Overview
Comment: | wip |
---|---|
Downloads: | Tarball | ZIP archive | SQL archive |
Timelines: | family | ancestors | descendants | both | v1.6584-ck5 |
Files: | files | file ages | folders |
SHA1: |
8357e84ed0fb81723ba9921c75cccedd |
User & Date: | matt on 2021-04-21 00:01:27 |
Other Links: | branch diff | manifest | tags |
Context
2021-04-21
| ||
06:15 | wip: db:lock-create-open, db:open-db and db:get-db check-in: 1863efefdc user: matt tags: v1.6584-ck5 | |
00:01 | wip check-in: 8357e84ed0 user: matt tags: v1.6584-ck5 | |
2021-04-20
| ||
08:56 | added checks on starting watchdog that it is a not started thread check-in: 31cd96b3b8 user: matt tags: v1.6584-ck5 | |
Changes
Modified commonmod.scm from [f2f99ff8c5] to [1116143dd3].
︙ | ︙ | |||
3282 3283 3284 3285 3286 3287 3288 | (debug:print-info 0 "failed to write out packet to " pktsdir ", exn=" exn) ;; don't care if this failed for now but MUST FIX - BUG!! (if (not (file-exists? pktsdir)) (create-directory pktsdir #t)) (with-output-to-file (conc pktsdir "/" uuid ".pkt") (lambda () (print pkt))))))))) | | > > > > > > > > > > > > > > > | 3282 3283 3284 3285 3286 3287 3288 3289 3290 3291 3292 3293 3294 3295 3296 3297 3298 3299 3300 3301 3302 3303 3304 3305 3306 3307 3308 3309 3310 3311 | (debug:print-info 0 "failed to write out packet to " pktsdir ", exn=" exn) ;; don't care if this failed for now but MUST FIX - BUG!! (if (not (file-exists? pktsdir)) (create-directory pktsdir #t)) (with-output-to-file (conc pktsdir "/" uuid ".pkt") (lambda () (print pkt))))))))) ;;====================================================================== ;; use-lt is use linktree "lt" link to find pkts dir (define (common:minimal-save-pkt pktalist pktsdir) (let-values (((uuid pkt) (alist->pkt pktalist common:pkts-spec))) (handle-exceptions exn (debug:print-info 0 "failed to write out packet to " pktsdir ", exn=" exn) ;; don't care if this failed for now but MUST FIX - BUG!! (if (not (file-exists? pktsdir)) (create-directory pktsdir #t)) (with-output-to-file (conc pktsdir "/" uuid ".pkt") (lambda () (print pkt)))))) ;; common:with-queue-db was here ;; common:load-pkts-to-db was here (define (common:get-pkt-alists pkts) (map (lambda (x) (alist-ref 'apkt x)) ;; 'pkta pulls out the alist from the read pkt pkts)) |
︙ | ︙ |
Modified http-transportmod.scm from [c8ea6cc265] to [c5ac09a5b9].
