Category Archives: Mac

MacBook Pro with Thunderbolt 3 and Dell docking stations

I recently got a new MacBook Pro with Thunderbolt 3 ports. These ports are great because they can carry power, USB and output to an external display simultaneously. Not a lot of accessories are available on the market yet though besides simple adapters from USB-C to things like USB, HDMI, VGA, Thunderbolt 2, or Ethernet. Belkin has announced the Belkin Thunderbolt 3 Express Dock HD, but it seems like it will be priced upwards of $300. Dell’s XPS 13 and 15 series however has included Thunderbolt 3 for a year now and Dell makes some nice docking stations for them:

Dell DA200: small portable device with Gigabit Ethernet, USB 3, VGA and HDMI.

Dell WD15: stationary device with a size similar to a paperback book, with a 130W or 180W power brick. Over the DA200, it adds mini-DisplayPort, 2x USB2, 2x USB3, Speaker and Headset outputs. It also passes power through to the computer.

Dell TB15: This model appears to have been recalled because it didn’t run stable and replaced with the TB16.

Dell TB16: stationary device shaped like a stack of a dozen CD cases, with a 180W or 240W power brick. Over the WD15, it adds DisplayPort. Note that it connects to the computer via Thunderbolt 3 instead of USB-C, which means it can deliver higher screen resolutions (see below).


First of all, the DA200 just works. VGA and HDMI run up to a resolution of 2048×1152 pixels at 60 Hz (even though Dell says it only supports 1920×1080), Ethernet works without installing a driver (it contains the same Realtek chip, PCI ID 0bda:8153, that the official Belkin USB-C adapter uses). HDCP-encrypted HDMI works fine, as confirmed by starting a Netflix video.

Next up, the WD15. I first hooked it up via mini-DisplayPort and was disappointed to find that it only runs up to 2048×1152 pixels. Switching to HDMI alleviated the problem and my screen ran at its native 2560×1440 pixels — though the colors were all messed up because the computer was outputting YCrCb while the screen was interpreting that as RGB, but that is easily fixed by creating an EDID override. HDCP works just fine — even though Dell says the dock doesn’t support it. Dell also says that the dock can do 4K resolutions (3840×2160 pixel) only at 30 Hz, so if you want to hook up a 4K display, don’t get this dock. The dock has a power button, but it didn’t surprise me to find out that it didn’t power up the Macbook. The Macbook did, however, automatically turn on when plugging in the USB-C cable, even in clamshell mode. One oddity I found was that the front left USB 3 port would only deliver power, not data, while a device was plugged into the rear USB 3 port — while I did not find this documented anywhere, it seems likely that this is a hardware limitation and not a Mac issue — see below for more information. Audio works fine too (it’s a Realtek chip, PCI ID 0bda:4014), but you need to use Audio MIDI Setup and click the “Configure Speakers” to assign left/right either to the first stream (headphone output on front) or the second stream (audio output on the back):

 

Finally, the Dell TB16. I didn’t have this one for testing because the WD15 suffices for my application. Dell says that you can run two 4K displays or one 5K (5120×2880 pixel) display, all at 60 Hz. I assume everything else will work just as it does with the WD15. For USB purposes however, I believe this dock contains a PCIe-attached USB 3.1 chip as the USB-C Alternate Mode Partner Matrix shows that a superspeed USB signal cannot be carried if high-resolution video is transferred. According to one of my readers, it doesn’t work out of the box unfortunately.


USB 3 problems

It seems like I can’t reliably use all USB 3 ports on the WD15 simultaneously. Sometimes they all work, but after the next reboot or unplug/replug cycle, once ceases to work. I had initially believed this to be a hardware limitation, but it appears to be a bug in Apple’s USB 3 drivers. sudo dmesg shows messages like these:

000069.247426 IOUSBHostHIDDevice: IOUSBHostHIDDevice::interruptRetry: resetting device due to IO failures

000069.551400 AppleUSB20Hub@14440000: AppleUSBHub::deviceRequest: resetting due to persistent errors

So the OS shuts down the USB hub. If you google these messages, you’ll find a large number of reports of it occurring with all kinds of USB hubs, but no solutions. So Apple just needs to get their drivers fixed.

