- 109: Base_systems, Site , Fri 27-Oct-2000 10:22:52 MDT, Corrected ops.config
The swap from array 2 to array 3 was completed at 17:25 PDT, Sept 15
This was mistakenly entered into ops.config in local time. I changed
this today (10/26/00) to 00:25 GMT, Sept 16.
- 93: Base_systems, Site , Mon 25-Sep-2000 18:19:31 PDT, Changed prep.config for ops4
We made a wiring error on ops4, so I changed prep.config and
rebooted cosmos around 1820 PDT, September 25.
The changes were:
Sonic Old channel New Channel
S1 202 203
S2 203 204
S3 204 202
- 92: Base_systems, Site , Mon 25-Sep-2000 18:17:21 PDT, Started ops4
We started reconfiguring the array from geometry 3 to geometry 4
around 1300 PDT, September 24 and completed it around 1730 PDT,
September 25. I changed ops from 3 to 4 at 0000 (GMT?) September 25.
- 61: Base_systems, Site , Wed 13-Sep-2000 15:20:07 PDT, how to monitor the breezecom performance
Use the Ambra notebook computer to
check on the status of the wireless
lan units. The Ambra is connected
to the network, and has software to
talk to the breezecom unit in the
trailer, which is known as the access
point.
If not already on, power on the Ambra.
You have to hold in the power button
on the back of the unit for a few seconds.
Choose the trackball option when
booting.
After W95 is running, run the the
DS11 program from the icon on the desktop.
Enter the aster password in the
"community" box, in the lower left
corner of the screen.
Next, click on "network autodiscovery"
in the upper right corner. The IP addresses
of the breezecom devices will be listed in
the box below the autodiscovery button.
Click on any of the IP addresses to check
the status of that unit. The access
point has AP listed after it; the remote
units have SA designations.
If you choose an SA unit, you get a nice
metter indication of the link quality.
One of the links shows 100%, the other
fluctuates between 90-98%.
Don't necessarily trust the locations listed in
the address for a given SA.
- 48: Base_systems, Site , Mon 11-Sep-2000 16:13:15 PDT, aster backup
aster disks backed up to exabyte tape #4.
- 47: Base_systems, Site , Mon 11-Sep-2000 15:50:19 PDT, how to do aster system backups
This procedure dumps the file systems mounted on aster
to an exebyte tape. The tape drive on clover is used
for writing the tape.
A scrap of paper in the top of the box containing the
backup tapes is used to record the numerical cycling
of the backup tapes.
steps:
1. put the next log tape into /dev/rmt/0 on clover.
2. log onto aster as root.
3. cd /usr/local/adm/backup
4. ./rbackup clover root /dev/rmt/0mn
It will tape forever to finish the backups.
- 46: Base_systems, Site , Mon 11-Sep-2000 15:45:46 PDT, system down for 15 minutes
Aster was shut down for about 40 minutes, from
1235 until 1310 PDT. This was to remove the magma serial port
card, so that it could be taken back to NCAR to
be used on the VTMX project.
The GPS signal was switched to /dev/cua/a on aster.
- 43: Base_systems, Site , Sun 10-Sep-2000 13:56:45 PDT, monitoring status
To check the status of aster processes, adamserver, nc_server, ingest
covar, and archive, enter (from aster):
check_aster
The adams and aster processes all write messages to aster's syslog.
The syslog is displayed on aster's console and written to
/var/log/local/isff.log
on aster. $ASTER/log/isff.log is a link to /var/log/local/isff.log.
This link only works on aster.
- 41: Base_systems, Site , Sun 10-Sep-2000 12:48:11 PDT, Changing ops
To change ops:
1. cd $ASTER/projects/SGS00
2. make a new directory for the ops:
mkdir ops99
3. copy 5 config files to it (channel_config, prep.config,
archive_config, ingest.conf, covar.config).
Ideally the only thing that should change during SGS are
the variable names, so you only need to edit prep.config
and covar.config.
4. edit ops.config. Enter a new ops name and start/stop times.
Remember to set the stop time of the previous ops period
# pre-project testing
ops0 2000 jun 7 00:00:00 - 2000 sep 1 00:00:00
# array #1
ops1 2000 sep 1 00:00:00 - 2000 sep 9 18:00:00
# array #2
ops2 2000 sep 9 18:00:00 - 2000 oct 1 00:00:00
5. Check the ops:
getops now
showproject
6. reboot the adams. From aster:
mxreset cosmos
enter a comment for syslog
mxreset daisy
7. watch the console on aster for errors as the adams are rebooting.
- 20: Base_systems, Site , Fri 01-Sep-2000 15:53:18 PDT, Sys Admin info
Networking:
192.168.80.*
The 16 port ethernet hub underneath the printer serves the
192.168.80.* net, which is where all new hosts should be added.
Aster is a DHCP server, so you can configure systems to use dhcp
and just plug them into the hub.
128.117.80.*
This is the net between aster and the adams. We want to keep
extraneous traffic off of it. The 4 port hub in the trailer,
on the shelf to the right of aster serves this network.
cosmos -> Breezecom SA -> RF \ / aster
\__ Breezecom AP -> 4port hub /
/ \
daisy -> Breezecom SA -> RF / \ ambra
NIS: passwd, hosts, group, and automount maps etc are provided via NIS.
To change them, edit the file on aster and do a NIS make (see below).
The passwd and shadow files are in aster:/var/yp, the
rest are on /etc, as normal.
NIS make: After making a change, do the following as root on aster:
cd /var/yp
/usr/ccs/bin/make
This reports an error about a "k" target, which you can ignore.
Since you can use DHCP, you probably won't need to add
any hosts.
Automounting: the auto mount maps are on aster: /etc/auto_master,
auto_home and auto_data. Using the net automount map,
any shared disk can be automounted as /net/host/mountprint
where host is the host name, and mountpoint is the
mount point on that host.