︙ | ︙ | |||
23 24 25 26 27 28 29 30 31 32 33 34 35 36 | (declare (uses commonmod)) (declare (uses configfmod)) (declare (uses mtargs)) (declare (uses mtver)) (declare (uses dbmod)) (declare (uses stml2)) (declare (uses portloggermod)) (module http-transportmod * (import scheme (prefix sqlite3 sqlite3:) chicken.base | > | 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 | (declare (uses commonmod)) (declare (uses configfmod)) (declare (uses mtargs)) (declare (uses mtver)) (declare (uses dbmod)) (declare (uses stml2)) (declare (uses portloggermod)) (declare (uses pkts)) (module http-transportmod * (import scheme (prefix sqlite3 sqlite3:) chicken.base |
︙ | ︙ | |||
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 | commonmod configfmod debugprint mtver dbmod stml2 portloggermod ) ;; (require-extension (srfi 18) extras tcp s11n) ;; ;; ;; (use srfi-1 posix regex regex-case srfi-69 hostinfo md5 message-digest posix-extras) ;; ;; (use spiffy uri-common intarweb http-client spiffy-request-vars intarweb spiffy-directory-listing) ;; ;; Configurations for server (tcp-buffer-size 2048) (max-connections 2048) | > < < < < < < < < < < < < < < < < | | > | 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 | commonmod configfmod debugprint mtver dbmod stml2 portloggermod pkts ) ;; (require-extension (srfi 18) extras tcp s11n) ;; ;; ;; (use srfi-1 posix regex regex-case srfi-69 hostinfo md5 message-digest posix-extras) ;; ;; (use spiffy uri-common intarweb http-client spiffy-request-vars intarweb spiffy-directory-listing) ;; ;; Configurations for server (tcp-buffer-size 2048) (max-connections 2048) (define (http-transport:make-server-url hostport) (if (not hostport) #f (conc "http://" (car hostport) ":" (cadr hostport)))) ;;====================================================================== ;; S E R V E R ;; ====================================================================== ;; NOTE: http-transport:launch is the entry point ;; -> http-transport:run ;; -> http-transport:try-start-server -> http-transport:try-start-server (until success) (define (http-get-function fnkey) (hash-table-ref/default *http-functions* fnkey (lambda () "nothing here yet"))) (define (http-transport:run hostn) ;; Configurations for server (tcp-buffer-size 2048) |
︙ | ︙ | |||
199 200 201 202 203 204 205 | (send-response body: ((http-get-function 'http-transport:html-test-log) $) headers: '((content-type text/HTML)))) ((equal? (uri-path (request-uri (current-request))) '(/ "dashboard")) (send-response body: ((http-get-function 'http-transport:html-dboard) $) headers: '((content-type text/HTML)))) (else (continue)))))))) | | | 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 | (send-response body: ((http-get-function 'http-transport:html-test-log) $) headers: '((content-type text/HTML)))) ((equal? (uri-path (request-uri (current-request))) '(/ "dashboard")) (send-response body: ((http-get-function 'http-transport:html-dboard) $) headers: '((content-type text/HTML)))) (else (continue)))))))) #;(handle-exceptions exn (debug:print 0 *default-log-port* "Failed to create file " start-file ", exn=" exn) (with-output-to-file start-file (lambda ()(print (current-process-id))))) (http-transport:try-start-server ipaddrstr start-port))) ;; This is recursively run by http-transport:run until sucessful ;; |
︙ | ︙ | |||
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 | ;; (define (http-transport:client-connect iface port server-id) (let* ((api-url (conc "http://" iface ":" port "/api")) (api-uri (uri-reference (conc "http://" iface ":" port "/api"))) (api-req (make-request method: 'POST uri: api-uri)) (server-dat (vector iface port api-uri api-url api-req (current-seconds) server-id))) server-dat)) ;; run http-transport:keep-running in a parallel thread to monitor that the db is being ;; used and to shutdown after sometime if it is not. ;; (define (http-transport:keep-running) ;; if none running or if > 20 seconds since ;; server last used then start shutdown ;; This thread waits for the server to come alive (debug:print-info 0 *default-log-port* "Starting the sync-back, keep alive thread in server") (let* ((sdat #f) (tmp-area (common:get-db-tmp-area)) (started-file (conc tmp-area "/.server-started")) (server-start-time (current-seconds)) (server-info (let loop ((start-time (current-seconds)) (changed #t) (last-sdat "not this")) (begin ;; let ((sdat #f)) (thread-sleep! 