Disabling “secured” IPv6 addresses is macOS 10.12 Sierra

On older macOS versions, every network interface would have one IPv6 address autogenerated from its MAC address, easily identified by the characteristic “ff:fe” bytes in the middle of the host part:
$ ifconfig en0
[...]
ether 10:dd:b1:9f:6b:ba
inet6 fe80::12dd:b1ff:fe9f:6bba%en0 prefixlen 64 scopeid 0x4
inet6 2001:7c0:2012:4a:12dd:b1ff:fe9f:6bba prefixlen 64 autoconf
[...]

Since macOS 10.12 however, these were replaced with randomly-generated “secured” addresses:
$ ifconfig en0
[...]
ether 10:dd:b1:9b:d0:67
inet6 fe80::46:3b36:146:9857%en0 prefixlen 64 secured scopeid 0x4
inet6 2001:7c0:2012:4a:4e6:f1d1:dd90:c6b4 prefixlen 64 autoconf secured
[...]

Very little is known about these, besides a single mailing list post that discovered them. If you are running a server, you’ll want your IPv6 address to be deterministic so you can register it in DNS. Therefore, we need to revert to pre-10.12 behavior:

$ echo net.inet6.send.opmode=0 >> /etc/sysctl.conf
$ reboot

If you look at the source code of the XNU kernel (Search for the IN6_IFF_SECURED flag) and the IPConfiguration service in macOS 10.11 (the 10.12 source code hasn’t been released yet), you can see that the new behavior was already there, just not enabled by default like it is now. Also, we now know that the change wasn’t made to reflect RFC 7217 (Semantically Opaque Interface Identifiers) behavior, but rather implements RFC 3972 (Cryptographically Generated Addresses).

Triple-booting a Clamshell iBook

I have a first-generation Clamshell iBook, manufactured in January 2000. These shipped with 64 MB of on-board RAM, a 6 GB hard drive and a CD drive.
I got it in summer 2005 when my high school decommissioned these computers. Running Mac OS X 10.3.9 at the time, it was still a very usable computer, especially after I upgraded the RAM to 320 MB. I used it as my main computer for a year and a half, until I replaced it with a new Intel-based Mac.

Now, in 2016, I found a 40 GB IDE laptop hard drive and decided to put that into the Clamshell and try to make it usable again.
As usual, iFixit has a good tutorial. You have to take the entire machine apart and end up with around 50 screws, but it’s surprisingly easy. You only need a Torx T8, a Philips #1 and a 5mm nut driver.
Now, on to putting some operating systems on the hard drive.

Installing Mac OS X 10.4

Here is what you need:

  • Mac OS X 10.4 retail install DVD
  • Mac OS X 10.3 retail install CD (the first one is sufficient). Any other bootable Mac OS X 10.3 CD might also work.
  • USB hard drive

Officially, Tiger is only supported on the latest revision of the Clamshell iBook, which has FireWire and a DVD drive. Mine only has a CD drive and does not support USB booting, which makes things exceedingly difficult.

  1. Take an image of the 10.4 install DVD and save it onto the USB hard drive using a different Mac.
  2. Boot the Clamshell off the 10.3 install CD
  3. Open up Disk Utility. Partition the disk with the following sizes and names: 16 MB linux-boot, 2 GB linux-swap, 12 GB linux, 20 GB Macintosh HD, 6 GB Install. Make sure to select the Install OS 9 drivers checkbox.
  4. Plug in the USB hard drive.
  5. Clone the 10.4 install image onto the Install partition on the internal drive.
  6. Open a Terminal and bless the hard drive: bless --folder "/Volumes/Install/System/Library/CoreServices" --bootinfo "/Volumes/Install/usr/standalone/ppc/bootx.bootinfo".
  7. Using the Terminal, pico /Volumes/Install/System/Installation/Packages/OSInstall.mpkg/Contents/OSInstall.dist and remove the PowerBook2,1 from the badMachines list.
  8. Open Startup Disk and set the iBook to boot from the Install partition.
  9. Reboot
  10. Install 10.4 onto the Macintosh HD partition, deselecting all printer drivers and languages you don’t need in order to save some space.
  11. After the reboot, run Software Update a few times.

