Installing OpenBSD is a relatively complex process, but if you have this document in hand and are careful to read and remember the information which is presented to you by the install program, it shouldn't be too much trouble. Before you begin, you should know the geometry of your hard disk, i.e. the sector size (note that sector sizes other than 512 bytes are not currently supported), the number of sectors per track, the number of tracks per cylinder (also known as the number of heads), and the number of cylinders on the disk. The OpenBSD kernel will provide limited information for SCSI disks, but not much. You should now be ready to install OpenBSD. The following is a walk-through of the steps you will take while getting OpenBSD installed on your hard disk. If any question has a default answer, it will be displayed in brackets ("[]") after the question. If you wish to stop the installation, you may hit Control-C at any time, but if you do, you'll have to begin the installation process again from scratch. Boot your machine from the CDROM or from the network. When presented with the boot prompt, enter the path of your kernel. (`bsd.rd' is a likely choice if you are booting over the network, otherwise check the CDROM insert for the pathname of the kernel.) You will then be presented with the OpenBSD kernel boot messages. You will want to read them, to determine your disk's name and geometry. Its name will be something like "sd0" or "hd0" and any geometry or size information will be printed on a line that begins with its name. As mentioned above, you will need your disk's geometry when creating OpenBSD's partitions. You will also need to know the name, to tell the install tools what disk to install on. If you cannot read the messages as they scroll by, do not worry -- you can get at this information later inside the install program. While booting, you will probably see several warnings. You should be warned that no swap space is present, and that init(8) cannot find /etc/rc. Do not be alarmed, these are completely normal. When you reach the prompt asking you for a shell name, just hit return. You will be presented with a welcome message and a prompt. At this time you should enter the command "install" to start the installation process. You will be asked which terminal type to use. There are only a few specific terminal types supported. If you are using a graphics display, just hit return to select the default (hp300h). If you are using an ASCII terminal, it should either be an HP or vt100 compatible terminal; nothing else is supported. The install program will then tell you which disks of that type it can install on, and ask you which it should use. The name of the disk is typically "hd0" for HPIB/CS80 drives or "sd0" for SCSI drives. If you do not wish to install on the first disk, you should use the appropriate number in place of "0". Reply with the name of your disk. Next you will have to edit or create a disklabel for the disk OpenBSD is being installed on. The installation script will invoke an editor allowing you to do this. Note that partition 'c' inside this disk label should ALWAYS reflect the entire disk, including any non-OpenBSD portions. The root file system should be in partition 'a', and swap is usually in partition 'b'. It is recommended that you create separate partitions for /usr and /var, and if you have room for it, also for /home. The swap partition (usually 'b') should have a type of "swap", all other native OpenBSD partitions should have a type of "4.2BSD". Block and fragment sizes are usually 8192 and 1024 bytes, but can also be 4096 and 512 or even 16384 and 2048 bytes. The install program will now label your disk and ask which file systems should be created on which partitions. It will auto- matically select the 'a' partition to be the root file system. Next it will ask for which disk and partition you want a file system created on. This will be the same as the disk name (eg. "hd0") with the letter identifying the partition (eg. "d") appended (eg. "hd0d"). Then it will ask where this partition is to be mounted, eg. /usr. This process will be repeated until you just hit return. At this point you will be asked to confirm that the file system information you have entered is correct, and given an opportunity to change the file system table. Next it will create the new file systems as specified, OVERWRITING ANY EXISTING DATA. This is the point of no return. After all your file systems have been created, the install program will give you an opportunity to configure the network. The network configuration you enter (if any) can then be used to do the install from another system using NFS, HTTP or FTP, and will also be the configuration used by the system after the installation is complete. If you select to configure the network, the install program will ask you for a name of your system and the DNS domain name to use. Note that the host name should be without the domain part, and that the domain name should NOT include the host name part. Next the system will give you a list of network interfaces you can configure. For each network interface you select to configure, it will ask for the IP address to use, the symbolic host name to use, the netmask to use and any interface-specific flags to set. No interface-specific flags should be required. After all network interfaces has been configured the install pro- gram will ask for a default route and IP address of the primary name server to use. You will also be presented with an opportunity to edit the host table. At this point you will be allowed to edit the file system table that will be used for the remainder of the installation and that will be used by the finished system, following which the new file systems will be mounted to complete the installation. After these preparatory steps has been completed, you will be able to extract the distribution sets onto your system. There are several install methods supported; FTP, HTTP, tape, CD-ROM, NFS or a local disk partition. To install from a tape, the distrib- ution sets must have been written to tape prior to running the installation program, either as tar images or as gzipped tar images. To install via FTP: To begin an FTP install you will need the following pieces of information. Don't be daunted by this list; the defaults are sufficient for most people. 1) Proxy server URL if you are using a URL-based ftp proxy (squid, CERN ftp, Apache 1.2 or higher). You need to define a proxy if you are behind a firewall that blocks outgoing ftp (assuming you have a proxy available to use). 2) Do you need to use passive mode ftp? Most modern ftp servers are capable of dealing with passive ftp connections. You only need to enable this option if you are behind a firewall that allows outgoing ftp but blocks incoming tcp ports > 1023. If in doubt say yes to this option. Note that you will not be asked about passive ftp if you are using a proxy. 