>From: address@hidden >Organization: SMSU >Keywords: 199902121845.LAA13194 McIDAS-X 7.40 GRIDPLOT PANEL Bill, >I'll send you X files (not the TV show, but let's list then and see >where I end up) > >1) /home/mcidas/src/eta.sh >2) /home/mcidas/src/eta_1.sh >3) /home/mcidas/data/ETA_L.BAT >4) /home/mcidas/data/ETA_1.BAT > >Yeah, that ought to get you laughing pretty good. No. >1. eta.sh simply initiates eta_1.sh and eta_2.sh (last one not sent) >2. eta_1.sh does the 0-18 hr eta products... >3. ETA_L.BAT sets up strings, then lists the grids available for > eta_1.sh to count and judge whether it should make new maps. >4. ETA_1.BAT is the actual mapping. OK. >eta.sh is initiated periodically by cron. OK, I see this in the cron list: cirrus:/home/mcidas/workdata> crontab -l #5 19 * * * /home/mcidas/archive2.script 04 12 * * * /home/mcidas/rotate.cmd 12 * * * * cp /home/mcidas/data/mc/curwx.html /home/mcidas/.plan 30 9,15 * * * /home/mcidas/src/mcscour.sh #01 * * * * /bin/rm /home/mcidas/workdata/ROUTEPP.LOG # The next two entries must by synchronized...format after surf 10,20 * * * * /home/mcidas/src/surf.sh 12 * * * * /home/mcidas/data/mc/format_htmls 55 8,20 * * * /home/mcidas/src/500mb.sh 45 7,19 * * * /home/mcidas/src/lift.sh 45 7,19 * * * /home/mcidas/src/mrf.sh 50 20,21,22,9,10,11,12 * * * /home/mcidas/src/eta.sh >/home/mcidas/workdata/eta.sh.output # eta (0,1,2=0z,10,11,12=12z) Since the failure was "flakey", I decided to take you up on your offer to logon to your machine and see what could be amiss. I noticed that there were some files in ~mcidas/workdata that were remnamts of old sessions: -rw-rwxr-- 1 mcidas unidata 4352 Feb 06 12:22 FRAMED.001* -rw-rwxr-- 1 mcidas unidata 55488 Feb 15 17:39 FRAMENH.001* -rw-rwxr-- 1 mcidas unidata 8192 Feb 06 12:22 FRNV.001* -rw-rw-r-- 1 mcidas unidata 8004 Feb 06 12:22 FRNVDIR.001 -rw-rw-r-- 1 mcidas unidata 10240 Feb 06 12:22 FRNVEXP.001 -rw-rwxr-- 1 mcidas unidata 12000000 Feb 15 17:39 SYSIMAGE.SAV* Files of these names are supposed to be created and destroyed on a session-by- session basis (even mini-sessions started by cron or ROUTE PP BATCH invocations). Without looking too hard at the scripts you sent me (I actually did look at them in some detail), I judged that the flakey behavior could be caused by the above files existing in the first directory in your MCPATH (/home/mcidas/workdata). Since these files should be deleted anyway, I went ahead and did so. The next thing for you to comment on is whether or not this solved your problem. Tom
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.