Overview
| Comment: | Fix for the > crash. Maybe... |
|---|---|
| Downloads: | Tarball | ZIP archive | SQL archive |
| Timelines: | family | ancestors | descendants | both | v1.6569-gt-crash-fix |
| Files: | files | file ages | folders |
| SHA1: |
5a05fc04ff63c390084de02ff6f51bcb |
| User & Date: | matt on 2021-01-26 14:00:48 |
| Other Links: | branch diff | manifest | tags |
Context
|
2021-01-26
| ||
| 20:05 | Merged in the removeal of fastmode (which did nothing) check-in: 4d5af74171 user: matt tags: v1.65-lazyqueue-items-rollup | |
| 14:00 | Fix for the > crash. Maybe... Leaf check-in: 5a05fc04ff user: matt tags: v1.6569-gt-crash-fix | |
|
2020-09-18
| ||
| 17:30 | added check for file existence before file delete ==/14/1.9/WARN/orion,mars/== NOTE: This is the last v1.65 before the split off. I.e code from before this point IS in the far future v1.65 branch. Code from this point to that branch might NOT be in the branch. check-in: 2769e4b7c9 user: mmgraham tags: v1.65, v1.6569 | |
Changes
Modified db.scm from [fb3a18f52f] to [09786a6ce2].
| ︙ | ︙ | |||
3213 3214 3215 3216 3217 3218 3219 |
(if newstatus (sqlite3:execute db "UPDATE tests SET status=? WHERE id=?;" newstatus test-id))
(if newcomment (sqlite3:execute db "UPDATE tests SET comment=? WHERE id=?;" newcomment ;; (sdb:qry 'getid newcomment)
test-id))))))
(mt:process-triggers dbstruct run-id test-id newstate newstatus))
;; NEW BEHAVIOR: Count tests running in all runs!
;;
| | | | | | 3213 3214 3215 3216 3217 3218 3219 3220 3221 3222 3223 3224 3225 3226 3227 3228 3229 3230 |
(if newstatus (sqlite3:execute db "UPDATE tests SET status=? WHERE id=?;" newstatus test-id))
(if newcomment (sqlite3:execute db "UPDATE tests SET comment=? WHERE id=?;" newcomment ;; (sdb:qry 'getid newcomment)
test-id))))))
(mt:process-triggers dbstruct run-id test-id newstate newstatus))
;; NEW BEHAVIOR: Count tests running in all runs!
;;
(define (db:get-count-tests-running dbstruct run-id) ;; fastmode)
(let* ((qry ;; (if fastmode
;; "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND NOT (uname = 'n/a' AND item_path = '') LIMIT 1;"
"SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND NOT (uname = 'n/a' AND item_path = '');")) ;; )
(db:with-db
dbstruct
run-id
#f
(lambda (db)
(let* ((stmth (db:get-cache-stmth dbstruct db qry)))
(sqlite3:first-result stmth))))))
|
| ︙ | ︙ | |||
3243 3244 3245 3246 3247 3248 3249 |
;; "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id NOT IN (SELECT id FROM runs WHERE state='deleted') AND NOT (uname = 'n/a' AND item_path = '');")
"SELECT count(id) FROM tests WHERE state in ('RUNNING','REMOTEHOSTSTART','LAUNCHED') AND run_id=?;"
run-id)))) ;; NOT IN (SELECT id FROM runs WHERE state='deleted');")
;; NEW BEHAVIOR: Look only at single run with run-id
;;
;; (define (db:get-running-stats dbstruct run-id)
| | | | | | 3243 3244 3245 3246 3247 3248 3249 3250 3251 3252 3253 3254 3255 3256 3257 3258 3259 3260 |
;; "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id NOT IN (SELECT id FROM runs WHERE state='deleted') AND NOT (uname = 'n/a' AND item_path = '');")
"SELECT count(id) FROM tests WHERE state in ('RUNNING','REMOTEHOSTSTART','LAUNCHED') AND run_id=?;"
run-id)))) ;; NOT IN (SELECT id FROM runs WHERE state='deleted');")
;; NEW BEHAVIOR: Look only at single run with run-id
;;
;; (define (db:get-running-stats dbstruct run-id)
(define (db:get-count-tests-running-for-run-id dbstruct run-id) ;; fastmode)
(let* ((qry ;; (if fastmode
;; "SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=? LIMIT 1;"
"SELECT count(id) FROM tests WHERE state in ('RUNNING','LAUNCHED','REMOTEHOSTSTART') AND run_id=?;")) ;; )
(db:with-db
dbstruct
run-id
#f
(lambda (db)
(let* ((stmth (db:get-cache-stmth dbstruct db qry)))
(sqlite3:first-result stmth run-id))))))
|
| ︙ | ︙ |
Modified rmt.scm from [39d97c528a] to [169f275811].