Installing Mac OS 9.2.2

Here is what you need:

Mac OS 9 is really easy to install, you just copy it to the hard drive.

  1. Mount the CD or image and copy System Folder and Applications (Mac OS 9) from the root directory to your Macintosh HD.
  2. Open the Startup Disk preference pane and select Mac OS 9.2.2.
  3. Reboot.
  4. If you used the NetBoot image, you will be asked for user name and password. Use NBUser and netboot.
  5. In order to disable the password prompt, go to the hard drive, System Folder and move the contents of Control Panels disabled to Control Panels. Now open the Multiple Users control panel, go to options and set it to local authentication. Then, disable multiple users entirely.
  6. Download and install QuickTime 6.0.3 for Mac.
  7. Open the Startup Disk control panel and select Mac OS 10.4.
  8. After the reboot, open the Classic preference pane and select the system folder on Macintosh HD.

Now you can use Classic and you can also do a native boot.

Installing Debian Linux 8 “Jessie”

Mac OS 9.2.2 hasn’t received updates in almost 15 years, and Mac OS X 10.4 is also 7 years beyond its update cycle. So how about a current operating system?

Here is what you need:

Except for the partitioning, the installation is quite straight-forward.

  1. Insert the CD and boot it by holding the C key while turning on the computer.
  2. Proceed through the installer.
  3. When asked for the partitioning, set the 16 MB partition to be NewWorld Boot and set the bootable flag. Set the 2 GB partition to be swap. Set the 12 GB partition to be ext3 and mounted at /.
  4. When asked for the packages to install, choose the Xfce desktop. It’s lightweight enough to run on this old hardware.
  5. After rebooting, run nano /etc/apt/sources.list and add the non-free repository to all entries in the file. Then, install the AirPort firmware: apt-get install linux-firmware-nonfree.

The resulting Debian install mostly worked for me, except for two things:

  • the AirPort card does not show up in NetworkManager
  • Fonts are a bit messed up. Bitmaps are drawn to the screen just fine.

I’m still looking for solutions to these two things.

Configuring the yaboot bootloader

Right now, your computer will boot into Linux all the time.
While the yaboot loader prompts you to hit x to boot into Mac OS X and m to boot into Mac OS 9, that doesn’t work as expected.
Instead of booting into Mac OS X, my iBook booted into the install partition.
And instead of booting into Mac OS 9, my iBook booted into Mac OS X.

So run nano /etc/yaboot.conf and adjust the macosx= line so that it refers to the same partition as the macos9= line. Exit the editor and run ybin -v to apply the changes. That fixes the first problem.

Now, reboot into Mac OS X and open the Startup Disk preference pane. Select Mac OS 9.2.2 and reboot.
While rebooting, hold down the option key and select the Linux partition at the boot picker.
Run nano /etc/yaboot.conf again and add brokenosx to a new line. Once more, run ybin -v.

Now, each boot loader entry does what you’d expect. The trick here is that brokenosx causes yaboot to directly load the Mac OS X booter for the macosx= entry. The macos= entry, on the other hand, will still cause the blessed system folder to be booted.

Fixing Mac OS 9 after installing Linux

One problem is still there: When you try to boot Mac OS 9, you are greeted by a blinking floppy with a question mark. This happens because the Debian partitioner destroys the Mac OS 9 drivers for the HFS+ partitions. However, the drivers can be reinstalled.

