Commit graph

Select branches

Hide pull requests

build/de-nix

chatbot

feature/license-resources

feature/scheduler-advisor

feature/x11-forwarding

main

merge-sf-20260831

merge-sf-20260903

working

Mono Color

• b38ef6de63 fix(testsuite): resize crr_build_config caller arrays -- bug-1977 feature/license-resources lei.chen 2026-10-28 11:58:09 +00:00
• 30c51ea147 external-license-resources 3.3b: elim load-index collector in xcecd lei.chen 2026-10-28 11:17:36 +00:00
• 700f3d4e1a scheduler tests: UT cover for E2 Phase 3 increments (adoption shared allocs + per-queue instances) lei.chen 2026-10-28 05:57:15 +00:00
• 13a28a387a wolf: log bug-1976 fixed (controller dispatch commit dropped allocs_shared) lei.chen 2026-10-28 05:40:03 +00:00
• 76ddd75fd1 E2 Phase 3: per-project/per-queue license quotas via limit vres + controller shared-alloc migration fix lei.chen 2026-10-28 05:36:43 +00:00
• 857f987f3c resources: XCE_RESF_BUILTIN 从未被置位 — resources.yaml 加 builtin: true 声明式键 lei.chen 2026-10-28 01:46:21 +00:00
• 805288d747 E2 Phase 2: external-license-resources 运维面 — stale 标记、last-updated 展示、xceadmin extres status lei.chen 2026-10-27 14:00:33 +00:00
• 3eb1b9716d wolf: mark bug-1973/1974/1975 fixed (commit 2fcacac2) lei.chen 2026-10-27 03:45:49 +00:00
• 2fcacac2cb E2 Phase 1: external-license-resources — xceesd daemon, extres role, EXTERNAL admission lei.chen 2026-10-27 03:45:28 +00:00
• ba42fdd82e wolf: mark bug-1972 fixed (commit f780a166) lei.chen 2026-10-26 14:03:05 +00:00
• f780a1660a A-batch admin-plane fixes: rusage unit name, XCE_JOBID identity, select WARN, CLI env, docs lei.chen 2026-10-26 14:02:53 +00:00
• 68eabd8a4a docs(spec): external-license-resources design proposal (E2) working lei.chen 2026-10-26 08:02:14 +00:00
• 8ae2b5dc47 advisor: probe LLM liveness every round with down/up transition logs lei.chen 2026-10-26 03:51:43 +00:00
• effe50a6ba cluster info: fill xce id cluster_name/status from live ctrlstatus lei.chen 2026-10-25 02:58:14 +00:00
• d028c8c315 admin-plane breakpoint 3: done-job live GC build/de-nix lei.chen 2026-10-23 15:47:10 +00:00
• fc33a5da16 admin-plane breakpoint 2: YAML resourcemap wiring + queue payload deep-copy lei.chen 2026-10-23 14:29:29 +00:00
• 563f8e6890 admin-plane breakpoint 1: config activation reaches live scheduler + queues hot-apply lei.chen 2026-10-23 09:24:05 +00:00
• d16954ad10 scheduler audit P1: picker hook validation + rejection WARNs + sim bootstrap by name lei.chen 2026-10-22 14:05:38 +00:00
• c194973d42 limit audit: correct PEND reason semantics + percent-map + project normalization lei.chen 2026-10-22 13:11:01 +00:00
• 993daa8f95 bug-1971: fix_commits field via buglog.py status lei.chen 2026-10-22 08:08:44 +00:00
• d8f738b08a bug-1971: mark fixed, record fix commit 68c3ccd7 lei.chen 2026-10-22 08:08:15 +00:00
• 68c3ccd773 bug-1971: preserve flaky test-failure evidence across xce-build.sh runs lei.chen 2026-10-22 08:08:01 +00:00
• e8cb6d5743 worker: handle pre-exec failure without dying; persist: skip snapshot queues not in config lei.chen 2026-10-22 07:17:01 +00:00
• 5b9d727086 Fix two flaky test failures: commit sweep timeout + array stats identity lei.chen 2026-10-22 02:51:42 +00:00
• 84f8d6b76a build: split compile -l ceiling from discounted test -j/-p caps lei.chen 2026-10-21 15:41:25 +00:00
• 28ccf8d35c build: recalibrate asan per-job memory budget 1536MB -> 512MB lei.chen 2026-10-21 15:38:46 +00:00
• e723c9c234 build: eliminate all remaining compile warnings lei.chen 2026-10-21 15:38:42 +00:00
• d4957bd290 build(scheduler): compile scheduler sources once -- shared sched_srcs_lib lei.chen 2026-10-21 14:07:14 +00:00
• 678163adb8 build: compile controller.capnp.c once -- libxce_dep passes only the header lei.chen 2026-10-21 13:55:33 +00:00
• a905baf013 test(scheduler): aggregate 28 white-box tests into one sched_unit_test lei.chen 2026-10-21 10:39:55 +00:00
• d167923632 build: eliminate all unused-variable/function warnings in our code lei.chen 2026-10-21 09:28:34 +00:00
• f1801fbef9 build: fix 7 risky warnings + slim debug profile to -g1 lei.chen 2026-10-21 08:32:41 +00:00
• edb63e6095 docs(readme): replace nix build instructions with the de-nix meson flow lei.chen 2026-10-21 03:44:43 +00:00
