New or Revised TAs on websco, 960601

New or Revised TAs on websco, 960601

Post by RadleyRob » Mon, 03 Jun 1996 04:00:00



The following problems are addressed in Technical Articles that appeared on
SCO's websco.sco.com site on 960601.

Some of these TAs are new, others are revisions of prior versions.

PROBLEM:  How can I use my old SCO MPX serial number and activation key
          to license additional CPUs on my SCO OpenServer 5 system?  

PROBLEM:  What is the Large System Supplement 1.0.0 and how can I
          obtain a copy of the supplement?

PROBLEM:  PROBLEM 1
          SCOAdmin fails to start.  It does not give an error message or
          write information to any of the "standard" logs about why it
          failed to initialize.  This happens in both GUI and CHARM modes.

          PROBLEM 2
          Root cannot access the Filesystem Manager under SCOAdmin.  When
          the root account attempts to access the FileSystem Manager, this
          error message is seen:

                  Failed to retrieve local host device list
                  Cannot continue filesystem manager startup

          PROBLEM 3
          A user is administering system backups under a non-root account.
          When they use the backup manager and attempt to specify local
          devices for the backup, they expect to see a pick-and-choose
          dialog box with the available devices.  Instead, they see these
          error messages:

                  Failed to retrieve local host device list
                  Cannot continue backup manager startup

          This user account has been granted "backup" primary
          authorizations.

          Why is the SCOAdmin not returning information about my local
          storage devices?  Why is SCOAdmin failing to initialize?

PROBLEM:  1. The floppy is improperly formatted:

          This may occur on systems with more than 16MB of memory.
          The format command indicates that a floppy diskette was
          successfully formatted, but accessing the diskette afterward
          indicates that it is unformatted.

          2. Pkgadd from diskette with 1:1 interleaving is very poor:

          The poor performance is attributable to the fact that the
          floppy driver performs reads on a sector-by-sector basis.
          The pkgadd utility may hang when trying to install an application.

          3. Desktop hang fix during access of the floppy device:

          When the Desktop MediaMgr is executed to access the floppy
          device, the window hangs with the floppy access light on.

PROBLEM:  I have an application that has been compiled and run on earlier
          releases of SCO UNIX System V/386.  However, when I run this same
          application on SCO OpenServer Release 5.0.0, it experiences screen
          corruption.

          If I compile the same source code on SCO OpenServer 5.0.0, this
          application runs successfully.

          Alternatively, if I pipe the output of the original application
          through cat(C) or tee(C), this will also run successfully on
          SCO OpenServer 5.0.0.

PROBLEM:  I have increased the NSTRPAGES parameter to 4000 in my kernel
          using configure(ADM) but when I reboot using the new kernel, I
          get the error messages:

 WARNING: table_alloc - Failed to get address space for mnode table

 CONFIG: strinit - Cannot alloc STREAMS mnode table (NSTRPAGES = 4000 too big)

 WARNING: maddlink -Cannot grow STREAMS mux node table.

          Networking will not work using this kernel.

PROBLEM:  I need a command that I can run from the command line that
          returns the system ID of my system.

          I can get the system ID using the License Manager, but I am
          looking for a command that I do not have to run interactively.

PROBLEM:  When using lpd(ADMN) to export printers, the rlpstat command
          only lists requests outstanding from the system on which rlpstat
          is invoked.

          Assume a configuration similar to:

            ----------            ----------
            |        |            |        |
            | client |            | server |             |---------|
            |        |-----//-----|        |-------------| printer |
            |        |            |        |             |---------|
            |        |   TCP/IP   |        |
            ----------            ----------

          Where "client," "server" and "printer" are the names of the
          systems and printer and they also denote their respective
          functions.  The printer is named "printer" on both systems.

          Both "client" and "server" can access the "printer": "server"
          using a physical connection (a serial or parallel port); "client"
          using a TCP/IP connection to the lpd(ADMN) daemon on "server."

          If a request is submitted on "server":

             server> lp -dprinter /etc/motd
             request id is printer-3

          rlpstat on the server will list the request:

             server> rlpstat printer
             printer is ready and accepting request
             printer-3            root                2409

          However, rlpstat invoked on "client" will not list the request:

             client> rlpstat printer
             client>

          Conversely, if a request is submitted from "client":

             client> lp -dprinter /etc/inittab
             request id is printer-6

          The request will be shown by rlpstat on the client:

             client> rlpstat printer
             printer is ready and accepting request
             printer-6            root                2409

          It will only be shown by rlpstat invoked on the server, if it
          is the job at the head of the print queue:

             server> rlpstat printer
             server>

             server> rlpstat printer
             printer-9            root              2409   Mar 29 13:53
             server>

PROBLEM:  When using certain World Wide Web browsers such as Microsoft
          Internet Explorer (as provided in Windows95) or Netscape Navigator
          to access manual pages served by an SCO OpenServer 5.0.0 system,
          attempts to access the manual pages fail with the error:

            No support for accept type(s): image/gif, image/x-xbitmap,
            image/jpeg, */*

          For example, open a connection to an SCO OpenServer 5.0.0 system
          using a URL similar to:

            http://example:457/dochome.html

          where "example" is replaced with the name of the SCO OpenServer
          5.0.0 system. Then follow the hypertext links to a manual page.

          Alternatively, the following URL:

            http://example:457/cgi-bin/man/man?accept+ADM

          (replacing "example" with the name of an SCO OpenServer 5.0.0
          system) is a direct link to an appropriate manual page.