You need:

  • Mac OS 9 install CD. Any other bootable Mac OS 9 CD might also work.

Now,

  1. Insert the CD and reboot the iBook while holding down the C key.
  2. Open Drive Setup, highlight the internal hard drive, go to the Functions menu and click Update Drivers.
  3. Reboot

Resetting the Startup Disk

Whenever you touch the Startup Disk preference pane in Mac OS X or the Startup Disk control panel in Mac OS 9, your system will no longer show the yaboot prompt when you turn it on. To fix that, do the following:

  1. Hold down the option key while booting and select the Linux partition
  2. Run ybin -v

Fixing Linux problems

http://ppcluddite.blogspot.de/2012/03/installing-debian-linux-on-ppc-part-iv.html is an excellent article that explains how to fix most issues that Debian has on PowerPC. For example, to fix the font rendering troubles, create an xorg.conf file (switch to a text terminal, run init 3, Xorg -configure, cp /root/xorg.conf.new /etc/X11/xorg.conf) and insert Option "RenderAccel" "false" into its Device section.

Digest authentication freezes Apache on OS X Server 5.0

When running a web site on OS X Server 5.0 for a while (on OS X 10.10.5 in my case), eventually you’ll notice hundreds of httpd processes in Activity Monitor. One or two might cause a bit of CPU load, while the others don’t do anything. When you try to load the web page, it is insanely slow and often throws a HTTP 500 error, a HTTP 502 Proxy Error, or the connection just times out. /var/log/apache2/error_log reports errors like

[Thu Nov 05 13:15:24.435549 2015] [mpm_prefork:error] [pid 60920] AH00161: server reached MaxRequestWorkers setting, consider raising the MaxRequestWorkers setting.

but that’s the only hint you get. To find out more, add the following lines inside the VirtualHost section of /Library/Server/Web/Config/apache2/sites/0000_127.0.0.1_34580_.conf and restart the Websites service in Server.app:

<Location /server-status>
SetHandler server-status
</Location>

Now, got to http://localhost/server-status and refresh it occasionally while traffic hits your web site. You’ll eventually see dozens of lines like the one below (starting with 52-0 in my case):

Scoreboard Key:
"_" Waiting for Connection, "S" Starting up, "R" Reading Request,
"W" Sending Reply, "K" Keepalive (read), "D" DNS Lookup,
"C" Closing connection, "L" Logging, "G" Gracefully finishing,
"I" Idle cleanup of worker, "." Open slot with no current process
Srv PID Acc M   CPU SS  Req Conn    Child   Slot    Client  VHost   Request

52-0    80825   0/2/2   C   0.45    51  0   0.0 0.02    0.02    ::1

All httpd processes you’re seeing in Activity Monitor are stuck in “Closing connection”, except for those that cause considerable CPU load. If the server were behaving correctly, you wouldn’t have as many processes and those that aren’t currently handling requests would either be “Waiting for Connection” or “Open slot with no current process”.

Let’s fire up a debugger to see what’s causing the processes to get stuck:

$ sudo lldb -p 80825
(lldb) process attach --pid 80825
Process 80825 stopped
* thread #1: tid = 0x3a69d4, 0x00007fff8fe35902 libsystem_kernel.dylib`__wait4 + 10, queue = 'com.apple.main-thread', stop reason = signal SIGSTOP
    frame #0: 0x00007fff8fe35902 libsystem_kernel.dylib`__wait4 + 10
libsystem_kernel.dylib`__wait4:
->  0x7fff8fe35902 <+10>: jae    0x7fff8fe3590c            ; <+20>
    0x7fff8fe35904 <+12>: movq   %rax, %rdi
    0x7fff8fe35907 <+15>: jmp    0x7fff8fe30c78            ; cerror
    0x7fff8fe3590c <+20>: retq   