0.01) (debug:print-info 0 *default-log-port* "Waiting for server alive signature") (mutex-lock! *heartbeat-mutex*) (set! sdat *server-info*) (mutex-unlock! *heartbeat-mutex*) (if (and sdat (not changed) (> (- (current-seconds) start-time) 2)) (begin | > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > | | | | | | > > > > > > > | | | | 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 | ;; (define (http-transport:client-connect iface port server-id) (let* ((api-url (conc "http://" iface ":" port "/api")) (api-uri (uri-reference (conc "http://" iface ":" port "/api"))) (api-req (make-request method: 'POST uri: api-uri)) (server-dat (vector iface port api-uri api-url api-req (current-seconds) server-id))) server-dat)) ;;====================================================================== ;; NEW SERVER METHOD ;;====================================================================== (define *srvpktspec* `((server (host . h) (port . p) (servkey . k) (pid . i) (ipaddr . a) (dbpath . d)))) (define (register-server pkts-dir pkt-spec host port servkey ipaddr dbpath) (let* ((pkt-dat `((host . ,host) (port . ,port) (servkey . ,servkey) (pid . ,(current-process-id)) (ipaddr . ,ipaddr) (dbpath . ,dbpath)))) (write-alist->pkt pkts-dir pkt-dat pktspec: pkt-spec ptype: 'server))) ;; ya, fake it for now ;; (define (register-server-in-db db-file) #t) ;; load up the db into inmem ;; (define (load-up-database db-file) (let* ((db (db:open-db db-file))) db)) (define (get-pkts-dir) (assert *toppath* "ERROR: get-pkts-dir called without *toppath* set. Exiting.") (let* ((pdir (conc *toppath* "/.meta/srvpkts"))) (if (file-exists? pdir) pdir (begin (create-directory pdir #t) pdir)))) ;; given a pkts dir read ;; (define (get-all-server-pkts pktsdir-in pktspec) (let* ((pktsdir (if (file-exists? pktsdir-in) pktsdir-in (begin (create-directory pktsdir-in #t) pktsdir-in))) (all-pkt-files (glob (conc pktsdir "/*.pkt")))) (map (lambda (pkt-file) (read-pkt->alist pkt-file pktspec: pktspec)) all-pkt-files))) (define (server-address srv-pkt) (conc (alist-ref 'host srv-pkt) ":" (alist-ref 'port srv-pkt))) (define (server-ready? server-address) ;; ping the server and ask it ;; if it ready #f) ;; from the pkts return servers associated with dbpath ;; NOTE: Only one can be alive - have to check on each ;; in the list of pkts returned ;; (define (get-viable-servers serv-pkts dbpath) (let loop ((tail serv-pkts) (res '())) (if (null? tail) res ;; NOTE: sort by age so oldest is considered first (let* ((spkt (car tail))) (loop (cdr tail) (if (equal? dbpath (alist-ref 'dbpath spkt)) (cons spkt res) res)))))) ;; from viable servers get one that is alive and ready ;; (define (get-the-server serv-pkts dbpath) (let loop ((tail serv-pkts)) (if (null? tail) #f (let* ((spkt (car tail)) (addr (server-address spkt))) (if (server-ready? addr) spkt (loop (cdr tail))))))) ;; am I the "first" in line server? I.e. my D card is smallest ;; use Z card as tie breaker ;; (define (get-best-candidate serv-pkts dbpath) (if (null? serv-pkts) #f (let loop ((tail serv-pkts) (best (car serv-pkts))) (if (null? tail) best (let* ((candidate (car tail)) (candidate-bd (string->number (alist-ref 'D candidate))) (best-bd (string->number (alist-ref 'D best))) ;; bigger number is younger (candidate-z (alist-ref 'Z candidate)) (best-z (alist-ref 'Z best)) (new-best (cond ((> best-bd candidate-bd) ;; best is younger than candidate candidate) ((< best-bd candidate-bd) ;; candidate is younger than best best) (else (if (string>=? best-z candidate-z) best candidate))))) ;; use Z card as tie breaker (if (null? tail) new-best (loop (cdr tail) new-best))))))) ;;====================================================================== ;; END NEW SERVER METHOD ;;====================================================================== ;; run http-transport:keep-running in a parallel thread