Monthly Archives: May 2008

msttcorefonts on FreeBSD and other stuff

For those who come from a Linux background, Debian and Ubuntu in particular, – msttcorefonts is named webfonts on FreeBSD.

I have just discovered that installing the port from package <i<removes firefox. (Don’t know if the same is observed when building from source.) Anyway, be warned: install webfonts before firefox, or you’ll waste another hour compling thing you shouldn’t be.

This is because portinstall cleans up by default. You can prevent this:

-w --noclean
Do not "make clean" before each build. See the -c option above.
-W --nocleanup
Do not "make clean" after each installation. See the -C option above.

So, I put an alias in my shell’s rc file and save time by sacrificing disk space.

Advertisements

Remove a port and all its dependencies

pkg_deinstall -R

From man pkg_deinstall:

-R
--upward-recursive Deinstall all those packages required by the given
packages as well.

pkg_deinstall is a part of portupgrade.

Firefox won’t start: typing “firefox”, nothing happens

Check if the ~/.mozilla folder is not owned by root. In such a case, absolutely nothing is printed when you want to start it:

% firefox
%

Firefox, Thunderbird, Sunbird report profile in use

An awesome explanation lies elsewhere.

One reason it may happen (which they don’t mention for the obvious rarity of it) is your drive letter (in Windows, of course) having changed.

That was the case for me after some re-partitioning done.

Unresolved trouble with hard drive in Debian

I had a really poorly partitioned FreeBSD slice, where I would always run out of space on /usr. Also, there are more pre-compiled desktop packages for Linux than there are for FreeBSD. I mean EclipseTrader and RSSOwl 2 mostly, but it’s also the BOINC server and BOINC project executables.

So, I decided to format the slice and try Debian again. Trying to find an unused CD or DVD was a pain.

After installing it, I immediately remembered the reason why I dumped Debian/Ubuntu in favor of FreeBSD in the first place: the nasty hard drive timeouts. But it’s not just timeouts: it’s the driver gradually reducing the mode from UDMA/133 to the lowest low of PIOs.