Executable module set to "/usr/sbin/httpd".
Architecture set to: x86_64-apple-macosx.
(lldb) bt
* thread #1: tid = 0x3a69d4, 0x00007fff8fe35902 libsystem_kernel.dylib`__wait4 + 10, queue = 'com.apple.main-thread', stop reason = signal SIGSTOP
  * frame #0: 0x00007fff8fe35902 libsystem_kernel.dylib`__wait4 + 10
    frame #1: 0x0000000109b3ee95 libapr-1.0.dylib`apr_proc_wait + 70
    frame #2: 0x000000010aa4290c mod_auth_digest_apple.so`cleanup_server_event + 73
    frame #3: 0x0000000109b3627a libapr-1.0.dylib`apr_pool_destroy + 82
    frame #4: 0x0000000109a74ce2 httpd`clean_child_exit + 50
    frame #5: 0x0000000109a74c59 httpd`child_main + 2393
    frame #6: 0x0000000109a73b2e httpd`make_child + 510
    frame #7: 0x0000000109a74181 httpd`perform_idle_server_maintenance + 1265
    frame #8: 0x0000000109a72887 httpd`prefork_run + 2471
    frame #9: 0x0000000109a26328 httpd`ap_run_mpm + 120
    frame #10: 0x0000000109a1185f httpd`main + 4687
    frame #11: 0x00007fff8d7435c9 libdyld.dylib`start + 1
(lldb) continue
Process 82070 resuming
(lldb) exit

Ah, so mod_auth_digest_apple.so is the culprit. So in all *.conf files in /Library/Server/Web/Config/apache2 and its subdirectories, replace every occurrence of AuthType Digest with AuthType Basic and comment out all lines containing mod_auth_digest_apple.so by prepending a # character. Restart the Websites service in Server.app. Watch http://localhost/server-status: everything should be fine now and no more connections stuck on “Closing connection”!

Printer Sharing randomly stops working due to memorystatus_thread killing cupsd

Printers shared in OS X 10.9, 10.10 or 10.11 randomly stop being accessible from remote computers. The system log reports that the cupsd process was terminated by memorystatus_thread:

Oct 26 07:14:33 robert kernel[0]: memorystatus_thread: idle exiting pid 4901 [cupsd]

This can also manually be triggered by executing sudo launchctl stop org.cups.cupsd.plist and manually and temporarily fixed by executing sudo launchctl start org.cups.cupsd.plist.

As a workaround, you can edit the CUPS LaunchDaemon to restart whenever a remote computer attempts to connect. On 10.9 or 10.10:

sudo /usr/libexec/PlistBuddy -c "Delete Sockets:Listeners:0:SockNodeName" /System/Library/LaunchDaemons/org.cups.cupsd.plist
sudo /usr/libexec/PlistBuddy -c "Delete Sockets:Listeners:1:SockNodeName" /System/Library/LaunchDaemons/org.cups.cupsd.plist
sudo launchctl unload /System/Library/LaunchDaemons/org.cups.cupsd.plist
sudo launchctl load /System/Library/LaunchDaemons/org.cups.cupsd.plist

On 10.11, first disable System Integrity Protection in Recovery mode, then run

sudo /usr/libexec/PlistBuddy -c "Add Sockets:Listeners:1 Dict" /System/Library/LaunchDaemons/org.cups.cupsd.plist/org.cups.cupsd.plist
sudo /usr/libexec/PlistBuddy -c "Add Sockets:Listeners:1:SockServiceName String" /System/Library/LaunchDaemons/org.cups.cupsd.plist/org.cups.cupsd.plist
sudo /usr/libexec/PlistBuddy -c "Set Sockets:Listeners:1:SockServiceName ipp "/System/Library/LaunchDaemons/org.cups.cupsd.plist/org.cups.cupsd.plist
sudo launchctl unload /System/Library/LaunchDaemons/org.cups.cupsd.plist
sudo launchctl load /System/Library/LaunchDaemons/org.cups.cupsd.plist

Multiple GPUs on unsupported Mac Pro

The first two generations of Apple’s Mac Pro, the MacPro1,1 and MacPro2,1, do not officially run OS X later than 10.7.5. However, there is a modified EFI bootloader available which emulates the EFI64 interface on EFI32 machines. The original version available at Google Code supports OS X 10.9, and there’s a newer one available at Github which also does OS X 10.10. You simply drop in a new boot.efi in two places and add your board ID to the list of supported systems.