• 017f41a010 build(packaging): bundle ttyd prebuilt; record license-compliance notes (DEPS.md) lei.chen 2026-10-21 03:33:10 +00:00
• 6bf26cbab3 fix(controller): post-purge job facts must be benign skips, not fence-engaging failures (bug-1970) lei.chen 2026-10-21 02:21:22 +00:00
• 50cf424146 build: remove nix dependency layer -- apt system libs + meson wrap subprojects lei.chen 2026-10-20 18:04:16 +00:00
• 1e5c85555a fix(worker): MPI remote-start path -- union alias UB, taskrun leak, xrun overflow feature/scheduler-advisor lei.chen 2026-10-20 06:36:45 +00:00
• a9c8044165 apiserver: derive array stats from live job store; fix CLI stats decode lei.chen 2026-10-20 05:22:20 +00:00
• 9dbb75adb3 apiserver: persist API keys across restarts via state file lei.chen 2026-10-20 03:12:13 +00:00
• e7fbedd047 docs: demo handbook - compare subcommand (2b) + deepseek model guidance lei.chen 2026-10-20 02:46:27 +00:00
• d4bd504ff5 advisor: add compare subcommand for echo-vs-LLM measurement lei.chen 2026-10-20 02:45:55 +00:00
• 2808742288 docs(advisor): stable install + restart script replaces /tmp nohup runbook lei.chen 2026-10-20 02:26:17 +00:00
• 42804166d8 docs(advisor): demo case 3 now supports the repeat-outage cycle lei.chen 2026-10-20 02:22:22 +00:00
• e97f6b1687 fix(advisor): disarm dedup key on full resolution so recurrences report lei.chen 2026-10-20 02:21:42 +00:00
• 321544f7be docs(advisor): manual demo case handbook for t1/t2/t3 cluster lei.chen 2026-10-20 01:57:14 +00:00
• 1691e49a05 feat(advisor): P1 outlets — findings push to SSE + MCP scheduler_advisor_findings lei.chen 2026-10-19 10:06:32 +00:00
• 3e3b6d7947 fix(apiserver): detail-only PEND reason promotion was unreachable on the C path lei.chen 2026-10-19 07:07:02 +00:00
• 4db1aa8aef fix(advisor): cluster-wide job view + change-triggered detection lei.chen 2026-10-19 07:06:53 +00:00
• 46bb23ea91 fix(schedulerclient): ListJobs sort + all-users filters (advisor P0 findings) lei.chen 2026-10-19 07:06:48 +00:00
• 2b57844087 docs(advisor): design doc with recorded-not-implemented Jev appendix lei.chen 2026-10-19 03:42:29 +00:00
• ab12f49256 feat(advisor): agent loop, decision log, runbook, run/once CLI lei.chen 2026-10-19 03:42:01 +00:00
• fbc81d61d1 feat(advisor): snapshot builder + deterministic detectors lei.chen 2026-10-19 03:33:54 +00:00
• 1ab96521b8 feat(advisor): Go module skeleton — config, provider contract, echo + openai providers lei.chen 2026-10-19 03:31:36 +00:00
• f94cc08680 fix(security): x11_cookie field-level RBAC; xcecli -X preflight before submit lei.chen 2026-10-18 07:19:26 +00:00
• f995123e6c Merge feature/x11-forwarding into working (X11 forwarding for interactive jobs) lei.chen 2026-10-18 02:37:18 +00:00
• 85a22271c1 Add X11 forwarding for interactive jobs (xcecli -I -X, ssh -X style) feature/x11-forwarding lei.chen 2026-10-17 02:16:06 +00:00
• 4f188310ba worker: reap unattached TTY jobs via ttyd watchdog with real-pid resolution lei.chen 2026-10-16 03:38:24 +00:00
• 6ce7bdbe11 xcecli: backoff poll cadence and suspend tolerance for -I URL wait lei.chen 2026-10-15 11:21:02 +00:00
• 3545aa0d77 apiserver: apply the controller's jobPidInfo start-report push lei.chen 2026-10-15 10:08:23 +00:00
• 66a1d1e680 attach jobs end with their command: relay watcher, once-only start banner, worker-side ttyd cleanup lei.chen 2026-10-15 08:32:21 +00:00
• 601f04dc00 feat(xcecli): bare -I maps to TTY attach interactive mode lei.chen 2026-10-15 03:40:32 +00:00
• 6d5281daf1 fix(worker): VNC start failure reports JOB_FAILED instead of killing the worker; fix generate_url snprintf overflow lei.chen 2026-10-15 03:40:29 +00:00
• a8c3e91b59 feat(interactive): attach TTY sessions through the apiserver websocket relay lei.chen 2026-10-14 13:24:09 +00:00
• 74121eb72e fix(apiserver): map pidinfo from job start into the job model (interactive_job_url) lei.chen 2026-10-14 09:20:27 +00:00
• f661987a60 fix(scheduler-test): unlink the reconcile stdout-capture scratch files (bug-924 tail) lei.chen 2026-10-14 07:27:45 +00:00
• 0c8607345e merge: sync Rongsong's scheduler-framework (c5b172c7..3442a57d, 781 commits) merge-sf-20260903 lei.chen 2026-10-14 05:17:03 +00:00
• 3442a57dd9
  