3) The IP address (or hostname if you enabled DNS earlier in the install) of an ftp server carrying the OpenBSD 2.3 distribution. If you don't know, just hit return when asked if you want to see a list of such hosts. 4) The ftp directory holding the distribution sets. The default value of pub/OpenBSD/2.3/hp300 is almost always correct. 5) The login and password for the ftp account. The default will be correct unless you are doing non-anonymous ftp. For instructions on how to complete the installation via ftp, see the section named "Common URL installations" below. To install via HTTP: To begin an HTTP install you will need the following pieces of information: 1) Proxy server URL if you are using a URL-based http proxy (squid, CERN ftp, Apache 1.2 or higher). You need to define a proxy if you are behind a firewall that blocks outgoing http connections (assuming you have a proxy available to use). 3) The IP address (or hostname if you enabled DNS earlier in the install) of an http server carrying the OpenBSD 2.3 distribution. If you don't know, just hit return when asked if you want to see a list of such hosts. 4) The directory holding the distribution sets. There is no standard location for this; You should use the directory specified along with the server in the list of official http mirror sites that you received in step 3. For instructions on how to complete the installation via http, see the section named "Common URL installations" below. To install from tape: In order to install from tape, the distribution sets to be installed must have been written to tape previously, either in tar format or gzip-compressed tar format. You will also have to identify the tape device where the distribution sets are to be extracted from. This will typically be "nrst0" (no-rewind, raw interface). Next you will have to provide the file number of the set that is to be extracted. Note that the file number starts at 1, which is the first file written to the tape. The install program will not automatically detect whether an image has been compressed, so it will ask for that information before starting the extraction. To install from CD-ROM: When installing from a CD-ROM, you will be asked which device holds the distribution sets. This will be "cd0" for the first CD-ROM drive. Next you will be asked which partition on the CD-ROM the distribution is to be loaded from. This is normally partition "a". Next you will have to identify the file system type that has been used to create the distribution on the CD-ROM, this can be either FFS or ISO CD9660. The OpenBSD CD distribution uses the CD9660 format. You will also have to provide the relative path to the directory on the CD which holds the distribution, for the hp300 this is "2.3/hp300". For instructions on how to complete the installation from the CD-ROM distribution, see the section named "Common file system installations" below. To install from a NFS mounted directory: When installing from a NFS-mounted directory, you must have completed network configuration above, and also set up the exported file system on the NFS server in advance. First you must identify the IP address of the NFS server to load the distribution from, and the file system the server expects you to mount. The install program will also ask whether or not TCP should be used for transport (the default is UDP). Note that TCP only works with newer NFS servers. You will also have to provide the relative path to the directory on the file system where the distribution sets are located. Note that this path should not be prefixed with a '/'. For instructions on how to complete the installation from the CD-ROM distribution, see the section named "Common file system installations" below. To install from a local disk partition: When installing from a local disk partition, you will first have to identify which disk holds the distribution sets. This is normally "hdN" or "sdN" where N is a number 0 through 9. Next you will have to identify the partition within that disk that holds the distribution, this is a single letter between 'a' and 'p'. You will also have to identify the type of file system residing in the partition identified. Currently you can install from partitions that has been formatted as fast file system (ffs) or MS-DOS. You will also have to provide the relative path to the directory on the file system where the distribution sets are located. Note that this path should not be prefixed with a '/'. For instructions on how to complete the installation from the a local disk partition, see the next section. Common file system installations: The following instructions are common to installations from local disk partitions, NFS mounted directories and CD-ROMs. A list of available distribution sets will be listed. If any sets has already been extracted, those will be marked with an X. Enter the name of one distribution set at a time, until all desired distribution sets has been installed on your system. Common URL installations: Once you have entered the required information, the install program will fetch a file list and present a list of all the distribution sets that were found in the specified directory. (If no valid sets were found, you will be notified and given the option of unpacking any gzipped tar files found or getting a file list if none were found.) At this point you may individually select distribution sets to install or enter "all" to install all of the sets (which is what most users will want to do). You may also enter "list" to get a file list or "done" when you are done selecting distribution sets. (It is also possible to enter an arbitrary filename and have it treated as a file set). Once you have selected the file sets you want to install and entered "done" you will be prompted to verify that you really do want to download and install the files. Assuming you acquiesce, the files will begin to download and unpack. If not, you will be given the option of installing sets via one of the other install methods. When all the selected distribution sets has been extracted, you will be allowed to select which time zone your system will be using, all the device nodes needed by the installed system will be created for you and the file systems will be unmounted. For this to work properly, it is expected that you have installed at least the "base23" and "etc23" distribution sets. Congratulations, you have successfully installed OpenBSD 2.3. When you reboot into OpenBSD, you should log in as "root" at the login prompt. There is no initial password, but if you're using the machine in a networked environment, you should create yourself an account and protect it and the "root" account with good passwords. Some of the files in the OpenBSD 2.3 distribution might need to be tailored for your site. In particular, the /etc/sendmail.cf file will almost definitely need to be adjusted, and other files in /etc will probably need to be modified, as well. If you are unfamiliar with UN*X-like system administration, it's recommended that you buy a book that discusses it.