On 5/15/2006 8:39 AM, Gavin Hamill wrote:
Andrew Sullivan wrote:
On Mon, May 15, 2006 at 10:15:13AM +0100, Gavin Hamill wrote:
Hrm. I'd still bet on data corruption. You might try walking
through the table, getting all the data, in chunks, and see if you
get a failure.
I just did a "pg_dump -F p"of that table alone in one fell swoop, and it
worked perfectly; it's left me a nice
This is a useless test. If the table contains any corruption of the kind
we are suspecting, then the dump will simply contain NULL's in a PK. You
won't be able to reload that dump into a table that defines that column
NOT NULL, but the dump itself would happily work.
I have attached a patch that I am about to commit into REL_1_1_STABLE as
well as HEAD. It will cause slon_quote_literal() to return a NULL result
for a NULL input and the log trigger will bail out with an ERROR if one
of the old key columns contains the NULL value. This will cause a
transaction that happens to touch such a row to abort, instead of
causing a postgres core dump.
Jan
COPY "RoomNight" ("ID", "BookingID", "RoomID", "Price", "Date",
"DateBooked", "Cancelled", "GuestName", "Adults", "Children",
"AllocationStatusID", "SmokingRequest", "BedRequested", "Incomplete",
"RoomNumber") FROM stdin;
followed by all 800k rows .. no segfaults, no errors at all :/
Why does this happen roughly once a day?
I'd hoped the corefile would've been able to tell me that, but seemingly
not :(
gdh
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#================================================== [EMAIL PROTECTED] #
Index: slony1_funcs.c
===================================================================
RCS file: /usr/local/cvsroot/slony1/slony1-engine/src/backend/slony1_funcs.c,v
retrieving revision 1.33.2.4
diff -c -r1.33.2.4 slony1_funcs.c
*** slony1_funcs.c 6 Jan 2006 17:07:46 -0000 1.33.2.4
--- slony1_funcs.c 15 May 2006 14:36:48 -0000
***************
*** 781,786 ****
--- 781,789 ----
continue;
col_ident = (char
*)slon_quote_identifier(SPI_fname(tupdesc, i + 1));
col_value = slon_quote_literal(SPI_getvalue(old_row,
tupdesc, i + 1));
+ if (col_value == NULL)
+ elog(ERROR, "Slony-I: old key column %s.%s IS
NULL on UPDATE",
+
NameStr(tg->tg_relation->rd_rel->relname), col_ident);
cmddata_need = (cp - (char *)(cs->cmddata_buf)) + 16 +
(len_ident = strlen(col_ident)) +
***************
*** 845,850 ****
--- 848,856 ----
continue;
col_ident = (char
*)slon_quote_identifier(SPI_fname(tupdesc, i + 1));
col_value = slon_quote_literal(SPI_getvalue(old_row,
tupdesc, i + 1));
+ if (col_value == NULL)
+ elog(ERROR, "Slony-I: old key column %s.%s IS
NULL on DELETE",
+
NameStr(tg->tg_relation->rd_rel->relname), col_ident);
cmddata_need = (cp - (char *)(cs->cmddata_buf)) + 16 +
(len_ident = strlen(col_ident)) +
***************
*** 1041,1046 ****
--- 1047,1055 ----
int len;
int wl;
+ if (str == NULL)
+ return NULL;
+
len = strlen(str);
result = palloc(len * 2 + 3);
cp1 = str;
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general
