I am repeatedly getting invalid memory alloc request size errors on almost a daily basis and not always on the same table. To fix the troubled table, I find the record that either gives the memory error or makes postgres crash (gives a terminated by signal 11 error in the log) and delete it. Until I do that, I cannot select the data from the table by any means (select * , dump, copy to, etc).
Here is the output of a pg_dump getting the memory alloc error. I had successfully dumped this table the night before and there were no changes made to this table between the time it dumped successfully and failed. This table has 13,070 records.
pg_dump: dumping contents of table j005876_update
pg_dump: ERROR: invalid memory alloc request size 1095520841
pg_dump: SQL command to dump the contents of table "j005876_update" failed: PQendcopy() failed.
pg_dump: Error message from server: ERROR: invalid memory alloc request size 1095520841
pg_dump: The command was: COPY public.j005876_update (acct_id, list_id, order_id, cust_nbr, matchcode, title, gender, first_name, middle_name, last_name, full_name, care_of, addr1, addr2, addr3, city, county, state, postal_code, crrt, dpbc, country_3b, country_name, grp, class_id, segment, split_id, status_flags, drop_flag, suppress_code, mail_yn, onlabel, contest, part, print_version, letter_code, postage, source_code, mail_code, rec_nbr, list_seq) TO stdout;
pg_dump: *** aborted because of error
This error occurs most often on a table with about 60 million records. For this table we mostly do a nightly update of insert into's selecting data from daily work tables.
We are using Postgres 8.1.4 on an IBM box w/ 4 G of memory running Linux
Any help as to why this keeps happening is greatly appreciated. Please let me know if you need more information (I am fairly new to both Postgres and Linux so please be specific).
Thank you.
