OpenBSD Journal

OpenBSD 4.2 released

Contributed by merdely on from the this-ones-for-itojun dept.

The OpenBSD team is pleased to announce the release of OpenBSD 4.2. See the announcement for more information.

This release is dedicated to the memory of long-time developer Jun-ichiro "itojun" Itoh Hagino, who focused his life on IPv6 deployment for everyone.

Place an order worldwide or order in Europe, or if you only download from FTP then make a donation. If you do FTP your release, be sure to use a local mirror and not the main ftp server:

o ftp.kd85.com  Austria       o ftp.stacken.kth.se  Sweden
o ftp2.usa.openbsd.org  NYC, USA       o ftp3.usa.openbsd.org  CO, USA
o ftp5.usa.openbsd.org  CA, USA       o rt.fm  IL, USA

For greater convenience, the new 4.2 release also comes with roughly ~200MB full install ISO images called "install42.iso". If you download this file, you do not need the other standard install files.

Highlights of OpenBSD 4.2 are listed below.

  • New/extended platforms:
    • OpenBSD/sparc64.
      The PCIe UltraSPARC IIIi machines like the V215 and V245 are now supported.
    • OpenBSD/hppa.
      Four-digit B/C/J-class workstations like the B2000, C3750 or J6750 are now supported (in 32-bit mode).
    • OpenBSD/alpha.
      Add support in the alpha platform for a couple of new Alpha models, AlphaServer 1200 and 4100.

  • Platforms skipped this release:
    • OpenBSD/sgi.
      This architecture will not be released this time.

  • Install/Upgrade process changes:
    • New install method!!
      For the most popular architectures, the FTP sites have a ~200MB install ISO file, which contains the base set, permitting non-network installs.
    • Allow the specification of an NTP server during installation.
    • Allow no fsck'ing of clean non-root partitions during upgrade.
    • Check for INSTALL. to confirm sets are for the correct architecture.
    • Create and format the MSDOS partition for macppc installs in a more flexible and reliable way.

  • Improved hardware support, including:
    • Native Serial-ATA support:
      • ahci(4) driver for SATA controllers conforming to the Advanced Host Controller Interface specification.
      • jmb(4) driver for the JMicron JMB36x SATA II and PATA Host Controller.
      • sili(4) driver for SATA controllers using the Silicon Image 3124/3132/3531 SATALink chipsets.
    • The pciide(4) driver has had support added for newer chipsets, including:
      • Intel ICH8M PATA
      • JMicron JMB36x PATA
      • VIA CX700/VX700 PATA
    • The lm(4) driver now supports Winbond W83627DHG and W83627E HF-A Super I/O Hardware Monitors.
    • The siop(4) driver now has support for NCR 53C720/770 controllers in big endian mode. In particular this means that the onboard Fast-Wide SCSI on many hppa machines is supported now.
    • New tht(4) driver for Tehuti Networks 10Gb Ethernet controllers.
    • The malo(4) driver now supports Marvell 88W8385 802.11g based Compact Flash devices.
    • New uts(4) driver for USB touch screens, supported by the xtsscale(1) calibration utility.
    • The i810(4) X.Org driver and the PCI AGP driver now support Intel i965GM chips.
    • New led(4) driver for the front panel LEDs on the V215/245.
    • New bbc(4) driver providing support for the BootBus Controllers in UltraSparc III systems.
    • New pmc(4) driver for the watchdog(4) timer on the National Semiconductor PC87317 SuperIO chip.
    • New pyro(4) driver for the SPARC64 Host/PCIe bridge.
    • New astro(4) driver for the Astro Memory and I/O controller on hppa.
    • New elroy(4) driver for the Elroy PCI hostbridge on hppa.
    • New lcd(4) driver for the front panel LCD display on hppa.
    • New ssio(4) driver for the National Semiconductor PC87560 Legacy IO on hppa.
    • New pxammc(4) driver for the MMC/SD/SDIO controller on zaurus.
    • New xlights(4) driver for the front panel lights on the Xserve G4.
    • New sysbutton(4) driver for the system identification button on the Xserve G4.
    • New piixpcib(4) driver for System Management Mode initiated speedstep frequency scaling on certain pairings of the Intel PIIX4 ISA bridges and Intel Pentium 3 processors.
    • CPU frequency and voltage can now be scaled on all CPUs when running GENERIC.MP on a multiprocessor i386 or AMD64 machine with enhanced speedstep or powernow.
    • Intel enhanced speedstep is now supported on OpenBSD/amd64.
    • New support for the on die CPU temperature sensor found on the Intel Core family of processors.
    • The nvram(4) driver is now available on OpenBSD/amd64.

