Subject: pmr 8x333,b344 (fwd) OH WELL From: James A. Sledd Thought you might be interested in this one. Unfortunately it begins by saying, turn the key to the service position and boot with tape or diskettes. Forwarded message: > From root Mon Nov 29 15:45:41 1993 > Posted-Date: Mon, 29 Nov 1993 14:39:28 -0600 > Received-Date: Mon, 29 Nov 1993 15:45:37 -0500 > Date: Mon, 29 Nov 1993 14:39:28 -0600 > From: sladewsk@risc.austin.ibm.com (Mary Sladewski) > Message-Id: <9311292039.AA08444@risc.austin.ibm.com> > To: jsledd@mcneil.sas.upenn.edu > Subject: pmr 8x333,b344 > > > > 09/24/93 > > RECOVERING FROM A CORRUPTED SUPERBLOCK IN AIX 3.1 OR 3.2 > > > > SPECIAL NOTICES > > The problem for which you received this document is not con- > sidered a code warranty issue. This document is provided > as an aid by the Austin AIX Support Center. If you need > further assistance, contact your local branch office or > point of sale, or call 1-800-CALL-AIX for information about > support offerings. All of the above services may be > billable. Faxes on a variety of subjects may be ordered > free of charge from 1-800-IBM-4FAX. > > Comments about this document may be sent by fax to "Info > Feedback" at (512) 823-7634. IBM representatives can send > comments internally to ROUSHC at AUSVM8. > > The information contained in this document is distributed > "AS IS" without any warranties of any kind either expressed > or implied. IBM will not be responsible for any direct, > incidental, consequential, special or indirect damages. IBM > EXPRESSLY DISCLAIMS ANY IMPLIED WARRANTY OF MERCHANTABILITY > AND ANY IMPLIED WARRANTY OF FITNESS FOR A PARTICULAR > PURPOSE. > > The use of this information or the implementation of any of > these techniques is the sole responsibility of the customer > and depends on the customer's ability to evaluate and inte- > grate this information or implementation into the customer's > operational environment. > > PROCEDURE > > | If you receive one of the following errors from fsck, the > | problem may be a corrupted superblock. > > "fsck: Not an AIX3 file system." > > "fsck: Not an AIXV3 file system." > > "fsck: Not a recognized file system type." > > | Follow these steps to recover a corrupted superblock: > > 1. Turn the key to the Service position. > > 2. With bootable media OF THE SAME VERSION AND LEVEL AS THE > SYSTEM, boot the system. (If booting from diskettes, > insert the Display diskette when you see LED c07.) > > > > > > > > > > > Recovering from a Corrupted Superblock in AIX 3.1 or 3.2 1 > > > 09/24/93 > > ------------------------------------------------------------ > > | WARNING: If you boot a 3.2 system with 3.1 media, or boot a > | 3.1 system with 3.2 media, then you will not be able to use > | the standard scripts (getrootfs or /etc/continue) to bring > | your workstation into full maintenance mode. > > | Moreover, performing the scripts on a 3.1 system with 3.2 > | boot media may actually remove some files and prevent your > | system from booting successfully in normal mode until > | missing files (/etc/mount and /etc/umount) are replaced on > | the disk. > > ------------------------------------------------------------ > > Follow the prompts to the installation/maintenance menu. > > 3. Choose the maintenance shell (option 5 for AIX 3.1, > option 4 for AIX 3.2). > > 4. Determine the hdisk# to use with the getrootfs or > /etc/continue command. If you have only one disk, then > "hdisk0" is the proper hdisk# to use. If you have more > than one disk, run > > lqueryvg -Atp hdisk# | grep hd5 > > for each hdisk# (hdisk0, hdisk1, etc.) until you get > output that looks like: > > 00005264feb3631c.2 hd5 1 > > The exact output you get will be different but will > follow the form > > large_number.x hd5 1 > > You may find more than one disk has this output. These > will all be disks which belong to the rootvg volume > group. You may use any of the disks identified to be in > rootvg in the following step. > > 5. Now access the rootvg volume group by running > /etc/continue (for AIX 3.1) or getrootfs (for AIX 3.2). > ('#' is the number of the fixed disk, determined in step > 4). > > For AIX 3.1 only: /etc/continue hdisk# sh > For AIX 3.2 only: getrootfs hdisk# sh > > If you get errors from /etc/continue or getrootfs, do > not continue with the rest of this procedure. Correct > the problem causing the error. > > 6. Try rebuilding the log device: > > For AIX 3.1 only: /etc/aix/logform /dev/hd8 > For AIX 3.2 only: logform /dev/hd8 > > Answer YES when asked if you want to destroy the log. > > Recovering from a Corrupted Superblock in AIX 3.1 or 3.2 2 > > > 09/24/93 > > 7. Next, run the following commands to check and repair > file systems. (The "-y" option gives fsck permission to > repair file systems when necessary.) > > fsck -y /dev/hd1 > fsck -y /dev/hd2 > fsck -y /dev/hd3 > fsck -y /dev/hd4 > > For AIX 3.2 only, also run > > fsck -y /dev/hd9var > > 8. Type "exit" to exit from the shell. The