Megatest

Check-in [08b1dfe720]
Login
Overview
Comment:Missed case of relying on run_duration field - fixed. ==/FAIL/orion,mars/==
Downloads: Tarball | ZIP archive | SQL archive
Timelines: family | ancestors | descendants | both | v1.65-failed-testdat
Files: files | file ages | folders
SHA1: 08b1dfe720a9d4f286547beb0fe70ce2ad98ac8b
User & Date: mrwellan on 2020-09-24 15:50:34
Original Comment: Missed case of relying on run_duration field - fixed.
Other Links: branch diff | manifest | tags
Context
2020-10-05
06:43
Removed suspect change in server that seemed to be causing the obfusc-skel fails check-in: 134d87e018 user: matt tags: v1.71_old
2020-10-03
21:25
Pull in all v1.65 before patching with prereq fix. check-in: b50ac24a01 user: matt tags: v1.70-defunct
20:20
Patched in fix for prereqs-not-met issue. Closed-Leaf check-in: 7ee0b77302 user: matt tags: v1.65-refactor-fix-2, v1.65-failed-testdat (unpublished)
2020-09-24
23:28
Try locking transactions in no-sync db check-in: 44919a2a7f user: mrwellan tags: v1.65-transaction-locking
15:50
Missed case of relying on run_duration field - fixed. ==/FAIL/orion,mars/== Closed-Leaf check-in: 08b1dfe720 user: mrwellan tags: v1.65-failed-testdat
2020-09-23
15:16
Do not exit on failure to create directory - race conditons on NFS cause false fail scenarios - just keep going and cross your fingers... ==/FAIL/orion,mars/== check-in: c5b1d00b22 user: mrwellan tags: v1.6572, v1.65-failed-testdat
Changes

Modified db.scm from [d7c987aadc] to [2c7b396933].

1774
1775
1776
1777
1778
1779
1780


1781
1782
1783
1784
1785





















1786
1787
1788
1789
1790
1791
1792
1774
1775
1776
1777
1778
1779
1780
1781
1782





1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810







+
+
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+







	  (debug:print 0 *default-log-port* "ERROR: cannot read " infile)
          (debug:print 0 *default-log-port* "ERROR: run-dir is " run-dir)
          #f
          )
        (with-input-from-file infile read-lines)
	)))

;; check duration against test-run.dat file if it exists and update the value in
;; the db if necessary
;;  select end_time-now from
;;      (select testname,item_path,event_time+run_duration as
;;                          end_time,strftime('%s','now') as now from tests where state in
;;      ('RUNNING','REMOTEHOSTSTART','LAUNCHED'));

;;
(define (db:adjust-run-duration dbstruct test-id run-dir event-time run-duration)
  (let* ((datf             (conc run-dir ".mt_data/test-run.dat"))
	 (modt             (if (and (file-exists? datf)
				    (file-read-access? datf))
			       (file-modification-time datf)
			       #f)) ;; (+ event-time run-duration))))
	 (alt-run-duration (if modt
			       (- modt event-time)
			       #f)))
    (if (and alt-run-duration
	     (> alt-run-duration run-duration))
	(begin
	  (debug:print 0 *default-log-port* "Test " test-id " run duration mismatch. Setting to " alt-run-duration)
	  (db:with-db
	   dbstruct #f #f
	   (lambda (db)
	     (sqlite3:execute db "UPDATE tests SET run_duration=? WHERE id=?;" alt-run-duration test-id)
	     #t)))
	#f))) ;; #f = we did NOT adjust the time
	      
(define (db:find-and-mark-incomplete dbstruct run-id ovr-deadtime)
  (let* ((incompleted '())
	 (oldlaunched '())
	 (toplevels   '())
          ;; The default running-deadtime is 720 seconds = 12 minutes.
          ;; "(running-deadtime-default (+ server-start-allowance (* 2 launch-monitor-period)))" = 200 + (2 * (200 + 30 + 30))
         (deadtime-trim (or ovr-deadtime (configf:lookup-number *configdat* "setup" "deadtime")))
1826
1827
1828
1829
1830
1831
1832

1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843











1844
1845
1846

1847
1848

1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859











1860
1861
1862
1863
1864
1865
1866
1844
1845
1846
1847
1848
1849
1850
1851











1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864

1865
1866
1867
1868











1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886







+
-
-
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+


-
+


+
-
-
-
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
+
+
+







	 ;; HOWEVER: this code in run:test seems to work fine
	 ;;              (> (- (current-seconds)(+ (db:test-get-event_time testdat)
	 ;;                     (db:test-get-run_duration testdat)))
	 ;;                    600) 
	 ;; (db:delay-if-busy dbdat)
	 (sqlite3:for-each-row 
	  (lambda (test-id run-dir uname testname item-path event-time run-duration)
	    (if (not (db:adjust-run-duration dbstruct test-id run-dir event-time run-duration))
	    (if (and (equal? uname "n/a")
		     (equal? item-path "")) ;; this is a toplevel test
		;; what to do with toplevel? call rollup?
		(begin
		  (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		  (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
		(begin
		  (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted))
		  (debug:print-info 0 *default-log-port* "Found old test in RUNNING state, test-id="
				    test-id" exceeded running-deadtime "running-deadtime" now="(current-seconds)
				    " event-time="event-time" run-duration="run-duration))))
		(if (and (equal? uname "n/a")
			 (equal? item-path "")) ;; this is a toplevel test
		    ;; what to do with toplevel? call rollup?
		    (begin
		      (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		      (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
		    (begin
		      (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted))
		      (debug:print-info 0 *default-log-port* "Found old test in RUNNING state, test-id="
					test-id" exceeded running-deadtime "running-deadtime" now="(current-seconds)
					" event-time="event-time" run-duration="run-duration)))))
	  stmth1
	  run-id running-deadtime) ;; default time 720 seconds
       
	    
	 (sqlite3:for-each-row 
	  (lambda (test-id run-dir uname testname item-path event-time run-duration)
	    (if (not (db:adjust-run-duration dbstruct test-id run-dir event-time run-duration))
	    (if (and (equal? uname "n/a")
		     (equal? item-path "")) ;; this is a toplevel test
		;; what to do with toplevel? call rollup?
		(begin
		  (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		  (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
		(begin
		  (debug:print-info 0 *default-log-port* "Found old test in REMOTEHOSTSTART state, test-id=" test-id
				    " exceeded running-deadtime "running-deadtime" now="(current-seconds)" event-time="event-time
				    " run-duration="run-duration)
		  (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted)))))
		(if (and (equal? uname "n/a")
			 (equal? item-path "")) ;; this is a toplevel test
		    ;; what to do with toplevel? call rollup?
		    (begin
		      (set! toplevels   (cons (list test-id run-dir uname testname item-path run-id) toplevels))
		      (debug:print-info 0 *default-log-port* "Found old toplevel test in RUNNING state, test-id=" test-id))
		    (begin
		      (debug:print-info 0 *default-log-port* "Found old test in REMOTEHOSTSTART state, test-id=" test-id
					" exceeded running-deadtime "running-deadtime" now="(current-seconds)" event-time="event-time
					" run-duration="run-duration)
		      (set! incompleted (cons (list test-id run-dir uname testname item-path run-id) incompleted))))))
	  stmth2
	  run-id remotehoststart-deadtime) ;; default time 230 seconds
	 
	 ;; in LAUNCHED for more than one day. Could be long due to job queues TODO/BUG: Need override for this in config
	 ;;
	 ;; (db:delay-if-busy dbdat)
	 (sqlite3:for-each-row