rlr@toccata.rutgers.edu (Rich Rosen) (12/04/89)
Having just gotten back on the air (sort of) with my UNIX-PC, let me first thank all the people who responded to my questions back towards the beginning of this year, when I had a friend post some questions for me. (Of course, I forgot to thankthose who sent answers, so apologies and thanks, belatedly.) Next, now that I've finally got some things that I both want to and *can* do with this box, I have a few questions. 1. When I'm using vi's map function, either in my .exrc or interactively from within vi, I find that mapped keys that are control characters do not operate at all. On my old base machine at work I had set up some mapped macros for doing a lot of silly things, many of which were assigned to ^R or ^O or ^Z^Z, for example. NONE of them currently work. I checked to make sure they are defined correctly and they are, apparently. (In other words, I didn't do an ASCII file transfer from my old machine that literally wrote carot-R instead of a cntl-R, there's really a cntl-R there.) Is this a known bug or am I doing something wrong somewhere? 2. Recently I started playing with /etc/rc to shorten the bringup time of the box. I didn't want to comment out the fsck entirely, I just wanted it to be skipped over if the machine had previously shut down cleanly. I saw a trick that I think accomplishes this in a SUN /etc/rc. It looks for a file called /fastboot, and if it's there, /etc/rc skips over the fsck, and then rm -f's /fastboot. The shutdown procedures (/etc/shutdown and an /etc/qshut shell I hacked up) re-create the file /fastboot for next time. If perchance the machine craps out, /fastboot is not recreated, and the full fsck of /etc/rc takes place. My question: is this a reasonable and safe practice on my part? Is there a better way, and are there other ways to quicken the startup process that are safe and reasonable? 3. Not having a lot of practical hands-on experience OPERATING a UNIX box, I needed a lot of help in understanding what UUCP does before I was willing to start even attempting to play with it. I got that help from the Nutshell handbook "Managing UUCP and Usenet". From that book I also learned of seemingly obvious (to most) things, like how getty works. The book describes the methods used for SysV type systems without uugetty to manage the use of one port for incoming and outgoing, most of which sound like kludges. Could someone explain what exactly the UNIX-PC does to accomplish this? (I told you they were silly questions! :-) 4. I need (once again, since I lost the info) the info on how to contact "THE STORE", and also how to get a hold of HDB/BNU UUCP. 5. Finally, since I'm not UUCP connected to anybody yet, I need some means of uploading and downloading files. This machine I'm logged into now does not have UMODEM, seemingly the only supported way of getting/transmitting files from/to my machine. Until I get better services, I need at least a PD version of UMODEM for this Sun box, in order to first GET that other better software down to my machine. Anyone know where I can get such a thing, or does anyone have a simpler way? 6. This time REALLY finally: I've read about this new 3.51d release. I still have the base 3.5 installed. Woulod I be able to upgrade to 3.51d directly? Thanks in advance to any and all who respond. -- "Time to eat all your words, swallow your pride, open your eyes..." Rich Rosen rlr@toccata.rutgers.edu --
jcm@mtunb.ATT.COM (John McMillan) (12/05/89)
In article <Dec.3.14.05.03.1989.9674@toccata.rutgers.edu> rlr@toccata.rutgers.edu (Rich Rosen) writes: : >6. This time REALLY finally: I've read about this new 3.51d release. I still >have the base 3.5 installed. Woulod I be able to upgrade to 3.51d >directly? No: it is only for upgrading from a 3.51 base. john mcmillan -- att!mtunb!jcm
lenny@icus.islp.ny.us (Lenny Tropiano) (12/06/89)
In article <Dec.3.14.05.03.1989.9674@toccata.rutgers.edu> rlr@toccata.rutgers.edu (Rich Rosen) writes: [...] |>2. Recently I started playing with /etc/rc to shorten the bringup time of the |>box. I didn't want to comment out the fsck entirely, I just wanted it to be |>skipped over if the machine had previously shut down cleanly. I saw a trick |>that I think accomplishes this in a SUN /etc/rc. It looks for a file called |>/fastboot, and if it's there, /etc/rc skips over the fsck, and then rm -f's |>/fastboot. The shutdown procedures (/etc/shutdown and an /etc/qshut shell |>I hacked up) re-create the file /fastboot for next time. If perchance the |>machine craps out, /fastboot is not recreated, and the full fsck of /etc/rc |>takes place. My question: is this a reasonable and safe practice on my |>part? Is there a better way, and are there other ways to quicken the startup |>process that are safe and reasonable? |> [...] This is probably a safe way of doing things. I wrote a small set of boot scripts and a program called "fsokay" which will only fsck the filesystems if it's considered "dirty" (or not OK). This is done by seeing what is in /etc/mnttab at boot-time, and fsck those filesystems. At shutdown time the /etc/mnttab is cleared through a series of "umount" calls (for filesystems other than "/" root) and a "> /etc/mnttab". If you are interested in seeing *my* solution to this, grab the /etc/fsokay package from the ICUS Archives (see the posting about the archives that was posted earlier this month...). fsokay.sh.Z CS Improved UNIX-pc /etc/rc script and a -Lenny -- | Lenny Tropiano ICUS Software Systems [w] +1 (516) 589-7930 | | lenny@icus.islp.ny.us Telex; 154232428 ICUS [h] +1 (516) 968-8576 | | {ames,pacbell,decuac,hombre,sbcs,attctc}!icus!lenny attmail!icus!lenny | +------- ICUS Software Systems -- PO Box 1; Islip Terrace, NY 11752 -------+