  docs(spec): project-config: a built-in default project always exists Rongsong Shen 2026-10-13 07:17:22 +00:00
• d769b746fa
  
  docs(spec): host state never changes pools or shares; the scheduler skips unusable hosts Rongsong Shen 2026-10-13 07:16:53 +00:00
• d1be84cd66
  
  docs(spec): project shares at a location; pool capacity, defaults, visibility Rongsong Shen 2026-10-13 07:09:03 +00:00
• e02310c02d
  
  docs(spec): project open items: jobs count, inner projects, aliases, accounting Rongsong Shen 2026-10-13 07:03:20 +00:00
• c6a938476b
  
  docs(spec): project-fairshare: proposals agreed by the user Rongsong Shen 2026-10-13 06:59:10 +00:00
• 16b9059116
  
  docs(spec): fairshare belongs to projects; add project-fairshare brief Rongsong Shen 2026-10-13 06:50:34 +00:00
• f3d4dba970
  
  docs(spec): projects own shares of their parent's resources; project-guarantee becomes project-borrowing Rongsong Shen 2026-10-13 06:38:39 +00:00
• 136e4a0132
  
  docs(spec): peer-authentication-hardening: decide SSH-style ed25519 keys with temporary keys and revocation Rongsong Shen 2026-10-13 04:32:20 +00:00
• 497148a732
  
  docs(roadmap): peer-authentication-hardening lists the ed25519 option Rongsong Shen 2026-10-13 04:28:37 +00:00
• 274eddfb55
  
  docs(spec): peer-authentication-hardening: option D, SSH-style ed25519 key pairs Rongsong Shen 2026-10-13 04:28:23 +00:00
• 20ebef6d28
  