The resulting system works perfectly fine, which makes me wonder why Apple didn’t come up with a solution like this themselves. In any case, it extends the life of 2006 and 2007 Mac Pros beyond last year’s end-of-support for OS X Lion.

Since OS X 10.7 and higher included graphics drivers that not only supported the official Apple-supplied GPUs with EFI-compatible firmwares, but pretty much any off-the-shelf Nvidia or AMD GPU, GPUs have become quite easy to upgrade in Mac Pros (the classic tower cheese grater Mac Pro, not the new black trash can Mac Pro). The only thing you lose is the boot screen, so you still need to keep around that original GPU to debug the machine if it doesn’t boot.

These upgraded GPUs work fine with the modified bootloader as well, however if you try to install multiple GPUs (e.g. if you want to drive more than two displays or develop CUDA code and would like to run it in the debugger), only one of them will actually output video.

The solution to make multiple GPUs work in Macs running OS X versions they don’t officially support is surprisingly simple:

sudo /usr/libexec/PlistBuddy -c "Add :IOKitPersonalities:AppleGraphicsDevicePolicy:ConfigMap:Mac-F4208DA9 string none" /System/Library/Extensions/AppleGraphicsControl.kext/Contents/PlugIns/AppleGraphicsDevicePolicy.kext/Contents/Info.plist
sudo /usr/libexec/PlistBuddy -c "Add :IOKitPersonalities:AppleGraphicsDevicePolicy:ConfigMap:Mac-F4208DC8 string none" /System/Library/Extensions/AppleGraphicsControl.kext/Contents/PlugIns/AppleGraphicsDevicePolicy.kext/Contents/Info.plist
sudo touch /System/Library/Extensions

So here we are, running a 2006 MacPro1,1 with two EVGA Nvidia GT610 1GB cards, driving three Apple Cinema Displays.

Fixing OS X Server Push Mail

OS X Server 10.7 and later support push mail for iOS devices. This mechanism is neither based on IMAP IDLE (which iOS doesn’t support) nor Exchange ActiveSync (EAS), but on Apple’s Push Notification Service (APNS) infrastructure.

After setting up Mail using the GUI in OS X Server 10.10 Yosemite, I wondered why push didn’t work. From my understanding, it should happen automatically. The only indications something was wrong were the following lines in /Library/Logs/Mail/push_notify.log:

Feb 21 20:13:27 server.example.com push_notify[22848]: ApplePushServiceProvider: Warning: no device map found for 3F2504E0-4F89-41D3-9A0C-0305E82C3301

as well as XAPPLEPUSHSERVICE missing from the IMAP capabilities list:

$ openssl s_client -quiet -connect localhost:993
* OK [CAPABILITY IMAP4rev1 LITERAL+ SASL-IR LOGIN-REFERRALS ID ENABLE IDLE AUTH=PLAIN AUTH=LOGIN] Dovecot ready.

This is often the point where you have to break out the disassembler to find out what is wrong. Luckily however, Dovecot is open source, including the modifications Apple made to support APNS. Tracing through the code, the message above is logged if /Library/Server/Mail/Data/mta/guid_device_maps.plist does not contain a section for the user to which the incoming email is addressed. This section is written when Dovecot receives an XAPPLEPUSHSERVICE command. This command is probably only sent by a client when the XAPPLEPUSHSERVICE capability is reported by the server. The reason why the server didn’t report the capability was a simple incorrect (default) setting, easily fixable using

sudo serveradmin settings mail:imap:aps_topic_enabled = yes

Push mail immediately started working for me after this command, and the capability is correctly reported:

$ openssl s_client -quiet -connect localhost:993
* OK [CAPABILITY IMAP4rev1 LITERAL+ SASL-IR LOGIN-REFERRALS ID ENABLE IDLE XAPPLEPUSHSERVICE AUTH=PLAIN AUTH=LOGIN] Dovecot ready.

