Our LAN administrator is documenting our LAN and has asked me for the MAC address of the HP 3000. I have looked through the NMMGR screens and can't find it. I have found the MAC addresses for all of our DTCs, but not for the HP 3000. Anyone know where to find it?
Replies from Jeff Mikolai, Lee Gunter, Mike Hornsby:
The linkcontrol command does the trick, as in:
linkcontrol @;status=all (Hornsby also pointed out that the linkcontrol command is a
good tool to use when
performance is slow and nothing else on the system seems to be responsible.)
I'm having problems with 5.5 PowerPatch2 and Stage/iX. The "create stage" part fails. Examination of STAGELOG.INSTALL.SYS shows that all individual files are staged OK, but the "validate" fails because no "complete" is being done.
For the Stageman problem the following will complete the staging:
1. STAGEMAN> expert on will boot from this staging area.
2. STAGEMAN$ COMPLETE [stage_name]
3. STAGEMAN$ VALIDATE [stage_name]
4. STAGEMAN$ SET [stage_name]
5. START NORECOVERY
According to the man page for man, (man man) the Posix version of man
works like other
Unix versions in that you can add man pages simply by adding files to the
proper directory.
E.g. man[0-9]/*.[0-9]. I recently installed Samba which seems to have
installed the man
pages properly, however, whenever I try to look at them (man samba), man
reports:
ts: not foundand that's all I get. What am I doing wrong?
--More--(EOF)
Mark Klein replies
You can do this one of two ways:
Use the GNU groff from the FREEWARE tape;
Create the following script, call it 'ts', and place it somewhere in
your path:
#! /bin/shWe had a job "runaway" with the CPU 100 percent active, but no
disk activity -- your
typical "loop forever" programming construct. Faced with rebooting the
system to regain
control (I couldn't even get a ":" on the console), I decided to try the
Control-A at the
console, which worked. At the "=" I entered
/bin/nroff -man -e -Tascii $5ABORTJOB #Jn,
which to my surprise actually
worked! So, what other commands are available from the "=" prompt?
HP's Jeff Vance replies:
Here is the current list of control-A commands:
ABORTIO And I don't know where this is stated in our documentation,
although online HELP
does show the control-A syntax for commands that are legal in a control-A
environment
(namely, running on PROGEN's stack).
ABORTJOB
LOGOFF
LOGON
RECALL
REPLY
SHUTDOWN
HP's Goetz Neumann replies:
If you have a terminal with nobody logged on yet, it should be
possible to use a
:HELLO MANAGER.SYS;PRI=BS to get a session that can preempt
the runaway JOB. This
should work since the HELLO handling process (JSMAIN) is running in the BS
queue and
should be able to preempt any "normal" 100 percent CS-queue looping
job/session. Of
course this kind of logon should be used for diagnostic and troubleshooting or
CPU-lightweight tasks only, otherwise you can cause a system hang.
Another method many people use is running GlanceXL permanently, and "pushing" runaway jobs into the ES queue. If you have HP's Workload Manager you can ALTPROC the runaway process to a "zero-cpu" workgroup, i.e. a group with maxcpupercent = 0.
Be aware that for a job/session to successfully terminate (react on an ABORTJOB) it actually needs to get to the CPU. So first "starve it", then "shoot it", then "feed it" so "it dies."
Where can I find the magic number for FINFO to get a file's creation time?
HP's Jeff Vance replies:
Sorry, you'll have to wait for MPE/iX 5.5 Express 3 (Ed. note: Express
3's feature are now documented in the Communicator article
posted on the Jazz Web server, http://jazz.external.hp.com). New
FINFO functionality
includes number of extents, number of sectors, file creation time and
number of accessors.
How can I programmatically prevent MPE from requesting a lockword if it is not specified in the HPFOPEN formal designator?
Bruce Toback and Don Seay reply separately:
Try specifying a null lockword in the filename, as in filename/ or
filename/.pub.sys. I
believe that will return an error to HPFOPEN, instead of prompting for a
lockword if the
file has a lockword.
Our network group is changing the IP address for the Domain Name Server. Is there anyplace on the HP 3000 that I need to reference this change? I looked at HOSTS.NET.SYS, and we do not have a listing for the old IP address.
Several people jumped all over this including (and hopefully not
missing anyone) Andreas
Schmidt, Per Ostberg, Michael Gueterman, Joe Geiser and Ted Ashton:
Definitely in RESLVCNF.NET.SYS. Also in the /etc/resolv.conf file, if
it is not linked to
RESLVCNF.NET.SYS file. If neither of these files are set up with a
"nameserver" entry of the
old DNS server's IP address, then the 3000 was not using DNS, but rather
the HOSTS file
exclusively.
I have been asked if we will be able to access a new Unix host from the
DTC we currently
use to access two HP 3000 MPE/iX host systems. The DTCs are managed with
OPENVIEW
DTCMGR (WIN 3.1). If this is possible, which version of DTCMGR and Openview
would be
required?
Gilles Schipper replies:
You should have absolutely no problem accessing a Unix host with
whatever version
of DTCMGR and Openview you are currently using to access your existing HP
3000's.
Keep in mind, however, that unless you are configuring your DTC's to utilize a DNS server (somewhere on your network -- but not your HP 3000s), you must connect to your Unix boxes by IP address, rather than the node name that you currently use when connecting to your HP 3000s.
For example, without a DNS server somewhere, you must:
connect hpnodename (to connect to HP 3000) OR
connect 999.999.999.999 (where 999.999.999.999 is the IP
address of your Unix box)
What is the product ID to configure a FDDI card in SYSGEN on an HP 3000?
John Skelton replies:
You don't configure a FDDI card in SYSGEN. You configure it in NMMGR.
In the NS
part of NMMGR you define a FDDI "LINK" and give it the actual physical path
of the card.
I'm working on a project that requires the use of FTP. I want to grab
several files from a
Unix box and plop them onto our HP 3000. My problem lies in the MGET
command. When
I try to use it (ftp> mget eop*) to retrieve all files beginning with "eop"
I get errors like:
"An invalid character was found in an MPE syntax name."
Use "ftp> mget ./eop*". The "./" is necessary to tell MPE you are dealing with the Posix name space.
I've gotten FTP to work inside my company, but when I try to go outside I get a SOCKET Timeout Error. I know that the communications people in my company have a DNS server and a firewall somewhere. Could that be my problem?
Mark Bixby replies:
I'd guess any of the following problems:
1) The nameserver entries in RESLVCNF.NET.SYS on your 3000 aren't valid.
2) A firewall could be preventing external DNS resolutions. Since you're not running BIND/iX (I assume), your 3000 must query another machine that's running a DNS server, and the firewall is blocking *that* DNS machine from going outside, not your 3000.
3) The non-3000 DNS machine may not be properly configured with the root cache info from ftp://rs.internic.net /domain/named.root. In this scenario, you would only be able to resolve internal names hosted locally on the same DNS server or others within your organization.
Can you do OS upgrades via remote console?
HP's Lars Appel replies:
I do this frequently at work. We have the remote consoles plugged into
a DTC and
configured back-to-back. This way I can connect to the remote console from
my desk via
DTC port. Pretty useful. Includes full Ctrl-A and Ctrl-B access. But I
still have to walk into
the EDP room to insert/swap DDS tapes. I've even done an update of a
customer system via
modem and remote console.