Linux Utilities

1. On the Service Desktop, select Utilities > Shutdown Applications.

2. Open a Shell and become <root> in the OC window.

3. Enter: su -

   Enter the super user (root) password.

4. Enter the following:

   root@hostname # setdate

   note:

   Type q to quit anytime. Press Enter to proceed.

   To be accurate, this tool prompts you to enter the second. Watch your clock or PC carefully to enter the proper value, and press the Enter key at the right second to set the accurate time. Press the Enter key to proceed.

   Enter the current year (1980 -2030) <2007>

   Enter the current Month (1-12) <04>

   Enter the current Day (1-30) <14>

   Enter the current Hour (Military Time) (0-23) <18>

   Enter the current Minute (0-59) <13>

   Enter the current Second (0-59) <00>

   Updating the time on the OC and DARC, Please Wait...

   Current OC date: Fri Apr 14 15:18:00 CDT 2007

   Could not retrieve the DARC TimeZone.

   Please make sure the timeZones for the OC and DARC are the same.

   Setdate completed with NO ERRORS.

   [root@hostname]#

5. Close the shell by typing: exit

6. Type: st to restart application software.

Storelog is run when there is insufficient disk space on the OC disk to bring up the system to Applications Level. During Applications startup, the disk management process checks for adequate disk space; if there is insufficient space, storelog will come up automatically in a shell prompting the user. The threshold level for the OC that prompts the need for storelog is when the partitions approach 97-98% full.

Storelog is a tool that recovers system disk space by removing files not essential to the operation of the system. Storelog provides an option to save the files, (core, log, data) to DVD prior to removing them from the system disks. Removing these “system log” files does not add image space, but should allow the applications to startup.

The storelog tool can be run standalone in a shell by simply typing storelog, or from the Service Desktop select ERRORLOG, and choose STORELOG.

In addition to the available command switches for sprsnap, there is a command switch, -r, that gives an on-screen warning about not executing this program during patient data acquisition.

Usage: sprsnap -h, -s, -M, -i

• -h: describe usage of sprsnap

• -s: run in silent mode, no questions

• -a: run in fully automated mode (HAST)

• -M: save to DVD automatically, no prompt

• -i: save log files only

• -o: ignore file size threshold

FILES SPRSNAP SAVES

Core files from the OC

/usr/g/service/log/core*

/usr/g/bin/core*

/usr/tmp/core*

UNIX kernel core files from the OC

/var/adm/crash/*

/usr/g/service/log from the OC

/var/adm/*SYSLOG* files from the OC

/var/adm/install*

Install log files from the OC

/usr/g/ctuser/logfiles/sdclog

SDC log files from the OC

ScanRx info files from the OC

/usr/g/service/log/exam*.protocol

/usr/g/service/log/exam*.scan.request

/usr/g/service/log/gesys_'uname -n'.log

/usr/g/queue

Miscellaneous information such as disk space, process status and showprods in Queue directory.

Scan files

/usr/g/data_manqagement/ex*/ex*_hdr

SDC log from the OC

usr/g/ctuser/logfiles/sdclog

Miscellaneous files form OC

/usr/g/service/log/*.timers

/usr/g/vin/*.timers*

/usr/g/en_US/app_defaults/archive/SCSI.fol

/usr/g/en_US/app_defaults/devices/camera.dev

/usr/g/config/INFO

/usr/g/config/dataacq/cfg

/usr/g/config/dataacq.init

/usr/g/config/scan_file_mgr.cfg

/usr/g/config/scan_disk_io.dfg

/usr/g/config/scanfilemgr.init

Use quicksnap to save a snapshot of the system so engineering can analyze it and try to understand the cause of system performance problems, system crashes, etc. The data for quicksnap is saved to a directory in the following location: /usr/g/service/log. The directory is named as follows: [hostname].[date-time string].inst. For example, msec7x2.06110294239.inst.

To run quicksnap from the desktop:

1. From any desktop toolbox, click on the Quick Snap button. A popup warning is displayed.

2. Click on the Okbutton on the warning screen. The screen remains on the monitor until the snapshot data collection is completed (usually under one minute).

To run quicksnap from a command line, open a UNIX shell and type one of the following commands:

• quicksnap -olc saves the snapshot data in the same manner as the desktop-based operation, but gives an on-screen indication of each step as it is performed.