| ︙ | ︙ | |||
659 660 661 662 663 664 665 | (map (lambda (run-id) (rmt:send-receive 'test-get-paths-matching-keynames-target-new run-id (list run-id keynames target res testpatt statepatt statuspatt runname))) run-ids)))) (define (rmt:get-prereqs-not-met run-id waitons ref-test-name ref-item-path #!key (mode '(normal))(itemmaps #f)) (rmt:send-receive 'get-prereqs-not-met run-id (list run-id waitons ref-test-name ref-item-path mode itemmaps))) | | | | | | 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 | (map (lambda (run-id) (rmt:send-receive 'test-get-paths-matching-keynames-target-new run-id (list run-id keynames target res testpatt statepatt statuspatt runname))) run-ids)))) (define (rmt:get-prereqs-not-met run-id waitons ref-test-name ref-item-path #!key (mode '(normal))(itemmaps #f)) (rmt:send-receive 'get-prereqs-not-met run-id (list run-id waitons ref-test-name ref-item-path mode itemmaps))) (define (rmt:get-count-tests-running-for-run-id run-id) (rmt:send-receive 'get-count-tests-running-for-run-id run-id (list run-id))) (define (rmt:get-not-completed-cnt run-id) (rmt:send-receive 'get-not-completed-cnt run-id (list run-id))) ;; Statistical queries (define (rmt:get-count-tests-running run-id) (rmt:send-receive 'get-count-tests-running run-id (list run-id))) (define (rmt:get-count-tests-running-for-testname run-id testname) (rmt:send-receive 'get-count-tests-running-for-testname run-id (list run-id testname))) (define (rmt:get-count-tests-running-in-jobgroup run-id jobgroup) (rmt:send-receive 'get-count-tests-running-in-jobgroup run-id (list run-id jobgroup))) |
| ︙ | ︙ |
Modified runs.scm from [030b929939] to [a0489a126c].
