lei.chen/myxce
1
0
Fork 1
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix(raft-ha-core-safety): close 5.10a — WAL segment rotation + endpoint UAF blocking real snapshot catch-up

Browse source 
Chain of fixes needed to get --force-snapshot soak to a real live PASS
(raft-ha-core-safety task 5.10a / ha-entity-snapshots task 7.1):

1. WAL segment rotation was gated on a hardcoded 64MB compile-time
   constant; trim_wal_segments() never deletes the active (currently-open)
   segment. At soak job counts a segment never rotated, so nothing was
   ever physically trimmable and ha_evaluate_repl_action always chose
   STREAM_WAL over SEND_SNAPSHOT for an empty-rejoining node. Fixed by
   threading a wal_segment_bytes override through persist_config_t
   (mirrors the --max-wal-records pattern: config.h default, persist.h
   setter, persist.c 3 usage sites + minimum-floor enforcement, 6-point
   CLI wiring through main.c/myconf.h/myapp.c/cluster.c) plus a new
   --wal-segment-bytes CLI flag. run-ha-soak.sh's --force-snapshot mode
   now sets WAL_SEGMENT_BYTES=4096 to force real rotation at soak scale.

2. This unmasked a genuine heap-use-after-free in INVOKE_MSG_HANDLER
   (include/netio.h): the macro calls ep->msg_handler(ep) synchronously
   (not via the vtask dispatcher, so the bug-1298 vtask ref/unref pairing
   doesn't cover it) then writes ep->fsm_in.pkt = NULL. A stale vote-reply
   handler chain (raft_on_vote_reply -> raft_cb_become_follower ->
   endpoint_unref) could drop the endpoint's last ref and free it inside
   the handler call, so the trailing write hit freed memory. This is the
   first-ever live exercise of a node receiving a snapshot transfer while
   mid-election. Fixed by holding a ref across the synchronous call, same
   pattern task 5.8 established for the async path.

With both fixes: --force-snapshot soak now PASSES live (committed
divergence=0, job_loss=0, confirmed real "HA: snapshot" install on the
restarted node) — first genuine live proof of this scenario. 3-round
--chaotic regression clean. Full 217-test controller suite green
before/after every fix in the chain.

Builds on the delaymgr starvation fix (b9ad018) and walcompare
checkpoint-floor fix (f3c86a6) from earlier in this investigation.
This commit is contained in:
Rongsong Shen 2026-07-02 04:11:20 +00:00
parent b9ad01878c
commit 1439e2655e
11 changed files with 115 additions and 40 deletions
Download patch file Download diff file Expand all files Collapse all files

8
src/controller/myapp.c
View file

@ -177,6 +177,14 @@ int myapp_run(myconf_t *myconf)
* deterministic record count instead of the 10000-record default. */
persist_config_set_max_wal_records(&pcfg, myconf->max_wal_records);
/* --wal-segment-bytes physical-segment-size override (0 = keep the
* persist default). raft-ha-core-safety task 5.10a / ha-entity-
* snapshots task 7.1: a segment is only trimmable once it rotates
* out, so forcing a real snapshot-triggered catch-up in the soak
* needs a small deterministic segment size instead of the 64 MB
* production default. */
persist_config_set_wal_segment_bytes(&pcfg, myconf->wal_segment_bytes);
persist_set_workdir(&c.persist, workdir, name);
rc = persist_init(&c.persist, &pcfg, &c);
if (rc != 0) {