-
Type: Bug
-
Resolution: Duplicate
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
StorEng - Refinement Pipeline
While trying to reproduce WT-10717 on kodkod-aws I hit an assertion failure in reconciliation:
WT_ASSERT_ALWAYS(session, tombstone->txnid != WT_TXN_ABORTED && __wt_txn_upd_visible_all(session, tombstone) && upd_select->upd == NULL, "Tombstone has been aborted, but the previously tombstoned update is not on " "the update chain");
The stack is:
(gdb) bt #0 0x00007ffbdadf0e87 in raise () from /lib/x86_64-linux-gnu/libc.so.6 #1 0x00007ffbdadf27f1 in abort () from /lib/x86_64-linux-gnu/libc.so.6 #2 0x00007ffbdba66724 in __wt_abort (session=0x14df680) at ../src/os_common/os_abort.c:30 #3 0x00007ffbdbab2c36 in __rec_fill_tw_from_upd_select (session=0x14df680, page=0x7ffb9c0b02b0, vpack=0x7ffba7ff5ff8, upd_select=0x7ffba7ff5f48) at ../src/reconcile/rec_visibility.c:768 #4 0x00007ffbdbab0be3 in __wt_rec_upd_select (session=0x14df680, r=0x7ffb90261020, ins=0x0, rip=0x7ffb9c0b1688, vpack=0x7ffba7ff5ff8, upd_select=0x7ffba7ff5f48) at ../src/reconcile/rec_visibility.c:860 #5 0x00007ffbdbaa0cee in __wt_rec_row_leaf (session=0x14df680, r=0x7ffb90261020, pageref=0x7ffb680b2c20, salvage=0x0) at ../src/reconcile/rec_row.c:768 #6 0x00007ffbdbab8c34 in __reconcile (session=0x14df680, ref=0x7ffb680b2c20, salvage=0x0, flags=355, page_lockedp=0x7ffba7ff640e) at ../src/reconcile/rec_write.c:278 #7 0x00007ffbdbab7cd5 in __wt_reconcile (session=0x14df680, ref=0x7ffb680b2c20, salvage=0x0, flags=355) at ../src/reconcile/rec_write.c:99 #8 0x00007ffbdb9fd20b in __evict_reconcile (session=0x14df680, ref=0x7ffb680b2c20, evict_flags=4) at ../src/evict/evict_page.c:822 #9 0x00007ffbdb9fb83b in __wt_evict (session=0x14df680, ref=0x7ffb680b2c20, previous_state=3 '\003', flags=4) at ../src/evict/evict_page.c:182 #10 0x00007ffbdb9faedd in __wt_page_release_evict (session=0x14df680, ref=0x7ffb680b2c20, flags=0) at ../src/evict/evict_page.c:91 #11 0x00007ffbdb84e9bf in __cursor_reset (cbt=0x7ffb9020eaf0) at ../src/include/cursor_inline.h:291 #12 0x00007ffbdb84e7a7 in __wt_btcur_reset (cbt=0x7ffb9020eaf0) at ../src/btree/bt_cursor.c:619 #13 0x00007ffbdb9709ce in __curfile_reset (cursor=0x7ffb9020eaf0) at ../src/cursor/cur_file.c:280 #14 0x000000000041a267 in table_op (tinfo=0x1ed74a0, intxn=true, iso_level=ISOLATION_SNAPSHOT, op=MODIFY) at ../test/format/ops.c:909 #15 0x00000000004167ca in ops (arg=0x1ed74a0) at ../test/format/ops.c:1250 #16 0x00007ffbdb5b66db in start_thread () from /lib/x86_64-linux-gnu/libpthread.so.0
Since this was running a reproducer on my machine I have a core dump and full format operations tracing turned on.
- duplicates
-
WT-10522 Fix a bug that leads to an unexpected aborted tombstone on update chain
- Closed
- is duplicated by
-
WT-10773 failed: format-stress-pull-request-test on ubuntu2004-msan [wiredtiger @ e95b9815]
- Closed
- related to
-
WT-10522 Fix a bug that leads to an unexpected aborted tombstone on update chain
- Closed