Weekly log #172 for week ending 03-Jul-1998

Moved Kate's workstation to her new temporary office. Got Mike Black to patch up one of the old network outlets in there so it could be connected directly back to our hub in the telemetry room.
Helped Katrin with locating her equipment in the telemetry room while the old room is remodeled.
Installed Solaris on the Machine Formerly Known As Terra10. The new nodename is 'nitya', and it will be used as a workstation for the new Oracle database person who is starting next week.
Set up access to the AlliedSignal intranet web site for ATSC personnel here.
Got Gimp [Gnu Image Manipulation Program] and installed it on iron and ehzsouth.
Wrote up how to do Apache web server access restrictions for Lisa. This is now on the internal documentation web pages.
Installed the Claymore script on ehzsouth and flint to catch the cookbook hackers who have been trying to break in recently.

Weekly log #172 for week ending 10-Jul-1998

Made a change in the new Terra10 system memory configuration to fix a problem with the rtem startup script.
Upgraded ssh [secure shell] on unix systems to version 1.2.22.
The internal disk on Avalon failed. Replaced it with a spare external disk.
The monitor on Avalon failed. It was swapped with the monitor on Indio, and one of the spare black-and-white monitors was put on Indio.
Did the iron hardware swap and upgrade, replacing the old Ultra-1 system with a new Ultra-5 box. This required a reinstall of ClearCase.
The web server log on flint had a lot of errors associated with an incorrect path specification for background files used in the earthquake reports web pages. Fixed the path to clear this problem.
Called OES about the fact that we have not received any EDIS transmissions for several weeks.
Configured the old iron as a new node called 'rust' for use by Vince Quitoriano.
Installed fvwm on rust for Vince.

Per a request by the SCIGN group, built and installed the most current version of Xemacs [20.4] on ehzsouth.
Connected spring to the 63 subnet.
Set up two SMTP distribution lists for duty personnel to use in notifying the world about who is the current duty seismologist.
Katrin reported a problem with networking on spring. This turned out to be a typographical error in the default router specification. Fixing the typo cleared the problem.
Figured out how to set up a common mail area for the Trinet unix development machines. Set up sylmar and onyx to test this out.
Bob got a new disk to replace the TANK_09 disk on Carizo, which has been logging sporadic errors. Did the disk swap on Friday of this week.

Set up the new unix mail scheme on the rest of the Trinet machines.
Got quotes for some AUI to 10Base-T transceivers to adapt the older equipment on the 63 subnet to work with the 10Base-T hub we have for that subnet.
Got briefed by Phil about what to do to tend to the Digital Online System while he is on vacation next week.
Accidentally overwrote one of the system configuration files on flint and had to recover it off of a backup tape.
Got a demo copy of the current version of the Markvision Printer utility to control the Lexmark printer. Set up some printer queues to test.
Larry Watanabe from ITS came by and dropped off the paperwork for the Catalyst 5000 network switch in the basement of the Yellow House.
Made accounts on jet and iron for Joe Franck.

Set up a pair of queues for the Lexmark printer on ehzsouth. This will allow USGS users to print through queues on a machine that is under USGS control, which may help in the future when there are problems printing on the color printer.
Egill and Bob Busby came by and outlined their plans for doing datalogger Multishear upgrades this week. This requires changing some configuration files and restarting the comserv processes on jet. The first station was done on Tuesday.
Tejon crashed on Wednesday afternoon. This turned out to have been a fault with the system disk. Cycling power on the disk brought it back and allowed the system to reboot. No errors were logged before the crash, which indicates that the disk failed all at once.
During the recovery from the Tejon crash, we wondered why no 'cannot ping Tejon' pages were sent. It turned out that the paging daemon on Bigone had died at 8:20 on Wednesday morning when the system disk on Bigone had briefly been full. Restarting the paging software brought out all the queued pages for the day.
Installed the new VMS CSLG licenses on all systems.

***