnail_html', 'my_post_image_html', 10, 3 );
function my_post_image_html( $html, $post_id, $post_image_id ) {
	$html = '<a href="' . get_permalink( $post_id ) . '" title="' . esc_attr( get_post_field( 'post_title', $post_id ) ) . '" rel="nofollow">' . $html . '</a>';
	return $html;
}

For all running Snow Leopard and wondering why the sparsebundle you create doesn’t work with Snow Leopard…

The sparsebundle doesn’t have to be named like computername_0014e3856bd0.sparsebundle. There does have to be a file included in the sparsebundle referring to the UUID of your machine.

A guide on how to create the sparsebundle can be found at http://www.insanelymac.com/forum/index.php?showtopic=184462

Is there a HOWTO for enabling all this services over VPN? Preferable pptp?

It works great over ethernet, but once connected over VPN(pptp) it stop working.

Regards Runar

So how does one get this thumbnail in a post that gets wrapped by text?