May 30 17:18:49 heaven kernel: ata2: SATA max UDMA/133 cmd 0x1F0 ctl 0x3F6 bmdma 0xFFA0 irq 14
May 30 17:18:49 heaven kernel: scsi1 : ata_piix
May 30 17:18:49 heaven kernel: ATA: abnormal status 0x7F on port 0x1F7
May 30 17:18:49 heaven kernel: ata3: PATA max UDMA/100 cmd 0x170 ctl 0x376 bmdma 0xFFA8 irq 15
May 30 17:18:49 heaven kernel: scsi2 : ata_piix
May 30 17:18:49 heaven kernel: ata3.00: ATA-6, max UDMA/100, 156301488 sectors: LBA48
May 30 17:18:49 heaven kernel: ata3.00: ata3: dev 0 multi count 16
May 30 17:18:49 heaven kernel: usb 5-6: new high speed USB device using ehci_hcd and address 2
May 30 17:18:49 heaven kernel: usb 5-6: configuration #1 chosen from 1 choice
May 30 17:18:49 heaven kernel: ata3.01: ATAPI, max UDMA/33
May 30 17:18:49 heaven kernel: ata3.00: configured for UDMA/33
May 30 17:18:49 heaven kernel: usb 4-2: USB disconnect, address 2
May 30 17:18:49 heaven kernel: ata3.01: configured for UDMA/33
May 30 17:18:49 heaven kernel: Vendor: ATA Model: HTS541080G9AT00 Rev: MB4O
May 30 17:18:49 heaven kernel: Type: Direct-Access ANSI SCSI revision: 05
May 30 17:18:49 heaven kernel: Vendor: TSSTcorp Model: CD/DVDW SN-S082D Rev: SS03
May 30 17:18:49 heaven kernel: Type: CD-ROM ANSI SCSI revision: 05
May 30 17:18:49 heaven kernel: ACPI: PCI Interrupt 0000:06:01.0[A] -> GSI 16 (level, low) -> IRQ 169
...
May 30 17:18:49 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:18:49 heaven kernel: sda: Write Protect is off
May 30 17:18:49 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:18:49 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:18:49 heaven kernel: sda: Write Protect is off
May 30 17:18:49 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:20:57 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:20:57 heaven kernel: ata3: soft resetting port
May 30 17:20:57 heaven kernel: ata3.00: configured for UDMA/33
May 30 17:20:57 heaven kernel: ata3.01: configured for UDMA/33
May 30 17:20:57 heaven kernel: ata3: EH complete
May 30 17:20:57 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:20:57 heaven kernel: sda: Write Protect is off
May 30 17:20:57 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:20:57 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:20:57 heaven kernel: sda: Write Protect is off
May 30 17:20:57 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:24:34 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:24:57 heaven kernel: ata3: soft resetting port
May 30 17:24:58 heaven kernel: ata3.00: configured for UDMA/33
May 30 17:24:58 heaven kernel: ata3.01: configured for UDMA/33
May 30 17:24:58 heaven kernel: ata3: EH complete
May 30 17:24:58 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:24:58 heaven kernel: sda: Write Protect is off
May 30 17:24:58 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:24:58 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:24:58 heaven kernel: sda: Write Protect is off
May 30 17:24:58 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:26:45 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:27:00 heaven kernel: usb 4-2: USB disconnect, address 3
May 30 17:27:08 heaven kernel: ata3: soft resetting port
May 30 17:27:08 heaven kernel: ata3.00: configured for UDMA/33
May 30 17:27:08 heaven kernel: ata3.01: configured for UDMA/33
May 30 17:27:08 heaven kernel: ata3: EH complete
May 30 17:27:08 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:27:08 heaven kernel: sda: Write Protect is off
May 30 17:27:08 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:27:09 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:27:09 heaven kernel: sda: Write Protect is off
May 30 17:27:09 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:28:45 heaven kernel: ata3.01: limiting speed to UDMA/25
May 30 17:28:45 heaven kernel: ata3: port is slow to respond, please be patient
May
30 17:28:45 heaven kernel: ata3: soft resetting port
May 30 17:28:45 heaven kernel: ata3.00: configured for UDMA/25
May 30 17:28:45 heaven kernel: ata3.01: configured for UDMA/25
May 30 17:28:45 heaven kernel: ata3: EH complete
May 30 17:28:45 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:28:45 heaven kernel: sda: Write Protect is off
May 30 17:28:45 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:28:45 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:28:45 heaven kernel: sda: Write Protect is off
May 30 17:28:45 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:30:56 heaven kernel: ata3.01: limiting speed to UDMA/16
May 30 17:31:27 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:31:27 heaven kernel: ata3: soft resetting port
May 30 17:31:27 heaven kernel: ata3.00: configured for UDMA/16
May 30 17:31:27 heaven kernel: ata3.01: configured for UDMA/16
May 30 17:31:27 heaven kernel: ata3: EH complete
May 30 17:31:27 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:31:27 heaven kernel: sda: Write Protect is off
May 30 17:31:27 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:31:27 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:31:27 heaven kernel: sda: Write Protect is off
May 30 17:31:27 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:32:13 heaven kernel: ata3.01: limiting speed to PIO4
May 30 17:32:44 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:32:44 heaven kernel: ata3: soft resetting port
May 30 17:32:44 heaven kernel: ata3.00: configured for PIO4
May 30 17:32:44 heaven kernel: ata3.01: configured for PIO4
May 30 17:32:44 heaven kernel: ata3: EH complete
May 30 17:32:44 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:32:44 heaven kernel: sda: Write Protect is off
May 30 17:32:44 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:32:44 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:32:44 heaven kernel: sda: Write Protect is off
May 30 17:32:44 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:33:16 heaven kernel: ata3.01: limiting speed to PIO3
May 30 17:33:23 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:33:46 heaven kernel: ata3: soft resetting port
May 30 17:33:46 heaven kernel: ata3.00: configured for PIO3
May 30 17:33:46 heaven kernel: ata3.01: configured for PIO3
May 30 17:33:46 heaven kernel: ata3: EH complete
May 30 17:33:47 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:33:47 heaven kernel: sda: Write Protect is off
May 30 17:33:47 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:33:47 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:33:47 heaven kernel: sda: Write Protect is off
May 30 17:33:47 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:40:17 heaven kernel: ata3.01: limiting speed to PIO2
May 30 17:40:24 heaven kernel: ata3: port is slow to respond, please be patient
May 30 17:40:47 heaven kernel: ata3: soft resetting port
May 30 17:40:47 heaven kernel: ata3.00: configured for PIO2
May 30 17:40:47 heaven kernel: ata3.01: configured for PIO2
May 30 17:40:47 heaven kernel: ata3: EH complete
May 30 17:40:47 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:40:47 heaven kernel: sda: Write Protect is off
May 30 17:40:47 heaven kernel: SCSI device sda: drive cache: write back
May 30 17:40:47 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 17:40:47 heaven kernel: sda: Write Protect is off
May 30 17:40:47 heaven kernel: SCSI device sda: drive cache: write back
...
May 30 18:00:18 heaven -- MARK --
May 30 18:03:04 heaven kernel: ata3: port is slow to respond, please be patient
May 30 18:03:04 heaven kernel: ata3: soft resetting port
May 30 18:03:04 heaven kernel: ata3.00: configured for PIO2
May 30 18:03:04 heaven kernel: ata3.01: configured for PIO2
May 30 18:03:04 heaven kernel: ata3: EH complete
May 30 18:03:04 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 18:03:04 heaven kernel: sda: Write Protect is off
May 30 18:03:04 heaven kernel: SCSI device sda: drive cache: write back
May 30 18:03:04 heaven kernel: SCSI device sda: 156301488 512-byte hdwr sectors (80026 MB)
May 30 18:03:04 heaven kernel: sda: Write Protect is off
May 30 18:03:04 heaven kernel: SCSI device sda: drive cache: write back

