DATE: December 9, 1993

TO: 140-ft operators

FROM: Ron Maddalena

SUBJECT: Some changes in MacII and Sun procedures.

We have been running the new archive system for the last few months and I would like to make the following changes to the procedures you have been following. Some of these things you are already doing but I'd like to formalize these thoughts for you. Perform all steps in the order given--the order of the steps is as important as the actions involved.

Case 1: Before dumping an archive:

Before making an archive tape, please do the following:

Stop taking data

Perform a DUMMY calibration observation (i.e., a CAL scan with DUMMY as a source name).

Start the dumping of the archive.

After the archive program says that it is "dumping" the archive, you can start taking data again.

Continue with the instructions below for Case 3.

If you don't perform these steps properly, occasionally data might not make it onto the archive tape.

Case 2: If Sun or MacII need to be shutdown:

If you or any other staff member need to shut down the MacII or Sun (e.g., if someone is about to do a backup, there will be a planned power outage, hung processes, etc.), you should do the following:

Stop taking data.

Perform a DUMMY calibration observation (i.e., a CAL scan with DUMMY as a source name).

Shut down the computer using the instructions in my previous memo.

Continue with the instructions below for Case 3.

Don't let staff or observers shut down the computer without first preforming these steps. Most staff are unaware that these precautions need to be taken. Only Nancy or I are aware of when these instructions shouldn't be followed (and the cases are not easily described and are few). Remember, the operators and staff intimately involved with the 140-ft must take responsibility for when the Sun or MacII are to be rebooted. Other staff members and astronomers must adhere to your decisions.

Case 3: After Sun or MacII have been rebooted or Archive made or after maintenance:

After any instance in which the Sun or MacII are rebooted, or after an archive tape has been made, or at the end of maintenance day, you should:

Make sure MacII and Sun communication is up (i.e., no error messages every few seconds on either machine). See Case 4 if the communications doesn't seem to want to start up properly.

Perform a few DUMMY calibration observation (i.e., a CAL scan with DUMMY as a source name).

The first or second DUMMY scan may not make it across. If so, wait until communications re-establish themselves and try again.

If after the third or forth try data is still not getting across to the Sun without an error message, try to follow the steps in Case 4.

If two or more scans get across OK, start taking data.

If you don't perform these steps, data is being lost!!


Case 4: If MacII and Sun communications fail to connect:

If the Sun and MacII link fails to restart itself properly and you keep getting error messages every few seconds for a few minutes on either the Sun or MacII, try the following:

Shutdown the MacII.

Shutdown the Sun.

Reboot MacII and wait until it has completely come up.

Reboot the Sun and wait for it to completely come up.

If things still give you trouble, call for help, else start taking data.

Case 5: Other MacII problems:

If you are getting CHECKSUM or SYNC BIT errors on the MacII, the problem lies in the Modcomp and MacII link and has nothing to do with the Sun. Errors of this type are displayed on the MacII screen and are designated by the messages 1*error, 2*error, 3*error, or 4*error where the '*' indicates a lower-case letter (e.g., 4cerror). If you see these error messages, sit back and see if they continue with the next record. If they persist, do the following:

Reboot the Modcomp and try to take data.

If problems persist, reboot MacII, wait for communications to re-establish themselves, and start taking data.

Call for help if even after rebooting MacII and Modcomp the problem continues.