  • New tools:
  • New functionality:
    • FFS2, the updated version of the fast file system.
    • ftp(1) now can send cookies loaded from a netscape-like cookiejar, supports proxies requiring a password, and has a keep-alive option to avoid over-aggressive control connection dropping.
    • pkg_add(1) has been vastly improved. It is more robust, outputs more consistent error messages, and can deal with a lot more update scenarios gracefully. It also has much better look-up capabilities for multiple entries in PKG_PATH, stopping at the first directory with suitable candidates.
    • ftp-proxy(8) is now able to automatically tag packets passing through the pf(4) rule with a supplied name.
    • Kernel work queues, workq_add_task(9), workq_create(9), workq_destroy(9) provides a mechanism to defer tasks to a process context when it is impossible to run such a task in the current context.
    • ifconfig(8) now understands IP address/mask in CIDR notation.
    • Add IP (v4 and v6) load balancing to carp(4), similar to the ARP balancing.
    • sensorsd(8) now supports a zero-configuration monitoring and has a more intuitive logging for all sensors that automatically provide sensor state. It also features advancements in user-specified monitoring, including monitoring of sensors that are periodically flagged as invalid.
    • sensorsd.conf(5) now supports sensor matching by sensor type, allowing easier configuration.
    • The i386 bootloader can now load amd64 kernels.
    • The amd64 bootloader can now load i386 kernels.

  • Assorted improvements and code cleanup:
    • Large (>1TB) disk and partition support in the disklabel and buffer cache code and in the userland utilities that manipulate disk blocks. Note that some parts of the system are not 64-bit disk block clean yet, so partitions larger than 2TB cannot be used at the moment.
    • Large (>2^32-1 sectors) SCSI disk sizes now probed.
    • Thread support for the Objective-C library (libobjc).
    • carp route handling has been fixed, solving some problems the routing daemons were exhibiting.
    • Various improvements in pf increase performance drastically, stateful passing more than twice as fast than before.
    • A change in how the kernel random pool is stirred increases performance with network interface cards that support interrupt mitigation a lot.
    • i386 TLB handling improved to avoid possible corruption on Core2Duo processors.
    • Rework TLB shootdown code for i386 and amd64, gives fairly large speed improvements.
    • i386 and amd64 use the BIOS disk geometry when creating default disklabels.
    • Default disk geometry changed from 64 heads/32 sectors to 255 heads/63 sectors.
    • More USB devices with various quirks coerced to work.
    • Archive/Wangtek cartridge tape drives (wt*) no longer supported.
    • rcs has improved GNU compatibility.
    • make(1) has improved stability of -j option.

  • OpenSSH 4.7:
    • Prevent ssh(1) from using a trusted X11 cookie if creation of an untrusted cookie fails.
    • sshd(8) in new installations defaults to SSH Protocol 2 only. Existing installations are unchanged.
    • The SSH channel window size has been increased, and both ssh(1) sshd(8) now send window updates more aggressively. These improves performance on high-BDP (Bandwidth Delay Product) networks.
    • ssh(1) and sshd(8) now preserve MAC contexts between packets, which saves 2 hash calls per packet and results in 12-16% speedup for arcfour256/hmac-md5.
    • A new MAC algorithm has been added, UMAC-64 (RFC4418) as "umac-64@openssh.com". UMAC-64 has been measured to be approximately 20% faster than HMAC-MD5.
    • A -K flag was added to ssh(1) to set GSSAPIAuthentication=Yes
    • Failure to establish a ssh(1) TunnelForward is now treated as a fatal error when the ExitOnForwardFailure option is set.
    • ssh(1) returns a sensible exit status if the control master goes away without passing the full exit status.

  • OpenBGPD 4.2:
    • Include support for Four-octet AS Number Space.
    • Allow matching on communities using 0 in the AS part.
    • Filtering on IPv6 prefixes is now possible.
    • Various bugs in the encoding of multiprotocol updates were fixed.
    • Allow the use of pkill -HUP bgpd to reload the config.
    • bgpctl can filter prefix output by community now.

