Re: [Proposal] Global temporary tables

Andrew Bille <andrewbille@gmail.com> 于2021年10月28日周四 下午6:30写道：

Thanks, the "group by" is fixed

Yet another crash (on v58 patches), reproduced with:

psql -t -c "create global temp table t(b text) with(on_commit_delete_rows=true); create index idx_b on t (b); insert into t values('test'); alter table t alter b type varchar;"
server closed the connection unexpectedly
        This probably means the server terminated abnormally
        before or while processing the request.
connection to server was lost

Thank you for pointing that out.

This is due to an optimization point: ALTER Table reuses the relfilenode of the old index.

I have banned this optimization point for GTT, I am not entirely sure it is appropriate, maybe you can give some suggestions.

Please review the new code(v59).

Wenjing

 

with trace:

[New LWP 569199]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Core was generated by `postgres: andrew postgres [local] ALTER TABLE                                 '.
Program terminated with signal SIGABRT, Aborted.
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
50      ../sysdeps/unix/sysv/linux/raise.c: No such file or directory.
(gdb) bt
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
#1  0x00007f197493f859 in __GI_abort () at abort.c:79
#2  0x00005562b3306fb9 in ExceptionalCondition (conditionName=0x5562b34dd740 "reln->md_num_open_segs[forkNum] == 0", errorType=0x5562b34dd72c "FailedAssertion", fileName=0x5562b34dd727 "md.c", lineNumber=187) at assert.c:69
#3  0x00005562b3148f15 in mdcreate (reln=0x5562b41abdc0, forkNum=MAIN_FORKNUM, isRedo=false) at md.c:187
#4  0x00005562b314b73f in smgrcreate (reln=0x5562b41abdc0, forknum=MAIN_FORKNUM, isRedo=false) at smgr.c:335
#5  0x00005562b2d88b23 in RelationCreateStorage (rnode=..., relpersistence=103 'g', rel=0x7f196b597270) at storage.c:154
#6  0x00005562b2d5a408 in index_build (heapRelation=0x7f196b58dc40, indexRelation=0x7f196b597270, indexInfo=0x5562b4167d60, isreindex=true, parallel=false) at index.c:3038
#7  0x00005562b2d533c1 in RelationTruncateIndexes (heapRelation=0x7f196b58dc40, lockmode=1) at heap.c:3354
#8  0x00005562b2d5360b in heap_truncate_one_rel (rel=0x7f196b58dc40) at heap.c:3452
#9  0x00005562b2d53544 in heap_truncate (relids=0x5562b4167c58, is_global_temp=true) at heap.c:3410
#10 0x00005562b2ea09fc in PreCommit_on_commit_actions () at tablecmds.c:16495
#11 0x00005562b2d0d4ee in CommitTransaction () at xact.c:2140
#12 0x00005562b2d0e320 in CommitTransactionCommand () at xact.c:2979
#13 0x00005562b3151b7e in finish_xact_command () at postgres.c:2721
#14 0x00005562b314f340 in exec_simple_query (query_string=0x5562b40c2170 "create global temp table t(b text) with(on_commit_delete_rows=true); create index idx_b on t (b); insert into t values('test'); alter table t alter b type varchar;") at postgres.c:1239
#15 0x00005562b3153f0a in PostgresMain (dbname=0x5562b40ed6e8 "postgres", username=0x5562b40ed6c8 "andrew") at postgres.c:4497
#16 0x00005562b307df6e in BackendRun (port=0x5562b40e4500) at postmaster.c:4560
#17 0x00005562b307d853 in BackendStartup (port=0x5562b40e4500) at postmaster.c:4288
#18 0x00005562b3079a1d in ServerLoop () at postmaster.c:1801
#19 0x00005562b30791b6 in PostmasterMain (argc=3, argv=0x5562b40bc5b0) at postmaster.c:1473
#20 0x00005562b2f6d98e in main (argc=3, argv=0x5562b40bc5b0) at main.c:198

On Mon, Oct 25, 2021 at 7:13 PM wenjing <wjzeng2012@gmail.com> wrote:

I missed whole row and system column. It has been fixed in v58.

Please review the new code(v58) again