Обсуждение: backup server core when redo btree_xlog_insert that type is XLOG_BTREE_INSERT_POST

My database version is Postgresql 13.2 , backup server core when redo  a  btree_xlog_insert that type is XLOG_BTREE_INSERT_POST

#0 0x00002aab66695d86inmemmove ssse3 krom /lib64/libc.so.6

#1 0x00000000004f5574in_bt_swap_posting(newitem=0x125d998,oposting=0x2aabcd55dcc8ostingoff=13) at nbtdedup.c:796

#2 0x000000000050cf08 in btree_xlog_insert(isleaf=true,ismeta=false, posting=true,r ecord=0x122d7b8)at nbtxlog.c:224

43 0x000000000050ed53 in btree_redo(record=0x122d7b8) at nbtxlog.c:969

44 0x00000000005487b5 in StartupXLOG()at log.c:7320

#5 0x00000000008292ab in StartupProcessMain () atstartup.c:204

46 0x000000000055c916 in AuxiliaryProcessMain(argc=2,argv=0x7fff071f8a00) at bootstrap.c:443

#7 0x000000000082820d in StartChildProcess(type=StartupProcess) at postmaster.c:5492

48 0x00000000008236ce in PostmasterMain(argc=1argv=0x11fb2e0)at postmaster.c:1404

#9 0x00000000007398f7 in main(argc=1,argv=0x11fb2e0)at main.c:210

gdb) f 1

#1 0x00000000004f5574in btswap_posting(newitem=0x125d998,oposting=0x2aabcd55dcc8ostingoff=13) at nbtdedup.c.796

796

gdb) p((Size)((oposting)->t_info & 0x1FFF))         //oposting index tuple size 

 $12 = 16

gdb) p nhtids

$13= 17

gdb) p postingoff

$14=13

gdb) p nmovebytes

$15=18

IndexTuple

_bt_swap_posting(IndexTuple newitem, IndexTuple oposting, int postingoff)

{

int nhtids;

char    *replacepos;

char    *replaceposright;

Size nmovebytes;

IndexTuple nposting;

nhtids = BTreeTupleGetNPosting(oposting);

       ===  nhtids = 17, postingoff = 13

Assert(_bt_posting_valid(oposting));

Assert(postingoff > 0 && postingoff < nhtids);

/*

* Move item pointers in posting list to make a gap for the new item's

* heap TID.  We shift TIDs one place to the right, losing original

* rightmost TID. (nmovebytes must not include TIDs to the left of

* postingoff, nor the existing rightmost/max TID that gets overwritten.)

*/

nposting = CopyIndexTuple(oposting);

===  IndexTupleSize(oposting) = 16, nposting memory size 16

replacepos = (char *) BTreeTupleGetPostingN(nposting, postingoff);

replaceposright = (char *) BTreeTupleGetPostingN(nposting, postingoff + 1);

nmovebytes = (nhtids - postingoff - 1) * sizeof(ItemPointerData);

memmove(replaceposright, replacepos, nmovebytes);

===core here,  nmovebytes = 18, for nposting size is 16, so here is  out of memory

}

At 2024-11-21 23:58:03, "Peter Geoghegan" <pg@bowt.ie> wrote:
>On Thu, Nov 21, 2024 at 10:03 AM yuansong <yyuansong@126.com> wrote:
>> Should nhtids be less than or equal to IndexTupleSize(oposting)?
>>  Why is nhtids larger than IndexTupleSize(oposting) ? I think there should be an error in the master host writing the wal log.
>> Does anyone know when this will happen?
>
>It'll happen whenever there is a certain kind of data corruption.
>
>There were complaints about issues like this in the past. But those
>complaints seem to have gone away when more hardening was added to the
>code that runs during original execution (not the REDO routine code,
>which can only do what it is told to do by the WAL record).
>
>You're using PostgreSQL 13.2, which is a very old point release that
>lacks this hardening -- the current 13 point release is 13.18, so
>you're missing a lot. Had you been on a later point release you'd very
>probably have still had the issue with corruption (which could be from
>bad hardware), but you likely would have avoided the problem with the
>REDO routine crashing like this.
>
>-- 
>Peter Geoghegan

At 2024-11-27 18:53:20, "yuansong" <yyuansong@126.com> wrote:

we find crash reson

We have identified the cause of the crash: it was due to the XLOG_BTREE_INSERT_POST XLOG having an OffsetNumber offnum that was one less than what was stored in the index. I experimented with adding +1, and the index data remained normal in both cases. This issue is likely caused by concurrent operations on the B-tree, and upon reviewing the corresponding WAL logs, we found SPLIT_L and INSERT_LEAF operations on the same block before the crash. This might be a bug. I'm not sure if there's a related fix.

At 2024-11-21 23:58:03, "Peter Geoghegan" <pg@bowt.ie> wrote:
>On Thu, Nov 21, 2024 at 10:03 AM yuansong <yyuansong@126.com> wrote:
>> Should nhtids be less than or equal to IndexTupleSize(oposting)?
>>  Why is nhtids larger than IndexTupleSize(oposting) ? I think there should be an error in the master host writing the wal log.
>> Does anyone know when this will happen?
>
>It'll happen whenever there is a certain kind of data corruption.
>
>There were complaints about issues like this in the past. But those
>complaints seem to have gone away when more hardening was added to the
>code that runs during original execution (not the REDO routine code,
>which can only do what it is told to do by the WAL record).
>
>You're using PostgreSQL 13.2, which is a very old point release that
>lacks this hardening -- the current 13 point release is 13.18, so
>you're missing a lot. Had you been on a later point release you'd very
>probably have still had the issue with corruption (which could be from
>bad hardware), but you likely would have avoided the problem with the
>REDO routine crashing like this.
>
>-- 
>Peter Geoghegan

At 2024-11-27 18:53:20, "yuansong" <yyuansong@126.com> wrote:

we find crash reson

We have identified the cause of the crash: it was due to the XLOG_BTREE_INSERT_POST XLOG having an OffsetNumber offnum that was one less than what was stored in the index. I experimented with adding +1, and the index data remained normal in both cases. This issue is likely caused by concurrent operations on the B-tree, and upon reviewing the corresponding WAL logs, we found SPLIT_L and INSERT_LEAF operations on the same block before the crash. This might be a bug. I'm not sure if there's a related fix.

At 2024-11-21 23:58:03, "Peter Geoghegan" <pg@bowt.ie> wrote:
>On Thu, Nov 21, 2024 at 10:03 AM yuansong <yyuansong@126.com> wrote:
>> Should nhtids be less than or equal to IndexTupleSize(oposting)?
>>  Why is nhtids larger than IndexTupleSize(oposting) ? I think there should be an error in the master host writing the wal log.
>> Does anyone know when this will happen?
>
>It'll happen whenever there is a certain kind of data corruption.
>
>There were complaints about issues like this in the past. But those
>complaints seem to have gone away when more hardening was added to the
>code that runs during original execution (not the REDO routine code,
>which can only do what it is told to do by the WAL record).
>
>You're using PostgreSQL 13.2, which is a very old point release that
>lacks this hardening -- the current 13 point release is 13.18, so
>you're missing a lot. Had you been on a later point release you'd very
>probably have still had the issue with corruption (which could be from
>bad hardware), but you likely would have avoided the problem with the
>REDO routine crashing like this.
>
>-- 
>Peter Geoghegan