-
Type: Task
-
Resolution: Done
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
keithbostic if you could take a first look at this and let me know if you believe it is my problem. The stack looks like it could be a split bug.
I am running the current end of the async branch (changeset 685f4771d9c5f40a34abbf5cf48214548089e5d1) and I am running the following wtperf configuration, using async operations, on the AWS HD box. I hit this error 100% of the time, almost instantaneously. I run to the large HD drives mounted at /mnt/fast.
conn_config="cache_size=1G,statistics=(fast,clear),statistics_log=(wait=10)" async_threads=10 table_config="lsm=(chunk_size=100MB,merge_threads=2),type=lsm" icount=10000000 report_interval=5 run_time=120 populate_threads=1
The command I use:
./wtperf -h /mnt/fast/sue/<dir> -O ~sue/async.wtperf
Here's the error:
[1397506074:437987][31598:00677febff7f0000], file:test-000001.lsm: the 13 and 15 keys on page at [write-check] are incorrectly sorted
Here's the stack:
(gdb) bt #0 __verify_dsk_row (session=0x6f1d20, addr=0x4cbb80 "[write-check]", dsk=0x7fffe0001000) at ../src/btree/bt_vrfy_dsk.c:373 WT-1 0x000000000045be25 in __wt_verify_dsk_image (session=0x6f1d20, addr=0x4cbb80 "[write-check]", dsk=0x7fffe0001000, size=3029) at ../src/btree/bt_vrfy_dsk.c:142 WT-2 0x000000000045be8f in __wt_verify_dsk (session=0x6f1d20, addr=0x4cbb80 "[write-check]", buf=0x7fffe0000958) at ../src/btree/bt_vrfy_dsk.c:158 WT-3 0x000000000044d35f in __wt_bt_write (session=0x6f1d20, buf=0x7fffe0000958, addr=0x7fffeb7f58a0 "\342\006\016\201\344Ь7\352\n", addr_sizep=0x7fffeb7f59a8, checkpoint=0, compressed=0) at ../src/btree/bt_io.c:176 WT-4 0x00000000004688b5 in __rec_split_write (session=0x6f1d20, r=0x7fffe0000940, bnd=0x7fffea7e6790, buf=0x7fffe0000958, final=0) at ../src/btree/rec_write.c:2531 WT-5 0x0000000000466ba2 in __rec_split (session=0x6f1d20, r=0x7fffe0000940) at ../src/btree/rec_write.c:1788 WT-6 0x000000000046c7d5 in __rec_row_leaf_insert (session=0x6f1d20, r=0x7fffe0000940, ins=0x7fffc468c410) at ../src/btree/rec_write.c:4284 WT-7 0x000000000046bad7 in __rec_row_leaf (session=0x6f1d20, r=0x7fffe0000940, page=0x7fffd8001b20, salvage=0x0) at ../src/btree/rec_write.c:3903 WT-8 0x00000000004647ea in __wt_rec_write (session=0x6f1d20, ref=0x7fffd8001ae0, salvage=0x0, flags=0) at ../src/btree/rec_write.c:405 WT-9 0x00000000004582d2 in __sync_file (session=0x6f1d20, syncop=1) at ../src/btree/bt_sync.c:56 WT-10 0x0000000000458934 in __wt_bt_cache_op (session=0x6f1d20, ckptbase=0x0, op=1) at ../src/btree/bt_sync.c:298 WT-11 0x000000000041f079 in __wt_lsm_checkpoint_worker (arg=0x75d0a0) at ../src/lsm/lsm_worker.c:343
- related to
-
WT-1 placeholder WT-1
- Closed
-
WT-2 What does metadata look like?
- Closed
-
WT-3 What file formats are required?
- Closed
-
WT-4 Flexible cursor traversals
- Closed
-
WT-5 How does pget work: is it necessary?
- Closed
-
WT-6 Complex schema example
- Closed
-
WT-7 Do we need the handle->err/errx methods?
- Closed
-
WT-8 Do we need table load, bulk-load and/or dump methods?
- Closed
-
WT-9 Does adding schema need to be transactional?
- Closed
-
WT-10 Basic "getting started" tutorial
- Closed
-
WT-11 placeholder #11
- Closed