  • OpenOSPFD 4.2:
    • Added support for RFC 3137: OSPF Stub Router Advertisement.
    • It is possible to specify a carp demote group on interfaces and areas.
    • Added support for mapping route labels to AS-external route tags and vice versa.
    • Allow the use of pkill -HUP ospfd to reload the config.

  • HostStated 4.2:
    • Added support for Layer 7 load balancing (or relay).
    • Added support for reloading through hoststatectl or SIGHUP.
    • Added support for testing name-based virtual hosts.

  • Over 4,500 ports, minor robustness improvements in package tools.
    • Many pre-built packages for each architecture:
      i386: 4360 sparc64: 4205 alpha: 3779 sh: 817
      amd64: 4283 powerpc: 4230 sparc: 3283 m68k: 1107
      arm: 2131 hppa: 3215     
    • Highlights include:
      • Gnome 2.18.
      • GNUstep 1.14.
      • KDE 3.5.7 and koffice 1.6.3.
      • Xfce 4.4.1.
      • OpenMotif 2.3.0.
      • OpenOffice.org 2.2.1.
      • Mozilla Firefox 2.0.0.6.
      • PostgreSQL 8.2.4.
      • GHC 6.6.1 (amd64 and i386 only)

  • As usual, steady improvements in manual pages and other documentation.
  • The system includes the following major components from outside suppliers:
    • Xenocara (based on X.Org 7.2 + patches, freetype 2.2.1, fontconfig 2.4.2, expat 2.0.0, Mesa 6.5.2, xterm 225 and more)
    • Gcc 2.95.3 (+ patches) and 3.3.5 (+ patches)
    • Perl 5.8.8 (+ patches)
    • Our improved and secured version of Apache 1.3, with SSL/TLS and DSO support
    • OpenSSL 0.9.7j (+ patches)
    • Groff 1.15
    • Sendmail 8.14.1, with libmilter
    • Bind 9.3.4 (+ patches)
    • Lynx 2.8.5rel.4 with HTTPS and IPv6 support (+ patches)
    • Sudo 1.6.9p4
    • Ncurses 5.2
    • Latest KAME IPv6
    • Heimdal 0.7.2 (+ patches)
    • Arla 0.35.7
    • Binutils 2.15 (+ patches)
    • Gdb 6.3 (+ patches)

If you'd like to see a list of what has changed between OpenBSD 4.1 and 4.2, look at http://www.OpenBSD.org/plus42.html.

Thank you to all of the developers who make OpenBSD possible. Please be sure to make a donation to continue to make OpenBSD releases possible.

(Comments are closed)


Comments
  1. By SleighBoy (24.113.29.23) on http://www.code.cx/

    Halloween, 4.2 Release....

    Pumpkin!

    http://www.flickr.com/photos/sleighboy/1810434544/

    Comments
    1. By Leonardo Rodrigues (189.10.126.156) on

      > Halloween, 4.2 Release....
      >
      > Pumpkin!
      >
      > http://www.flickr.com/photos/sleighboy/1810434544/

      That was really nicely done =)

  2. By Anonymous Coward (212.183.136.193) on

    YAY!

  3. By Bitje (194.151.182.60) lj@nix.bz on

    This baby will go on my brand new HP notebook :D

  4. By Anonymous Coward (74.13.33.101) on

    It lists Xenocara, but not the old Xfree86 that has been tagging along releases since some drivers for older hardware were dropped by developers, does that mean the support for that older hardware is gone or that it has been set up for Xenocara?

    Comments
    1. By Brad (2001:4830:122b:3:216:41ff:fe17:6933) brad at comstyle dot com on

      > It lists Xenocara, but not the old Xfree86 that has been tagging along releases since some drivers for older hardware were dropped by developers, does that mean the support for that older hardware is gone or that it has been set up for Xenocara?

      The Xfree86 3.3.x drivers were dropped. It was only for really old crufty hardware and AFAIK this stuff should work Ok even if it is using the generic VGA driver.

  5. By Jeroen Janssen (j.janssen) jeroen@opennsd.net on http://opennsd.net

    Did anyone else notice the MD5 sum is 03dc43a1d18d3003843a1f13b3861917? Downloaded from 4 mirrors, they all give this MD5 sum, and not the one listed in "MD5".

    Comments
    1. By Jeroen Janssen (j.janssen) on http://opennsd.net

      > Did anyone else notice the MD5 sum is 03dc43a1d18d3003843a1f13b3861917? Downloaded from 4 mirrors, they all give this MD5 sum, and not the one listed in "MD5".

