Overview
Comment: | Check disk usage every 30 seconds instead of every 3 seconds |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | v1.64 |
Files: | files | file ages | folders |
SHA1: |
b60108422e9d9d99ada061488db95153 |
User & Date: | mrwellan on 2017-05-24 14:48:04 |
Other Links: | branch diff | manifest | tags |
Context
2017-05-24
| ||
14:50 | Bumped version to v1.6415 check-in: 18a4e536c2 user: mrwellan tags: v1.64 | |
14:48 | Check disk usage every 30 seconds instead of every 3 seconds check-in: b60108422e user: mrwellan tags: v1.64 | |
10:48 | Sync from test to server only when cpu changed, disk changed by more than 10% or more than 30 seconds past. check-in: 8ca2b352b3 user: mrwellan tags: v1.64 | |
Changes
Modified launch.scm from [4b0e8bb2f3] to [e2b2d5cae9].
︙ | ︙ | |||
315 316 317 318 319 320 321 | (if (steprun-good? logpro-used (launch:einf-exit-code exit-info)) (if (not (null? tal)) (loop (car tal) (cdr tal) stepname)) (debug:print 4 *default-log-port* "WARNING: step " (car ezstep) " failed. Stopping"))) (debug:print 4 *default-log-port* "WARNING: a prior step failed, stopping at " ezstep))))))) (define (launch:monitor-job run-id test-id item-path fullrunscript ezsteps test-name tconfigreg exit-info m work-area runtlim misc-flags) | > | > | | | | | | > > | | 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 | (if (steprun-good? logpro-used (launch:einf-exit-code exit-info)) (if (not (null? tal)) (loop (car tal) (cdr tal) stepname)) (debug:print 4 *default-log-port* "WARNING: step " (car ezstep) " failed. Stopping"))) (debug:print 4 *default-log-port* "WARNING: a prior step failed, stopping at " ezstep))))))) (define (launch:monitor-job run-id test-id item-path fullrunscript ezsteps test-name tconfigreg exit-info m work-area runtlim misc-flags) (let* ((update-period (string->number (or (configf:lookup *configdat* "setup" "test-stats-update-period") "30"))) (start-seconds (current-seconds)) (calc-minutes (lambda () (inexact->exact (round (- (current-seconds) start-seconds))))) (kill-tries 0)) ;; (tests:set-full-meta-info #f test-id run-id (calc-minutes) work-area) ;; (tests:set-full-meta-info test-id run-id (calc-minutes) work-area) (tests:set-full-meta-info #f test-id run-id (calc-minutes) work-area 10) (let loop ((minutes (calc-minutes)) (cpu-load (alist-ref 'adj-core-load (common:get-normalized-cpu-load #f))) (disk-free (get-df (current-directory))) (last-sync (current-seconds))) (let* ((over-time (> (current-seconds) (+ last-sync update-period))) (new-cpu-load (let* ((load (alist-ref 'adj-core-load (common:get-normalized-cpu-load #f))) (delta (abs (- load cpu-load)))) (if (> delta 0.1) ;; don't bother updating with small changes load #f))) (new-disk-free (let* ((df (if over-time ;; only get df every 30 seconds (get-df (current-directory)) disk-free)) (delta (abs (- df disk-free)))) (if (and (> df 0) (> (/ delta df) 0.1)) ;; (> delta 200) ;; ignore changes under 200 Meg df #f))) (do-sync (or new-cpu-load new-disk-free over-time))) (debug:print 4 *default-log-port* "cpu: " new-cpu-load " disk: " new-disk-free " last-sync: " last-sync " do-sync: " do-sync) (set! kill-job? (or (test-get-kill-request run-id test-id) ;; run-id test-name itemdat)) (and runtlim (let* ((run-seconds (- (current-seconds) start-seconds)) (time-exceeded (> run-seconds runtlim))) (if time-exceeded (begin (debug:print-info 0 *default-log-port* "KILLING TEST DUE TO TIME LIMIT EXCEEDED! Runtime=" run-seconds " seconds, limit=" runtlim) |
︙ | ︙ |
Modified tests/unittests/all-rmt.scm from [a8b587a3b0] to [47417391a0].
︙ | ︙ | |||
24 25 26 27 28 29 30 | (test #f #t (vector? (rmt:get-connection-info toppath))) ;; TODO: push areapath down. (test #f #t (string? (server:check-if-running "."))) ;; DEF (test #f #f (rmt:send-receive-no-auto-client-setup *runremote* 'get-keys #f '())) ;; DEF (rmt:kill-server run-id) ;; DEF (rmt:start-server run-id) (test #f '(#t "successful login")(rmt:login #f)) | < < < < < < | 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | (test #f #t (vector? (rmt:get-connection-info toppath))) ;; TODO: push areapath down. (test #f #t (string? (server:check-if-running "."))) ;; DEF (test #f #f (rmt:send-receive-no-auto-client-setup *runremote* 'get-keys #f '())) ;; DEF (rmt:kill-server run-id) ;; DEF (rmt:start-server run-id) (test #f '(#t "successful login")(rmt:login #f)) ;; DEF (rmt:login-no-auto-client-setup connection-info) (test #f #t (pair? (rmt:get-latest-host-load (get-host-name)))) ;; get-latest-host-load does a lookup in the db, it won't return a useful value unless ;; a test ran recently on host (test-batch rmt:get-latest-host-load "rmt:get-latest-host-load" |
︙ | ︙ | |||
110 111 112 113 114 115 116 | (tpt "%/%")) ;; target patt (test-batch rmt:get-runs-by-patt "rmt:get-runs-by-patt" (list (list "t=0" #t keys rnp tpt #f #f #f 0) (list "t=current" #f keys rnp tpt #f #f #f (+ 100 (current-seconds))) ;; should be no records from the future ) post-proc: (lambda (res) | | | 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 | (tpt "%/%")) ;; target patt (test-batch rmt:get-runs-by-patt "rmt:get-runs-by-patt" (list (list "t=0" #t keys rnp tpt #f #f #f 0) (list "t=current" #f keys rnp tpt #f #f #f (+ 100 (current-seconds))) ;; should be no records from the future ) post-proc: (lambda (res) ;; (print "rmt:get-runs-by-patt returned: " res) (and (vector? res) (let ((rows (vector-ref res 1))) (> (length rows) 0)))))) ;; (rmt:find-and-mark-incomplete run-id ovr-deadtime) ;; (rmt:get-main-run-stats run-id) ;; (rmt:get-var varname) |
︙ | ︙ |