HOW TO MANUALLY PURGE FILES =========================== - Check to see if the disk is getting close to full - There is a chkdsk (check disk) program running that notifies us by email if any of the partitions we check get over 90% - F5 contains all the high rate MDI data and fills up the quickest (usually) - To manually check the disk status, in an open window type: df -k - There is a column that is labeld %Used, check the status, check the /ftp, /ftp/.F1, /ftp/.F2, /ftp/.F3 and /ftp/.F5 directories for usage - If one or more of the disks is over 90% full, manual purge can be used - In an open window type: startmpurge - This script will start the manual purge process - After the manual purge process is started type: runmpurge - This script starts the actual purging of the files NOTE: It takes a while to run the process because it is going through each APID and purging the files - The scripts for all the purging processes can be found in the /ecs/exe directory - The startall script will call the automated purge process which runs every day at 15:00 UT - The file to change the automated purge process time can be found in the /home/ecs/ecs/ecs_config directory - To change the automated purge times, edit the file Purge_Schedule_Def_Setup_File - The file is set up like this --> File Description TIME File Path Purge_Schedule_Def_Setup_File Sample: SFDU_Files 1 15:00 /ftp/tlm_files/quicklook *.SFDU SFDU_Files 3 15:00 /ftp/tlm_files/quicklook *.DAT1 SFDU_tcf 28 15:00 /ftp/ancillary_data/tcf SO_AN_TCF*.SFDU DAT_sdr 3 15:00 /ftp/ancillary_data/sdr SO_AN_SDR*.DAT SFDU_sdr 3 15:00 /ftp/ancillary_data/sdr SO_AN_SDR*.SFDU Orbit_Data 7 15:00 /ftp/cdhf_files/orbit_data * Attitude_Data 7 15:00 /ftp/cdhf_files/attitude_data * - To change the purge schedule all you need to do is change the time, the quickest is to open the file and do a global replace for the string (ex: Ctrl R in xedit) NOTE: The purge process is always running if the ECS software is running, it will ocassionally fail but it happens very infrequently. If it the process does fail, you can restart the automated purge procress with the nstartpurge script, or run the manual purge. Created: 17 June 2009 by Gavin Westenburger