• quicksnap -gui works the same as the desktop-based operation described previously.

• quicksnap -save gives the capability to select previously “snapped” data and save it to DVD (requires a blank, formatted DVD cartridge).

If you have a MOD upon which you want to put system files, this is different from image files. You prepare the MOD by making a file system on it. System State and DD File Analysis will detect this condition and prepare the MOD in the drive for you. To prep a system MOD under other circumstances, open a shell and type: mkfsMOD (formatting takes about 3-5 min).

Initialize a blank DVD cartridge for the System State Save, quicksnap, or sprsnap programs to use.

1. Make sure the DVD cartridge is not write-protected before attempting to initialize it.

2. Open a UNIX shell and type: mkfsDVD

To save the system configuration information, characterization, calibration, protocols, etc. to a System State DVD, perform the following tasks:

1. Bring the system up if it is not already up.

2. Insert the System State DVD (or MOD).

3. Click on the SERVICE DESKTOP.

4. On systems with GRE (GOC3 or GOC4) consoles the System State selection is located under UTILITIES.

   note:

   Click on the PM icon.

5. Click on SYSTEM STATE.

6. Click on ALL. This will highlight Cals, Characterization, Reconfig Info, etc.

7. Click on SAVE.

   The save will take a few minutes. Review the output for errors or missing files; the scroll bar on the right works only when the tool isn’t busy performing some task, it may take a little while.

8. Click on DISMISS.

Saving scan files to DVD or MOD requires that the media first be formatted for a Linux (or UNIX) file system. Refer to Initializing a DVD for formatting an DVD and to Initializing a MOD Cartridge for formatting an MOD. The Scan file save/restore function provides the ability to save/restore patient scan data (not images) to a DVD or MOD. This is used to allow engineering to analyze and troubleshoot unusual or intermittent data acquisition problems.

Saving scan files onto the MOD puts the scan files in the following directory path on the MOD: /MOD/service_mod_data/SFfiles<suiteid>.exam.series.scan file,. The scan files are saved as an iq. (referred to as a .iq file) in the Sfiles directory.

Restoring scan files function reads the MOD directory path: /MOD/service_mod_data/Sfiles and presents the files in a list/select GUI for restoring. Restoring the scan files takes a copy of .iq file on MOD and puts it into the scandata disk partition.

Both the Save and Restore Scan files functions reside under RECONMGMT on the top level ExamRx Desktop.

The reserve function allows you to prevent the overwriting of scan data files. The files can then be selected at a later time for future storage and/or reconstruction. Otherwise, eventually new scan data overwrites all of the existing scan data files. The release function unreserves any scan files previously reserved, freeing that scan file to be overwritten.

Beginning with software version 05MW14.5, a new script for database management was created. The advantage of the reinstall.db script is that it reinstalls all images and recreates the database indexes, thus rebuilding the Terra database, instead of destroying the database and recreating it (as in resetImageDB_Linux).

Perform the following steps to use reinstall.db:

1. Shutdown the Applications (Service Desktop>Utilities>Application Shutdown).

2. Open a Unix shell and type: reinstall.db. As the process runs, a screen displays the following message: Please Wait ... Database is being reconstructed.

   note:

   Applications must be shut down to use reinstall.db.

   The script reinstall rate is approximately 50 images/sec.

   The total time to reinstall all images is dependent on the database size.

   The CT host console is not available for other tasks during the database image reinstallation.

   The CT host console must be rebooted after the reinstall.db. script completes.

The resetImageDB_Linux command can be used on any of the current software. This command rebuilds the sdc_image_pool partition so all images are removed. sdc_image_pool partition is actually /dev/md0, which is a raid device created from striping two 73Gb disk drives together.

Perform the following steps to use resetImageDB_Linux:

1. Select [Shutdown], then select [Restart].

   note:

   There have been some issues with the raid being lost if resetImageDB_Linux is run after shutting down applications and not rebooting. This is why a reboot is required prior to running resetImageDB_Linux.

2. Select [Cancel] when the attention: You have 5 seconds to cancel Applications Startup is displayed.

3. Open a Unix Shell and type su -<password>.

4. In the Unix Shell type: resetImageDB_Linux.

5. When prompted, type yes to continue with the database reset.

6. When resetImageDB is complete, type exit and st to start applications.