  docs(spec): peer-authentication-hardening: secret type and key distribution deferred for later discussion Rongsong Shen 2026-10-13 04:26:27 +00:00
• a2bf160cc3
  
  docs(spec): peer-authentication-hardening brief; production mode accepts user requests only from the apiserver Rongsong Shen 2026-10-13 04:25:17 +00:00
• 33edbeeaf0
  
  docs(spec): project-config: apiserver is the only project authorization point Rongsong Shen 2026-10-13 04:17:04 +00:00
• 88ca983f43
  
  docs(spec): project-config: project:submit RBAC implied by membership; project admins Rongsong Shen 2026-10-13 04:13:35 +00:00
• 332f86eaa1
  
  docs(spec): resource control under the project tree; retire limits.yaml Rongsong Shen 2026-10-13 04:07:00 +00:00
• df486a01c0
  
  docs(spec): application-profile: Future Discussion section (design questions, deferred features, housekeeping) Rongsong Shen 2026-10-13 03:47:12 +00:00
• 7382e94d24
  
  docs(spec): presets have no access control of their own Rongsong Shen 2026-10-13 03:33:25 +00:00
• e48d36054d
  
  docs(spec): defer per-profile access control to app-access-control Rongsong Shen 2026-10-13 03:31:00 +00:00
• 863ba98ac8
  
  docs(spec): application-profile hooks and starters; kill-path requirement; proposals 2-5 Rongsong Shen 2026-10-13 03:28:10 +00:00
• f2406caaa7
  
  docs(spec): application-profile: job starter wraps the full job, task starter wraps every task startup Rongsong Shen 2026-10-13 03:15:43 +00:00
• a50c0e236b
  
  docs(spec): application-profile params and annotations; 0-valued rusage dropped; app-submit-hook backlog Rongsong Shen 2026-10-13 03:04:32 +00:00
• f658b0a4b0
  
  docs(spec): application-profile review fixes Rongsong Shen 2026-10-13 02:55:23 +00:00
• 1f9aaae050
  
  docs(spec): structured profile resreq; per: optional (default task); merge rules at a glance Rongsong Shen 2026-10-13 02:49:53 +00:00
• 8f3a811a43
  
  docs(spec): rusage merge: union, job value wins, profile per-type; compound rule Rongsong Shen 2026-10-12 14:49:10 +00:00
• a68d2fc965
  
  docs(spec): presets record themselves in exdata.meta; job-labels backlog item Rongsong Shen 2026-10-12 13:26:09 +00:00
• 5cfe64218c
  
  docs(roadmap): config-history-audit backlog item; profile versioning/audit deferred from application-profile Rongsong Shen 2026-10-12 13:12:39 +00:00
• 67002b2a25
  
  docs(spec): application-profile v1 attributes: job starter added, others ship with their features Rongsong Shen 2026-10-12 13:08:29 +00:00
• 83c6eb8024
  
  docs(spec): xcecli-presets: server preset directory, pull name collisions, /xce/v1 routes Rongsong Shen 2026-10-12 10:33:31 +00:00
• d1e03055e4
  
  docs(spec): split execution modes from application-profile into job-sandbox-execution Rongsong Shen 2026-10-12 10:20:15 +00:00
• 8900f8876b
  
  docs(spec): application-profile: profile aliases; moving pending jobs deferred to job-modify Rongsong Shen 2026-10-12 10:18:06 +00:00
• c0bd701f48
  
  docs(spec): xcecli-presets: one file per preset, same layout on client and server Rongsong Shen 2026-10-12 10:15:34 +00:00
• 1a5ddc2642
  
  docs(spec): xcecli-presets: environment capture by name (XCE_ and secret exclusions) Rongsong Shen 2026-10-12 10:14:50 +00:00
• 8f8be20885
  
  docs(spec): application-profile: clean env for the outer pre/post-exec, reserved XCE_ prefix Rongsong Shen 2026-10-12 10:12:15 +00:00
• d88b9cb31f
  
  docs(spec): application-profile: dispatch-window handling and profile content hash Rongsong Shen 2026-10-12 10:09:38 +00:00