[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20040301: new ldm at stc cont

>From: "Anderson, Alan C. " <address@hidden>
>Organization: St. Cloud State
>Keywords: 200402182057.i1IKvTrV011544 McIDAS-X setup

Hi Alan,

>Thanks for the analysis.  I figured it was something rather simple on
>my end.

No worries.  It is the simple stuff that really trips us up most of
the time.

>After your message a couple of weeks back about the clock time,  I
>decided to change the system clock to UTC and I messed up.   I know
>that I also need to run ntp or something similar to maintain it.

We use ntpdate run out of cron about every 15 minutes.  ntpd can
be run of course as well (but not at the same time as ntpdate)

>I notice that DATALOC LIST  RTPTSRC   shows  LOCAL DATA.   I will want
>this to be available from other terminals  so will need to change it
>to full host name.

I changed this when I logged onto cyclone this morning.  You can
change it back to cyclone.stcloudstate.edu when you get the chance.
By the sway, the DATALOC setting for a user will not have any effect
on the running of the remote interface that is setup.  You could
leave the DATALOC to LOCAL-DATA for 'mcidas' and everything would
keep working fine.

>The log file problem was also simple.   I had  used the entry shown in
>the LDM Preinstall material on the web, copied the line just as
>written:  local0.debug /usr/local/ldm/logs/ldmd.log
>
>After checking,  restarting syslogd  and still no log file,  I put in
>tabs after  local0.debug and then it worked.   I know that tabs must
>sometimes be used and that spaces will not work but did not think of
>that when it was written.

Yes, tabs are required on some system, but not on others.  This
makes it hard to advise folks when they run more than one OS.

>So, except to change the DATALOC  listing for  RTPTSRC  I think it is
>ready.   I will check my access from the other terminals  that cyclone
>will serve  and then we will be able to turn off waldo.    Waldo has
>been a very reliable machine.    A  Gateway pc running  Solaris v 7 for
>intel, We have never had any hardware failures.   I think it has run
>for 5 years or more.

That is good service from a PC.

>Now I have another question regarding another new machine.   We just
>received a  Sunfire dual processor server  with  2  SCSI disks  each
>about  70 Gb.    I am not sure what the Ram is,  I think about   10
>Gb.

Sounds great.  atm.geo.nsf.gov is a Sunfire 280R, and it is feeding
close to 100 downstream LDMs at the moment.

>What suggestions do  you  or others (Mike S. ?  ) have about how we
>should  partition the disks and set up file systems.    In the past,
>based on comments from Mike, I have just made one  large  file system
>on  /  (except for  swap).     On  cyclone, which also has  2 disks,
>I created the root  fs  on one disk  and  put  /var/data ...   on the
>other.

I will send this to Mike for his expert advice.

>The  OS  Solaris 9 is preinstalled,  but I suspect  the initial file
>system set up is not what will serve us the best.    We will use this
>machine like cyclone,  to serve data via ADDE  to our other terminals
>and to  run  the McIDAS sessions for  a number of  SUNRAY terminals .
>Our initial results  doing this on cyclone  have been  good.   The
>SUNRAY terminals are a small client  device from Sun  and they display
>sessions that are run on the server.

OK.

>Thanks Tom

Glad to help when I can...

Tom
--
NOTE: All email exchanges with Unidata User Support are recorded in the
Unidata inquiry tracking system and then made publically available
through the web.  If you do not want to have your interactions made
available in this way, you must let us know in each email you send to us.