to monitor that the db is being ;; used and to shutdown after sometime if it is not. ;; (define (http-transport:keep-running) ;; if none running or if > 20 seconds since ;; server last used then start shutdown ;; This thread waits for the server to come alive (debug:print-info 0 *default-log-port* "Starting the sync-back, keep alive thread in server") (let* ((sdat #f) (tmp-area (common:get-db-tmp-area)) (started-file (conc tmp-area "/.server-started")) (server-start-time (current-seconds)) (pkts-dir (get-pkts-dir)) (server-key (server:mk-signature)) (db-file (conc *toppath* "/.db/" (or (args:get-arg "-db") "main.db"))) (server-info (let loop ((start-time (current-seconds)) (changed #t) (last-sdat "not this")) (begin ;; let ((sdat #f)) (thread-sleep! 0.01) (debug:print-info 0 *default-log-port* "Waiting for server alive signature") (mutex-lock! *heartbeat-mutex*) (set! sdat *server-info*) (mutex-unlock! *heartbeat-mutex*) (if (and sdat (not changed) (> (- (current-seconds) start-time) 2)) (begin (debug:print-info 0 *default-log-port* "Received server alive signature, now attempting to lock in server") ;; create a server pkt in *toppath*/.meta/srvpkts (register-server pkts-dir *srvpktspec* (get-host-name) (cadr sdat) server-key (car sdat) db-file) ;; now read pkts and see if we are a contender (let* ((all-pkts (get-all-server-pkts pkts-dir *srvpktspec*)) (viables (get-viable-servers all-pkts db-file)) (best-srv (get-best-candidate viables db-file)) (best-srv-key (if best-srv (alist-ref 'servkey best-srv) #f))) ;; am I the best-srv, compare server-keys to know (if (and (equal? best-srv-key server-key) (register-server-in-db db-file)) (load-up-database db-file) ;; ready to go! (bdat-time-to-exit-set! *bdat* #t)) ;; nope, we are not needed, exit when can do sdat)) (begin (debug:print-info 0 *default-log-port* "Still waiting, last-sdat=" last-sdat) (sleep 4) (if (> (- (current-seconds) start-time) 120) ;; been waiting for two minutes (begin (debug:print-error 0 *default-log-port* "transport appears to have died, exiting server") #;(common:save-pkt `((action . died) (T . server) (pid . ,(current-process-id)) (ipaddr . ,(car sdat)) (port . ,(cadr sdat)) (msg . "Transport died?")) *configdat* #t) (exit)) |
︙ | ︙ | |||
524 525 526 527 528 529 530 | (if (not server-going) ;; *dbstruct-db* (let ((watchdog (bdat-watchdog *bdat*))) (debug:print 0 *default-log-port* "SERVER: dbprep") (set! *dbstruct-db* (db:setup #t)) ;; run-id)) (set! server-going #t) (debug:print 0 *default-log-port* "SERVER: running, megatest version: " (common:get-full-version)) ;; NOTE: the server is NOT yet marked as running in the log. We do that in the keep-running routine. (if watchdog | | > > | | 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 | (if (not server-going) ;; *dbstruct-db* (let ((watchdog (bdat-watchdog *bdat*))) (debug:print 0 *default-log-port* "SERVER: dbprep") (set! *dbstruct-db* (db:setup #t)) ;; run-id)) (set! server-going #t) (debug:print 0 *default-log-port* "SERVER: running, megatest version: " (common:get-full-version)) ;; NOTE: the server is NOT yet marked as running in the log. We do that in the keep-running routine. (if watchdog (if (not (member (thread-state watchdog) '(ready running blocked sleeping dead))) (begin (debug:print-info 0 "Starting watchdog thread (in state "(thread-state watchdog)")") (thread-start! watchdog))) (debug:print 0 *default-log-port* "ERROR: *watchdog* not setup, cannot start it.")))) ;; when things go wrong we don't want to be doing the various queries too often ;; so we strive to run this stuff only every four seconds or so. (let* ((sync-time (- (current-milliseconds) start-time)) (rem-time (quotient (- 4000 sync-time) 1000))) (if (and (<= rem-time 4) |
︙ | ︙ | |||
