pgsql-bugs@postgresql.org writes:
> [avl@bdserver avl]$ nohup pg_dump -R -O leemail2 | gzip -c > /tmp/resp.leemail2_25_6_2002_1_00_pm.sql.gz &
> [avl@bdserver avl]$ jobs
> [1]+ Running nohup pg_dump -R -O leemail2 | gzip -c >/tmp/resp.leemail2_25_6_2002_1_00_pm.sql.gz &
> [avl@bdserver avl]$ ERROR: MemoryContextAlloc: invalid request size 4294967293
> PQendcopy: resetting connection
> SQL query to dump the contents of Table 'datamailentrada_historico' did not execute correctly. After we read all the
tablecontents from the backend, PQendcopy() failed. Explanation from backend: ''.
> The query was: 'COPY "datamailentrada_historico" TO stdout;
This looks like you have some corrupted data in
datamailentrada_historico :-(.
There is not much you can do about this except try to identify and
remove the damaged row(s). You can look in the mailing list archives
for previous discussions about systematic ways to identify the location
of the damage, but basically you want to home in on which row(s) can't
be selected without provoking the error.
Once you've removed the damaged rows, you may be fortunate enough to be
able to restore them from a recent backup. Good luck!
regards, tom lane