Stoyan Angelov has build up an impressive Solaris documentation site with links to Opensolaris, ZFS, SMF, name it, all the Solaris goodies are included.
One of the features of Solaris 10 which is no less important than DTrace, Zones, and SMF is the re-write and major speed-up of the network stack. Internally the project is called FireEngine, and BigAdmin carries an interesting round-up of this new stack.
OpenSolaris BrandZ is a framework that extends the Solaris Zones infrastructure to create Branded Zones, which are zones that contain non-native operating environments. Nils Nieuwejaar has a blog post where he installs a Debian zone with BrandZ.
Sun has offered various tools in the past to analyze Sun/Solaris systems for patches which are installed or missing, e.g. PatchDiag, PatchCheck, PatchPro, smpatch (see the Sun Patch Portal for details). Some of them are not actively maintained, some are huge and opaque, some don't run on older Solaris releases or stripped-down machines.
Patch Check Advanced (PCA) is a perl script which generates lists of installed and uninstalled patches for Sun/Solaris systems and optionally downloads and installs patches.
# w
2:51pm up 852 day(s), 18:35, 2 users, load average: 0.77, 0.61, 0.33
# init 0
I guess this is a record. I was afraid the server would reboot with fsck errors, but it came back online without a glitch. This is a Netra server running Solaris 7. My heart bleeds every time I must do this, but patching is a necessary thing...
Since mid november, Sun released Nevada build 27, which contained the source code of the anticipated ZFS file system. ZFS is IMO a radically new and revolutionary filesystem which completely eliminates the concept of volumes and the associated problems of partitions, all operations are copy-on-write transactions, so the on-disk state is always valid. There is no need to fsck a ZFS filesystem ever. Every block is checksummed to prevent silent data corruption, and the data is self-healing in replicated (mirrored or RAID) configurations, which is kinda neat.
So I decided to test drive the new x86 build; unfortunately, the Solaris installer is not for the weak at heart : the installer hardly doesn't upgrade, doesn't contain ZFS support, which is really sad, so your filesystems are created as UFS . I believe it is still impossible to put your root partition on ZFS, too, so I guess we're still stuck with UFS.
I installed the build in a VMware container, which makes the installer friggin' slow (it took over 7 hours to install), and I had to scrape my 256 MB RAM based notebook for all available memory : if I gave the VMware guest too much memory, it got terminated by the Linux OOM killer. Giving it 200 Meg RAM went fine, but then you're stuck with the textual console based installer. In short : use a machine with a minimum of 512 Meg if you're planning to install this in VMware. I'll be downloading the sparc build in the near future, to see how this behaves when installing it onto my Enterprise 3000 server.
Luckily, you don't need a JBOD or a million dollar RAID5 storage system to play around with ZFS : ZFS has the ability to use files as virtual devices! Instead of using a real disk, you can instead create files of 128MB or larger and use them just like a disk. This allows for debugging, testing, and experimentation with complex pool setups without having to require immense resources. Obviously this is gonna be slow. You've got ZFS on top of UFS, so, don't expect it to be speedy. But the point here isn't performance, its about being able to experiment, play, and learn with ZFS configurations that otherwise be impracticle if not impossible. As an example :
root@harad ~$ mkdir /vdev
root@harad ~$ mkfile 128m /vdev/disk1
root@harad ~$ mkfile 128m /vdev/disk2
root@harad ~$ mkfile 128m /vdev/disk3
root@harad ~$ zpool status
no pools available
root@harad ~$ zpool create oasis raidz /vdev/disk1
/vdev/disk2 /vdev/disk3
root@harad ~$ zpool status
pool: oasis
state: ONLINE
scrub: none requested
config:
NAME STATE READ WRITE CKSUM
oasis ONLINE 0 0 0
raidz ONLINE 0 0 0
/vdev/disk1 ONLINE 0 0 0
/vdev/disk2 ONLINE 0 0 0
/vdev/disk3 ONLINE 0 0 0
That the builds are *test* builds I had to discover unfortunately : I bumped into this bug which refuses to boot the kernel :
Reading beyond end of ramdisk start=0x2000 size=0x2000 failed to read superblock panic : can't mount boot archive
Some info about the upcoming Sun Niagara chips has been leaked; The Register got some output of an OpenSolaris box running on a 8-way Niagara chip, showing 32 threads at a time, which makes Niagara very interesting for webbased stuff. I definitely would like to see some Apache2 benchmarks on this stuff, and can't wait to have one of these in our computer room. The first server based on the Niagara chip is believed to be called the Sun Fire T200.
Ben Rockwood has some more info about the different Sun chipsets and their codenames. Interesting to see is that a Niagara chip is in fact a scaled down UltraSparc II.
I wanted to play around with smf (System Management Facility) in Solaris 10 on my laptop, when I noticed it wasn't present. Apparently, the Solaris Express build on my laptop was b63, which is quite outdated. Time for an upgrade, then. Sun hasn't simplified things by making three versions of Solaris available for download :
OpenSolaris is, off course, the most interesting one to follow. There is quite a bit of new stuff in the latest build, called Nevada :
OpenSolaris now contains BFU, an utility to upgrade the system without the need to download a whole ISO image (yay!). Off course, you can use Flash Archives to backup your previous system.
I find this very impressive :
# top
load averages: 501.82, 436.45, 366.07
621 processes: 524 sleeping, 75 running, 17 zombie, 5 on cpu
CPU states: 0.0% idle, 64.9% user, 35.1% kernel, 0.0% iowait, 0.0% swap
Memory: 4096M real, 142M free, 2349M swap in use, 2790M swap free
...
The box is a 4 CPU Sun Enterprise 10000 domain, running Solaris7. The server still feels very responsive, too.
C|net's news.com.com has reported that Sun Microsystems is releasing parts of the OpenSolaris code today licensed under the OSI-approved CDDL . The release consistes of over 5 million lines of code for the base system OS/Net (kernel and networking). OpenSolaris is based on Solaris 10, the current version of Sun's Unix Operating System. Back in January, Sun released the code for DTrace, a dynamic tracing tool for analyzing and debugging kernel and userland events. DTrace is one of the big features in Solaris 10. Some other highlights include the GRUB bootloader, SMF (Service Management Facility) which replaces init.d scripts, it starts up processes in parallel for faster boots as well as providing features for automatically restarting.
OpenSolaris provides support for x86/x86-64 processors as well as Sparc. The Blastware guys are working on Polaris, which is an OpenSolaris port to PowerPC. Sun has been working on opening Solaris for over a year now. The OpenSolaris project started with a pilot group of Sun and non-Sun users. During the pilot program a lot of info including screenshots could be found on various OpenSolaris member blogs. (My favorite is Ben Rockwood's blog).
Teamware is the source code management system Sun uses for Solaris and OpenSolaris. Which was designed by Larry McVoy (now of BitKeeper) while he was at sun. No word yet on if Teamware will be available for OpenSolaris developers or not. Sun also uses CollabNet for it's Open Source project websites so that might be a possibility as well.
Solaris 10 has been out for a while, and I even noticed it comes already preinstalled on our new V440 Sun servers. Sure, everyone now know about zones, ZFS and DTrace, but since there are 600 new features added to Solaris 10, there's still lots to discover. OSNews.com finally brings a Solaris sysadmin which talks about the forgotten new stuff in Solaris 10, which isn't sexy enough to make the marketing flyers, but still delivers a great advantage compared to Solaris8 or 9.