583 584 585 586 587 588 589 | (current-seconds))) (if (common:low-noise-print 120 "server continuing") (debug:print-info 0 *default-log-port* "Server continuing, seconds since last db access: " (- (current-seconds) last-access)) (let ((curr-time (current-seconds))) (handle-exceptions exn (debug:print 0 *default-log-port* "ERROR: Failed to change timestamp on log file " server-log-file ". Are you out of space on that disk? exn=" exn) | | | 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 | (current-seconds))) (if (common:low-noise-print 120 "server continuing") (debug:print-info 0 *default-log-port* "Server continuing, seconds since last db access: " (- (current-seconds) last-access)) (let ((curr-time (current-seconds))) (handle-exceptions exn (debug:print 0 *default-log-port* "ERROR: Failed to change timestamp on log file " server-log-file ". Are you out of space on that disk? exn=" exn) (if (and server-log-file (not *server-overloaded*)) (set-file-times! server-log-file curr-time curr-time))))) (loop 0 server-state bad-sync-count (current-milliseconds))) (else (debug:print-info 0 *default-log-port* "Server timed out. seconds since last db access: " (- (current-seconds) last-access)) (http-transport:server-shutdown port))))))) (define (http-transport:server-shutdown port) |
︙ | ︙ | |||
624 625 626 627 628 629 630 631 632 633 634 635 636 | (db:print-current-query-stats) (common:save-pkt `((action . exit) (T . server) (pid . ,(current-process-id))) *configdat* #t) (debug:print-info 0 *default-log-port* "Server shutdown complete. Exiting") (exit))) ;; all routes though here end in exit ... ;; ;; start_server? ;; (define (http-transport:launch) | > > > < < < < < < < < < < < < | | | | | 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 | (db:print-current-query-stats) (common:save-pkt `((action . exit) (T . server) (pid . ,(current-process-id))) *configdat* #t) (debug:print-info 0 *default-log-port* "Server shutdown complete. Exiting") (exit))) ;; Call this to start the actual server ;; ;; all routes though here end in exit ... ;; ;; start_server? ;; (define (http-transport:launch) (let* ((tmp-area (common:get-db-tmp-area)) (server-start (conc tmp-area "/.server-start")) (server-started (conc tmp-area "/.server-started")) (start-time (common:lazy-modification-time server-start)) (started-time (common:lazy-modification-time server-started)) (server-starting (< start-time started-time)) ;; if start-time is less than started-time then a server is still starting (start-time-old (> (- (current-seconds) start-time) 5)) (cleanup-proc (lambda (msg) (let* ((serv-fname (conc "server-" (current-process-id) "-" (get-host-name) ".log")) (full-serv-fname (conc *toppath* "/logs/" serv-fname)) (new-serv-fname (conc *toppath* "/logs/" "defunct-" serv-fname))) (debug:print 0 *default-log-port* msg) (if (common:file-exists? full-serv-fname) (system (conc "sleep 1;mv -f " full-serv-fname " " new-serv-fname)) (debug:print 0 *default-log-port* "INFO: cannot move " full-serv-fname " to " new-serv-fname)) (exit))))) (common:save-pkt `((action . start) (T . server) (pid . ,(current-process-id))) *configdat* #t) (let* ((th2 (make-thread (lambda () (debug:print-info 0 *default-log-port* "Server run thread started") (http-transport:run (if (args:get-arg "-server") (args:get-arg "-server") "-") )) "Server run")) |
︙ | ︙ | |||
698 699 700 701 702 703 704 | ;; (debug:print 0 *default-log-port* " Done.") ;; (exit 4)) ;; "exit on ^C timer"))) ;; (thread-start! th2) ;; (thread-start! th1) ;; (thread-join! th2)))) | < < < < < < < < < < < < | 817 818 819 820 821 822 823 824 825 826 827 828 829 830 | ;; (debug:print 0 *default-log-port* " Done.") ;; (exit 4)) ;; "exit on ^C timer"))) ;; (thread-start! th2) ;; (thread-start! th1) ;; (thread-join! th2)))) ;; Generate a unique signature for this server (define (server:mk-signature) (message-digest-string (md5-primitive) (with-output-to-string (lambda () (write (list (current-directory) |
︙ | ︙ |
Modified servermod.scm from [d4e6207e5f] to [4b44aabd8d].
︙ | ︙ | |||
55 56 57 58 59 60 61 | debugprint configfmod http-transportmod pkts ) | < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < < | 55 56 57 58 59 60 61 62 63 64 65 66 67 68 | debugprint configfmod http-transportmod pkts ) (define (server:make-server-url hostport) (if (not hostport) #f (conc "http://" (car hostport) ":" (cadr hostport)))) ;;====================================================================== ;; logic for getting homehost. Returns (host . at-home) |
︙ | ︙ |