| ︙ | ︙ | |||
319 320 321 322 323 324 325 |
;; Take advantage of a good place to exit if running the one-pass methodology
(if (and (> (runs:dat-can-run-more-tests-count runsdat) 20)
(args:get-arg "-one-pass"))
(exit 0))
(if (runs:dat-load-mgmt-function runsdat)((runs:dat-load-mgmt-function runsdat)))
| | | 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 |
;; Take advantage of a good place to exit if running the one-pass methodology
(if (and (> (runs:dat-can-run-more-tests-count runsdat) 20)
(args:get-arg "-one-pass"))
(exit 0))
(if (runs:dat-load-mgmt-function runsdat)((runs:dat-load-mgmt-function runsdat)))
(let* ((num-running (rmt:get-count-tests-running run-id))
(num-running-in-jobgroup (rmt:get-count-tests-running-in-jobgroup run-id jobgroup))
(job-group-limit (let ((jobg-count (configf:lookup *configdat* "jobgroups" jobgroup)))
(if (string? jobg-count)
(string->number jobg-count)
jobg-count))))
(if (> (+ num-running num-running-in-jobgroup) 0)
(runs:inc-can-run-more-tests-count runsdat)) ;; (set! *runs:can-run-more-tests-count* (+ *runs:can-run-more-tests-count* 1)))
|
| ︙ | ︙ | |||
1563 1564 1565 1566 1567 1568 1569 | )) extras) extras) '()))) (waitons (delete-duplicates (append (tests:testqueue-get-waitons test-record) extra-waits) equal?)) (newtal (append tal (list hed))) (regfull (>= (length reg) reglen)) | | | 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 | )) extras) extras) '()))) (waitons (delete-duplicates (append (tests:testqueue-get-waitons test-record) extra-waits) equal?)) (newtal (append tal (list hed))) (regfull (>= (length reg) reglen)) (num-running (rmt:get-count-tests-running-for-run-id run-id)) (testdat (make-runs:testdat hed: hed tal: tal reg: reg reruns: reruns test-record: test-record test-name: test-name |
| ︙ | ︙ | |||
1829 1830 1831 1832 1833 1834 1835 |
(rmt:set-var (conc "lunch-complete-" run-id) "yes")
;; now *if* -run-wait we wait for all tests to be done
;; Now wait for any RUNNING tests to complete (if in run-wait mode)
;; (if (runs:dat-load-mgmt-function runsdat)((runs:dat-load-mgmt-function runsdat)))
(thread-sleep! 10) ;; I think there is a race condition here. Let states/statuses settle
| | | | | 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 |
(rmt:set-var (conc "lunch-complete-" run-id) "yes")
;; now *if* -run-wait we wait for all tests to be done
;; Now wait for any RUNNING tests to complete (if in run-wait mode)
;; (if (runs:dat-load-mgmt-function runsdat)((runs:dat-load-mgmt-function runsdat)))
(thread-sleep! 10) ;; I think there is a race condition here. Let states/statuses settle
(let wait-loop ((num-running (rmt:get-count-tests-running-for-run-id run-id))
(prev-num-running 0))
;; (debug:print-info 13 *default-log-port* "num-running=" num-running ", prev-num-running=" prev-num-running)
(if (and (or (args:get-arg "-run-wait")
(equal? (configf:lookup *configdat* "setup" "run-wait") "yes"))
(> num-running 0))
(begin
;; Here we mark any old defunct tests as incomplete. Do this every fifteen minutes
;; (debug:print 0 *default-log-port* "Got here eh! num-running=" num-running " (> num-running 0) " (> num-running 0))
(if (> (current-seconds)(+ last-time-incomplete 900))
(let ((actual-num-running (rmt:get-count-tests-running-for-run-id run-id)))
(debug:print-info 0 *default-log-port* "Marking stuck tests as INCOMPLETE while waiting for run " run-id
". Running as pid " (current-process-id) " on " (get-host-name))
(set! last-time-incomplete (current-seconds)) ;; FIXME, this might be causing slow down - use of set!
(rmt:find-and-mark-incomplete run-id #f)
(debug:print-info 0 *default-log-port* "run-wait specified, waiting on " actual-num-running
" tests in RUNNING, REMOTEHOSTSTART or LAUNCHED state at "
(time->string (seconds->local-time (current-seconds))))))
;; (if (runs:dat-load-mgmt-function runsdat)((runs:dat-load-mgmt-function runsdat)))
(thread-sleep! 5) ;; (if (>= num-running max-concurrent-jobs) 5 1))
(wait-loop (rmt:get-count-tests-running-for-run-id run-id)
num-running))))
;; LET* ((test-record
;; we get here on "drop through". All done!
;; this is moved to runs:run-testes since this function is getting called twice to ensure everthing is completed.
;; (debug:print-info 0 *default-log-port* "Calling Post Hook")
;; (runs:run-post-hook run-id)
(debug:print-info 1 *default-log-port* "All tests launched")))
|
| ︙ | ︙ |