PHP 5: ldap_search never returns when searching Active Directory

I recently moved a PHP web application from a server running PHP 5.3 on Mac OS X 10.6 to a newer one with PHP 5.4 on Mac OS X 10.9. This caused the following code sample, run against an Active Directory server, to hang at the ldap_search() call:

$conn = ldap_connect('ldaps://' . $LDAPSERVER);
ldap_set_option($conn, LDAP_OPT_PROTOCOL_VERSION, 3);
$bind = @ldap_bind($conn, $LDAPUSER, $LDAPPW);
$result = ldap_search($conn, $LDAPSEARCHBASE, '(&(samaccountname=' . $searchuser . '))');
$info = ldap_get_entries($conn, $result);
ldap_close($conn);

Wiresharking the connection between web server and LDAP server (after replacing ldaps:// with ldap://) showed:

bindRequest(1) "$LDAPUSER" simplebindResponse(1) success searchRequest82) "$LDAPSEARCHBASE" wholeSubtree
searchResEntry(2) "CN=$searchuser,...,$LDAPSEARCHBASE" | searchResRef(2) | searchResDone(2) success [1 result]
bindRequest(4) "" simple
bindResponse(4) success
searchRequest(3) "DC=DomainDnsZones,$LDAPSEARCHBASE" wholeSubtree
searchResDone(3) operationsError (000004DC: LdapErr: DSID-0C0906E8, comment: In order to perform this operation a successful bind must be complete on the connection., data0,

So it’s binding, receiving a success response, searching and then receiving a response and a referrer to DC=DomainDnsZones,$LDAPSEARCHBASE. Next, it opens a new TCP connection and follows the referrer, but does an anonymous bind.

The solution is simple: just add

ldap_set_option($conn, LDAP_OPT_REFERRALS, FALSE);

after line 2. If for some reason you actually need to follow the referrer, have a look at ldap_set_rebind_proc, which lets you specify a callback which then does the authentication upon rebind.

Update August 2015: Same goes when using Net_LDAP3, which is used e.g. by Roundcube’s LDAP integration. Here you need to add the following:

$config['ldap_public']['public'] = array(
[...]
 'referrals' => false,
);

CUPS-to-CUPS printing with server-side processing and page_log

Printer sharing on Windows is easy: the client receives the driver from the server, presents the driver GUI and passes on an intermediate format along with the options selected in the driver to the server, which then renders the print job for the printer (usually into PostScript).

In the Unix (Mac OS X in my case, but Linux would be the same) world, CUPS is commonly used for printing. It’s very powerful, but I find the documentation severely lacks details about the exact way something is implemented in the code. Luckily, the code is open-source and Michael Sweet, the developer of CUPS who now works at Apple and still maintains CUPS, managed to create a very structured piece of software with code that’s reasonably easy to understand.

If you just add a CUPS server’s print queue as a new printer on a CUPS client, it will work fine, but you might run into some inconveniences:

Problems

  1. The job might get run through a vendor-supplied filter twice, once on the server and once on the client. This usually works fine, but the print job might significantly increase in size (observed on an HP LaserJet).

  2. The page_log on the server might not contain the number of pages and copies a job consisted of and list 1 for both instead.

  3. The page_log on the server might not contain things like page format, duplex status or attributes you manually added to PrintLogFormat.

Reasons

  1. This happens if the PPD both on the client and on the server contains a line starting with *cupsFilter, which links to a vendor-supplied filter. Such a filter usually produces a MIME type of application/postscript.

