As Scream! receives data blocks, it checks that each one is valid and self-consistent. Any errors it detects are highlighted in the ViewInfo window (see section 5.4, page 66.) They can also be logged to disk.
To specify this file, or to change which types of error are logged, choose File → Setup... from the main menu and move to the Event Log tab. Check Enable Event log to begin.
The top section of this window lets you select the types of event to be logged:
Block Sequence Errors : A block with an unexpected block number. Blocks received over a serial data link are marked with a block number, which increases by 1 for each new block. If Scream! receives a block whose number is out of sequence, it generates this error.
Suspicious System IDs : A block coming from an instrument which is reporting a System ID of zero (i.e. blank). Although the block itself is valid, there may be a problem with the device it came from.
Invalid Time Code : A block with an out-of-range time stamp (i.e. greater than 23:59:59.)
User Commands : A command initiated by the user which affects the operation of the instrument, e.g. locking, unlocking, centering, calibration, and configuration detection.
Invalid Compression Code : A block with a compression code other than 1, 2, or 4.
Invalid Block Duration : A block which is not a whole number of seconds long. GCF blocks must contain a whole number of seconds of data, calculated as number-of-samples × sample-rate.
End Sample Mismatch : A block whose final sample (reverse integrating constant) does not equal the first sample plus all the differences within the block.
Checksum Errors : A block whose checksum does not match the sum of all the bytes in the block.
Data Discontinuities : Any detected gap in a stream, with the reason if known. This includes reboots, clock syncs, and intentional gaps in triggered channels, as well as data lost owing to slow or poor-quality communications links. Status streams hold no duration information and are not checked.
By default, log files are created in the user's directory (Local Settings under Windows), under different names according to the source of the data:
local serial inputs are logged by port, in files named comportx.log ,
network servers are logged by IP address, in files named aa_bb_cc_dd.log (for the IP address aa.bb.cc.dd), and
file and SCSI disk replays are logged in the file replay.log .
Any events which do not relate to a particular stream are logged in the file scream.log . If you uncheck File for each data source, all events are logged to this file.
Log files can become quite large, and it is usual to limit their size. To do this, check Limit to... and set a value in the box. When a xxx.log file reaches this size, it will be renamed xxx.old, and a new, empty xxx.log file will be created. If the file xxx.old already exists, it is overwritten.
Finally, enter the name of a directory in the text box, or click Browse... to find a suitable directory for saving the log file(s).
You can also instruct Scream! to send e-mail to warn the operator when particular events happen. To set this up, choose File → Setup... from the main menu and move to the Email Options tab.
To e-mail, you will need access to a mail (SMTP) server which will accept messages from the address Scream_SMTP_client@your-hostname. Fill in the IP address or hostname of this server in the SMTP Server box. Enter the e-mail address which you want to receive this mail in the Recipient box.
The events you can monitor are:
Digitizer re-boots : Scream! can recognise the boot-up status messages of a Güralp digitizer or digital instrument. These are normal after a power cycle, or when a new configuration is uploaded to the digitizer. Ater a re-boot, you should receive Re-Sync and Data Discontinuity messages as well.
Re-Syncs : Whenever the digitizer's internal clock is resynchronized with GPS. These are normal after a re-boot, or an extended period of GPS outage.
Data Discontinuities : Any detected gap in a stream, with the reason if known. This includes reboots, clock syncs, and intentional gaps in triggered channels, as well as data lost owing to slow or poor-quality communications links. You should check the log files (see above) to help you determine the cause.
SAM disk full : SAM units send status messages when they stop recording data because their disk is full. SAM units in circular-buffer mode carry on recording, and do not send these messages.
Scream Start / Stop : Generated by Scream! when it starts, and when it shuts down cleanly. If a computer suffers a sudden loss of power or is not shut down properly, you will not see a Scream Stop message.
Scream recording stopped : Whenever recording fails for a particular stream, because of a write error on the local disk. Check that the disk is not full, and that you have write access to the relevant directories. If you have Auto Record – Enable active and have set recording to GCF format, replaying a previously-recorded GCF file will cause Scream! to attempt to record to the same file, causing this error. The file will be undamaged.
GPS Control Timeout : Generated if Scream! has previously received satisfactory GPS synchronization messages, but has not received any within the last 12 hours. Scream! will not send another GPS Control Timeout message until GPS operation resumes, at which point it will restart the 12-hour timer.
Digitizer Trigger Report : Whenever the digitizer triggers. Scream! determines this from lines in status messages, and not from the data streams. Accordingly, there may be a slight delay between a trigger and this message. The contents of the status lines are repeated in the message.
Mass Positions : Generated if an instrument reports a mass position over the value you specify, expressed as a percentage of the mass's travel. Most Güralp instruments can operate satisfactorily with mass positions up to 50% of their travel; some can operate with significantly higher mass positions.
Disk Full Warning : Generated if the amount of data on the hard disk exceeds a specified percentage of the size of the disk.