IDE/ATA disk being recognized as SCSI, timeouts, lags, 2 Mb/s I/O speed… No, really, I’d rather dump this kind of usability and switch back to my FreeBSD, “where everything is different from how it should be on a normal Linux box”.

Oh, and I’m calling it “hell” this time. Quite unoriginal, of course, but does anyone care.

All torrent downloads corrupted

I tried Transmission, Deluge, Azureus (for a short while, though, nowhere near sufficient), and all the downloads were “corrupted”. They downloaded fine, no error shown in the client, but when I tried to listen to MP3s (all creative-commons, of course) or watch the movies (all public domain, of course), I was getting some jagged up sound or glitches/skips in the video stream.

First I thought libtorrent or something was to blame, or that the clients used don’t implement hashing/checking. Dumped these thoughts really soon.

The answer is simpler: I had the download folder on a FAT32 partition. Moved to UFS – the glitches disappeared.

So, it might be something related to the msdosfs driver, I don’t know.

Running EclipseTrader on FreeBSD

Both EclipseTrader websites seem to be somewhat laking in documentation, so it was kind of hard. There’re no directions as to how to install ET like RSSOwl have, and with all the various ways to create an Eclipse project, you’re kind of left in the dark. Anyway, to the point.

First, I had the Eclipse IDE (eclipse-devel port) installed. Then, I made sure I had a clean open perspective. I then even restarted the IDE and checked the Plug-ins (Window -> Show View… -> Other… -> PDE -> Plug-ins) had no errors or what not.

Then just:

File -> Import… -> Existing projects into workspace -> From project archive file…

And that’s it. Just build it (if you have “Build automatically” set, “Build” will be grayed out), then run it:

Run -> Open run dialog…

Set:

  1. Run a product: net.sourceforge.eclipsetrader.product
  2. Plug-ins tab:
  • Check “Workspace”
  • Uncheck “Target platform”
  • Click “Add Required Plug-ins”

Apply, run.

Now, this might seem really simple, but without any instructions, it’s somewhat of a hell if you installed Eclipse yesterday. I was getting all sorts of funky errors, like net.sourceforge.eclipsetrader.yahoo not being able to initialize its Feed and HistoryFeed classes when trying to load a chart, etc..

Edit:

An alternative way would be to check out from SVN. (You’ll need subclipse to use that from Eclipse IDE.) See this message in the forum for instructions, and the reply to it for the SVN address.

RSSOwl2 M7 on FreeBSD

I managed to run it from the Eclipse IDE, after installing the java/eclipse-devel port (had to use the -devel, since I needed version 3.3.1.1 at least, as specified on the RSSOwl project page). Then it’s pretty straightforward, just follow the instructions.

I am yet to find out how to build a stand-alone version. Eclipse RCP seems to be relevant.

Evince: Undefined symbol “dbus_watch_get_unix_fd”

I updated Evince, and then started getting the following error:

% evince Zeki98.pdf
/libexec/ld-elf.so.1: /usr/local/lib/gio/modules/libgvfsdbus.so: Undefined symbol "dbus_watch_get_unix_fd"

I tried upgrading the gvfs port, but it was up-to-date. Without any further thinking, I upgraded dbus (from dbus-1.0.2_2 to dbus-1.1.20), and the error went away.

Stale dependency: fusefs-kmod

Edit: you can also read this <dead link>.

I now get this every time I want to do portversion. Caught it after updating the ports CVS (ports-sysutils, to be precise), followed by skipping the installation of libgweather dirtily (which is vaguely related in this case).

But I can’t install fusefs-kmod. The reason is that in the Makefile, a check is made to see whether

/usr/src/sbin/mount

exists. If not, it complains about userland sources not being installed.

The problem was “introduced” in this discussion. (It solved someone’s trouble all right.)

I don’t know why the heck so many applications that had a dependency on fusefs-kmod (a kernel module for FuseFS-based file systems) passed without installing it on the system.

Anyway, I “fixed” it with

portdb -F

then deleting the stale dependency.