  2. This happens if the job does not get run through the pstops filter by CUPS. CUPS bypasses that filter if the client submits the job with a MIME type of application/vnd.cups-postscript, i.e. it was already run through pstops on the client.

  3. This is either caused by the same things as (1) or (2), but I’m not sure which one.

Solution

Simply add the following lines to the PPD on the client. That way, it passes the job straight to the server for server-side processing.

*cupsFilter: "application/pdf 0 -"
*cupsFilter: "image/* 0 -"
*cupsFilter: "application/postscript 0 -"
*cupsFilter: "application/vnd.cups-postscript 0 -"
*cupsFilter: "application/vnd.cups-command 0 -"

By the way, if you use Mac OS X and let the “Add Printer” wizard automatically add a print queue from a remote CUPS server discovered via Bonjour, this is exactly what it does.

Notes

If you append something like %{SelectColor} to your PageLogFormat because that’s the attribute your printer uses to determine whether it should print in color or grayscale and you’d like to log that, please note that the default value (either as specified by the PPD or as specified by you via lpadmin -d printername -d SelectColor=Grayscale or via the CUPS web interface’s “Set Printer Defaults”) will never be written to the page_log. Only deviations from the default value will be logged. The defaults set on the server-side CUPS do not matter here, this is determined by the client-side CUPS.

Per the filter(7) documentation (italic comments were added by me):

Options passed on the command-line typically do not include the default choices the printer’s PPD file. […] use the ppdMarkDefaults [which sets all options to the defaults specified inside the PPD] and cupsMarkOptions [which sets the options to the values specified in the driver GUI] functions in the CUPS library to use the correct mapping, and ppdFindMarkedChoice [which reads from the options array composed from the defaults and the selected options] to get the user-selected choice.

CUPS on OS X hangs after a few days, reports “Internal Server Error”

If you set up CUPS on an OS X Server (version 10.8.5 in my case, but anything from 10.7 (where CUPS introduced sandboxing) through 10.9 (the current version) should exhibit this behavior), i.e. you enable Printer Sharing in System Preferences and run sudo cupsctl WebInterface=yes, and leave the system running for a few days, you’ll eventually run into the situation that http://localhost:631/printers will report “Internal Server Error”, and clients will no longer be able to print to the server.

Digging around CUPS’ debug log, you’ll see something like
D [27/Oct/2013:13:33:52 +0100] [CGI] sandbox_init failed: /private/tmp/05d735269fa67: No such file or directory (No such file or directory)
D [27/Oct/2013:13:33:52 +0100] PID 78980 (/usr/libexec/cups/cgi-bin/printers.cgi) stopped with status 1.

That missing file (named a different 13-digit hexadecimal name upon each restart) is the CUPS daemon’s sandbox profile.

Digging around further reveals that /var/log/daily.out contains exactly this file name:
Sun Oct 27 03:15:01 CET 2013
Removing old temporary files:
/tmp/05d735269fa67
[...]

All we need to do to prevent this from happening in the future is opening /etc/periodic/daily/110.clean-tmps in your favorite text editor and adding the line printed in bold:
set -f noglob
args="-atime +$daily_clean_tmps_days -mtime +$daily_clean_tmps_days"
args="${args} -ctime +$daily_clean_tmps_days"
args="${args} ! -group _lp ! -user _lp"
dargs="-empty -mtime +$daily_clean_tmps_days"
dargs="${dargs} ! -name .vfs_rsrc_streams_*"

Update February 2014: CUPS 1.7.1 is supposed to fix that issue; the release notes mention my reported bug. Now lets see how long it takes until Apple ships the updated CUPS with an OS X update.

Update March 2014: I just upgraded our server to OS X 10.9.2 and got CUPS 1.7.1 with it. Hooray, less than three months between bug reported and fix deployed. The sandbox profile now gets written to /var/spool/cups/tmp. In fact, that’s exactly what was changed in scheduler/conf.c in the CUPS source code: they added setenv("TMPDIR", TempDir, 1);