          With Microsoft Internet Explorer or Netscape Navigator, the browser
          displays an error as shown above.  With SCO Mosaic, or SCOhelp(XC),
          the manual page is returned correctly.

PROBLEM:  I am not seeing consistent behavior using the find(C) command.

          In the following example, I am trying to display all files with
          the string: "tune" in their names that are located within the
          subdirectory /etc/conf:

                    # find /etc/conf -name \*tune\* -print

          This command does not work; no files are returned.

          The find(C) commands:

                    # find /etc/conf/. -name \*tune\* -print
                    # find /etc/conf/.. -name \*tune\* -print

          do work; files are returned.

          The find command that does not work did work under releases previous
          to SCO OpenServer Release 5.0.

          Why does find no longer work as expected?        

PROBLEM:  I have installed an additional hard disk by running the "mkdev hd"
          utility, but the information entered for the controller and/or
          the disk number was incorrect.

PROBLEM:  Every 10-15 seconds, my /usr/adm/syslog file is updated with the
          error:

             telnetd: [pid] can't find utmp entry for ptty [xxx]

          where [pid] and tty number [xxx] will vary.

PROBLEM:  When attempting to determine the time the UNIX system booted, there
          is a discrepancy between the time stored in:

            /usr/adm/messages

          above the kernel drivers list, the time reported by the who(C)
          command:

            #  who -b

          and the time given in the 'Auto Boot Log' mail which root receives.

          Why does this occur ?

PROBLEM:  Is the Toshiba Satellite 100CS supported on SCO OpenServer Release 5?

PROBLEM:  When I run the cleantmp script it removes random files from any
          currently mounted filesystems.

PROBLEM:  I want to use the dir directive to boot(HW) to view the contents
          of a disk on my system. If I issue the command:

                dir hd(40)

          I get a listing of the /stand filesystem of my hard disk.  
          However, if I try to read the root filesystem on my hard disk
          using the command:

                dir hd(42)

          I get the error message:

                not a directory

                . not found
                Can't open hd(42)

PROBLEM:  My system has panics with the message:

                PANIC: finddbd: can't find page table entry

          What is causing this, and what actions should I take?

PROBLEM:  It is documented on page 205 of the SCO OpenServer Performance
          Guide that the maximum value for NSCRN (the number of Console
          Multiscreens) is 12.

          The multiscreen(M) manual page says that the default
          number of console multiscreens is 12, but the maximum
          is 16.

          Which is correct?

PROBLEM:  I cannot open the raw device on a SCSI hard disk that reports
          exactly 2048 cylinders to the operating system.  For example,
          if the root disk is SCSI, then the command "dd if=/dev/rhd00"
          produces "dd: read error: No such device or address.  The command
          "hd /dev/rhd00" produces no errors but returns a count of "0000".  
          This has been known to cause hard disk boot failure after
          installation, since masterboot code is written to the device
          by opening and writing to /dev/rhd00.  Reading or writing on
          /dev/hd00 in all these cases is successful.

PROBLEM:  This message appears on the console screen:

              Warning: table_grow - exec data table page limit of 25 pages
                                      (MAXEXECARGS) exceeded by 1 pages

PROBLEM:  I would like to create a backup copy of the tape distribution for
          my SCO OpenServer Release 5 product. How do I do this?

PROBLEM:  In previous releases of SCO UNIX software, the Sysamdsh(ADM) was
          used to control which password entity, the protected password
          database or /etc/password, was considered the "master" file
          in the event that discrepancies developed between the two files.

          However, I do not see this functionality in the SCOAdmin(F)
          Account Manager utility in SCO OpenServer Release 5.

          How do I control which password entity is the master file in
          the event of password file corruption?

PROBLEM:  The error below is displayed when running sort(C), or when
          attempting to relink the kernel:

                       sort: allocation error before sort

PROBLEM:  How do I configure a RAID 10 array using SCO Virtual Disk
          Manager Release 1.1.0?

          RAID 10 is a virtual disk consisting of striped mirrors
          (RAID 1 + RAID 0), which are usually duplexed (the disks
          use separate buses for I/O). The minimum number of pieces
          in a RAID 10 array is four, which consists of two two-disk
          mirrors striped together into one virtual disk. RAID 10 offers
          the best rating for I/O performance and rates very high in the
          Resilience category.

          The following example assumes that there are four 525 megabyte
          SCSI disks that are attached to two SCSI host adapters.  None are
          part of the root or bootable hard drive, and the entire disk space
          for each of the drives will be used in the RAID 10 array.

PROBLEM:  When restoring an SCO OpenServer Release 5.0 filesystem or
          directory structure using 'tar -x' onto an SCO Open Server 3.0
          server, I receive the error:  

            tar: ./<dir_name>/<fname> - cannot create  
            tar: cannot make directory ./<dir_name>/<fname>  

          This error will occur even if the subdirectory structure
          already exists on the 3.0 system.

PROBLEM:  I need to determine whether Release Supplements 5.0.0c, 5.0.0d,
          and/or the SCO OpenServer Networking Supplement Release 1.0.0
          have been installed on my SCO OpenServer Release 5 system.  These
          are sometimes referred to as "rs500c", "rs500d", and "net100a",
          respectively.

PROBLEM:  What is contained in Support Level Supplement (SLS) UOD426B,
          the SCO Year 2000 SLS for SCO UNIX System V/386 and SCO Open
          Desktop?