file systems > will automatically mount after you type "exit". > > | If you get the same fsck errors now during the automatic > | mounting as you did before starting this procedure, skip > | to step 11 now. > > 9. With the key in Normal position, enter the following: > > sync;reboot > > | If you still get the errors from fsck during the reboot, > | follow steps 1 through 4 again and then skip to step 10. > > 10. Now access the rootvg volume group (hd5) by running > /etc/continue (for AIX 3.1) or getrootfs (for AIX 3.2). > ('#' is the number of the fixed disk, determined in step > 4.) > > For AIX 3.1 only: /etc/continue hdisk# > For AIX 3.2 only: getrootfs hdisk# > > 11. Check the entry in block 1000 of the file system in > question (/dev/hd4 is used as an example): > > For AIX 3.1 only: /usr/lpp/bosinst/hdf /dev/hd4 1000 100 > For AIX 3.2 only: /usr/sbin/hdf /dev/hd4 1000 100 > > The output should look something like this: > > 00001000 43218765 00000000 00000800 0000001F |C!.e............| > 00001010 0001A000 10000000 2F000000 0000726F |......../.....ro| > 00001020 6F740000 000A0004 01000000 294CD89F |ot..........)L..| > 00001030 00000000 00000000 00000000 00000000 |................| > 00001040 00000000 00000000 00000000 00000000 |................| > > > You will notice the "43218765" in the second column of > the first line. This is the magic number. You'll also > see in the strings "/" and "root"; that is, the mount > point and "vol =" entry for the file system, respec- > tively. > > If the entry at address 1000 is something other than > 43218765, fsck will give an error indicating that it is > not an AIXV3 file system. > > > Recovering from a Corrupted Superblock in AIX 3.1 or 3.2 3 > > > 09/24/93 > > 12. Check the entry in block 1F000 of the file system in > question (/dev/hd4 is used as an example): > > For AIX 3.1 only: /usr/lpp/bosinst/hdf /dev/hd4 1F000 100 > For AIX 3.2 only: /usr/sbin/hdf /dev/hd4 1F000 100 > > The output should look something like this: > > 0001F000 43218765 00000000 00000800 00000000 |C!.e............| > 0001F010 0001A000 10000000 2F000000 0000726F |......../.....ro| > 0001F020 6F740000 FFFFFFFF 00000000 294CD89F |ot..........)L..| > 0001F030 00000000 00000000 00000000 00000000 |................| > 0001F040 00000000 00000000 00000000 00000000 |................| > > > If you see "43218765" in the second column of the first > line, then you can issue the following command to copy > this BACKUP superblock to the CURRENT superblock: > > dd count=1 bs=4k skip=31 seek=1 if=/dev/hd4 of=/dev/hd4 > fsck /dev/hd4 2>&1 | tee /tmp/fsck.errors > > You may look at the file /tmp/fsck.errors to see what > corrections fsck states need to be made. In most cases, > by copying the BACKUP superblock entry to the CURRENT > superblock entry, you will be able to recover your file > system. If this does not work, you will have to re- > create the file system, and restore the data from a > backup. > > END OF DOCUMENT Subject: aix-archive Date: Wed, 13 Oct 1993 18:37:50 -0400 From: awhite (Andrew White) Support the AIX repository at WUARCHIVE !!! Washington University in St.Louis makes available via anonymous FTP a veritable cornicopia of information and software. Anyone who has need or opportunity to access this archive can atest to its usefullness. I am one of the disciples of this archive and am soliciting AIX specific contributions from anyone who has a useful contribution which they wish to make generally available. Contributions must be truly public domain. You may make direct deposit to wuarchive:/archive/systems/aix/contribute or may mail your contributions to me at rcf@wuarchive.wustl.edu. Thanks for your support. Subject: IBM to support WABI on RS/6000 Date: Tue, 12 Oct 93 10:24:11 EDT From: Andrew White IBM has announced AIX 3.2.5, and with that product announcement, also provided this "Statement of Intent". I assume that means it is not yet available. The following was taking from Article 31751 on comp.unix.aix, and is IBM announcement 293-488. ------ STATEMENTS OF GENERAL DIRECTION IBM announces the following statements of intent: o To provide the ability to run Microsoft (2) Windows applications (using Wabi (3)) on AIX o The ability to run Apple Macintosh (4) applications on the AIX/6000 operating system o A new AIX/6000 desktop. IBM intends to provide customers the ability to execute Microsoft Windows applications (using Wabi) on AIX/6000. IBM also intends to provide customers the capability to execute Apple Macintosh applications on AIX/6000. These two application environments are part of IBM's continuing commitment to provide customers flexibility in their choice of applications and to help protect customers' current application investments. The new application environments, and additions to current offerings, provide customers the ability to run DOS applications and existing AIX applications based on the AIX/6000 operating system. IBM intends to integrate these application