      Sorry, forgot to mention this is about the i386 version.

    2. By Todd T. Fries (todd) on http://todd.fries.net/

      > Did anyone else notice the MD5 sum is
      > 03dc43a1d18d3003843a1f13b3861917?
      > Downloaded from 4 mirrors, they all give this MD5 sum,
      > and not the one listed in "MD5".

      I think you need to be more clear, or do better checking:

      $ grep 03dc43 MD5
      MD5 (install42.iso) = 03dc43a1d18d3003843a1f13b3861917
      $ md5 install42.iso
      MD5 (install42.iso) = 03dc43a1d18d3003843a1f13b3861917
      $

      This looks perfectly sane to me.

      Comments
      1. By Jeroen Janssen (J.Janssen) on http://opennsd.net

        > > Did anyone else notice the MD5 sum is
        > > 03dc43a1d18d3003843a1f13b3861917?
        > > Downloaded from 4 mirrors, they all give this MD5 sum,
        > > and not the one listed in "MD5".
        >
        > I think you need to be more clear, or do better checking:
        >
        > $ grep 03dc43 MD5
        > MD5 (install42.iso) = 03dc43a1d18d3003843a1f13b3861917
        > $ md5 install42.iso
        > MD5 (install42.iso) = 03dc43a1d18d3003843a1f13b3861917
        > $
        >
        > This looks perfectly sane to me.


        Yes, it does. For some reason the MD5 sum for install42.iso (i386) release on ftp,kd85.com gave another sum. Either I was stoned, or the rsync fucked up.

        Comments
        1. By Wim Vandeputte (wvdputte) wim@kd85.com on https://kd85.com/notforsale.html


          > Yes, it does. For some reason the MD5 sum for install42.iso (i386) release on ftp,kd85.com gave another sum. Either I was stoned, or the rsync fucked up.

          Most of the MD5 files were updated earlier today, I just synced them, they should all be OK now

          I got the files really early, without the update

          Comments
          1. By Jeroen Janssen (j.janssen) on http://opennsd.net

            >
            > > Yes, it does. For some reason the MD5 sum for install42.iso (i386) release on ftp,kd85.com gave another sum. Either I was stoned, or the rsync fucked up.
            >
            > Most of the MD5 files were updated earlier today, I just synced them, they should all be OK now
            >
            > I got the files really early, without the update

            Ah, ok, thanks. I was already thinking that the original ISO's from ftp.openbsd.org were damaged. But this explains a lot ;)

  6. By Kian (71.231.38.215) kian.mohageri@gmail.com on http://www.zampanosbits.com

    Thank you all for yet another amazing release.

  7. By and I have it along with a T-Shirt - yes ! Thanx all ! (129.35.231.2) peb.nowarra@bluewin.ch on

  8. By Dylan Cochran (68.33.25.161) on

    With regards to FFS2, does anyone know (or know where it's written down) what the differences are between it and UFS2? as in, is it a straight port or is it only implementing the features the developers like?

    Comments
    1. By Anonymous Coward (68.104.220.48) on

      > With regards to FFS2, does anyone know (or know where it's written down) what the differences are between it and UFS2? as in, is it a straight port or is it only implementing the features the developers like?

      Um, it's not quite what you're thinking.

    2. By ClickF1 (82.146.61.159) clickf1.kz@gmail.com on

      Ура товарищи !

    3. By tedu (38.99.3.113) on

      > With regards to FFS2, does anyone know (or know where it's written down) what the differences are between it and UFS2? as in, is it a straight port or is it only implementing the features the developers like?

      ffs2 is ufs2

  9. By Anonymous Coward (193.15.214.188) on

    Torrents of the release.

    Please seed forever if you want to support OpenBSD.

    Comments
    1. By ftp @ df.lth.se (194.47.245.200) on

      > Torrents of the release.
      >
      > Please seed forever if you want to support OpenBSD.
      >

      Or use the mirrors, we (ftp.df.lth.se) are hardly using any of our bandwidth :-)

Credits

Copyright © - Daniel Hartmeier. All rights reserved. Articles and comments are copyright their respective authors, submission implies license to publish on this web site. Contents of the archive prior to as well as images and HTML templates were copied from the fabulous original deadly.org with Jose's and Jim's kind permission. This journal runs as CGI with httpd(8) on OpenBSD, the source code is BSD licensed. undeadly \Un*dead"ly\, a. Not subject to death; immortal. [Obs.]