environments into the AIX/6000 desktop. IBM also intends to provide a new graphic desktop on the AIX/6000 operating system. This desktop is intended to be the next generation graphic user environment for the AIX/6000 operating system. This desktop offering will be based on the Common Desktop Environment being developed within the common open software environment process. These statements represent IBM's current intent and are subject to change or withdrawal. IBM is strongly committed to AIX leadership in support of industry standards and open system specifications. IBM intends to broaden its capabilities by formally submitting a future version of AIX for the RISC System/6000 for X/Open (5) Portability Guide Issue 4 (XPG4) branding in 1994. XPG4 is the fourth release of X/Open's specifications that define an open systems environment. X/Open is an international, non-profit corporation focused on unifying standards issues in open systems environments. The X/Open portability guides assist software and system developers by providing them with the information on the functions and interfaces users have identified as requirements for open systems. These statements represent IBM's current intent and are subject to change or withdrawal. (2) Registered trademark of Microsoft Corporation. (3) Trademark of Sun Microsystems, Inc. (4) Trademark of Apple Computer, Inc. (5) Trademark of X/Open Company Ltd. in the UK and other countries. Subject: Ethernet performance Date: Mon, 16 Aug 93 9:41:56 EDT From: Andrew White Jase, Tip from the net: In SMIT: Devices Communication Ethernet Adapter Adapter Change / Show Characteristics of an Ethernet Adapter ent0 [select] I have changed: Receive Data transfer OFFSET to 26 (from 92) Transmit Queue Size to 80 Receive Queue Size to 150 Apply changes to database only: yes Reboot is needed to to take effect (these can't be changed while tcp/ip is running). This is supposed to increase NFS performance significantly. Subject: Running jobs on our RS/6000's Date: Wed, 22 Sep 93 11:07:19 EDT From: Andrew White I justed wanted to bring to the attention of our consultants something which may have been missed when reading the Policy statement. Users (including us) are not supposed to run SIMULTANEOUS large jobs (eg, sas, stata, gauss, etc). That means that you should NOT do the following: % sas program1 & % sas program2 & % sas program3 & The above will run THREE SIMULTANEOUS SAS jobs which is a very bad thing because it makes the system run very sloooowly. Instead, run them SEQUENTIALLY by entering: % sas program1; sas program2; sas program3 & This will run program1, and then run program2 when program1 is finished, and then program3. This is mentioned in the policy statement; you may wish to review it. Subject: at commmand Date: Thu, 2 Dec 93 17:41:02 EST From: Mike Little > at command at 8:30 pm --> press Enter sas myprog then press Ctrl-d Your SAS program will be scheduled to be run at 8:30 pm. Look at the man pages for at for more info. Subject: mounting home directories Date: Mon, 8 Nov 93 18:54:51 EST From: Mike Little > I showed a user how to mount his Unix home directory as a DOS directory from a gdal network PC. Say, for user little, you enter: idmnt unix ssdc /mcneil/home/userid f: user little The word "unix" is arbitrary; it can be anything. You are then prompted for your Unix password. You can now access drive F: as a DOS drive, but you are storing your files on SSDC. *** BE SURE TO UNMOUNT YOUR HOME DIRECTORY WHEN YOU'RE DONE!!! *** d: change to another drive idumnt f: unmount the filesystem Logging out does NOT unmount the filesystem!!! Subject: "smart recall" for set -o emacs Date: Tue, 12 Oct 93 20:07:25 EDT From: Mike Little The "set -o emacs" function allows you to do an emacs-style search for previously entered commands. Say you had entered some really long Unix command about 30 commands ago, and you want to enter it again. One way to redisplay the command would be to enter Ctrl-p 30 times, but this can be tedious. Just do a reverse search, like you would in emacs! First press Ctrl-R. "^R" will appear at the prompt. Follow this immeadiately with some unique characters from the command string, then press enter. The first command containing this string will appear at the prompt! (I tried this out after finding that Matlab has a similar "smart recall" function, except it uses the up-arrow to go backwards.) Subject: real arrow keys for Unix command line Date: Thu, 21 Oct 93 10:00:10 EDT From: Mike Little I found, in AIX Frequently Asked Questions 1.401, a way to get arrow keys to work on Unix command lines. I tried it and it works. Can it be set up to automatically execute w/o people having to change their .profile file? - Mike p.s. The "^P" et al. are real control sequences. I did them in gemacs. # Set up *real* arrow keys for command line editing. alias __A= alias __B= alias __C= alias __D= alias __H= set -o emacs # Allow emacs line editing keys on command line.