Killing a data modifying transaction - Mailing list pgsql-general

Hi All,

I've got two transactions I tried to kill 3 days ago using "select
pg_cancel_backend(<pid>)", then SIGTERM, and have since then been
using 100% of a cpu core each. They were supposed to insert the
results of large unions with PostGIS and appear to have failed.
Could someone tell me what's the least worst option here please? If I
kill -9 will I corrupt my data directory?  (I've searched for an
answer in the FAQ and the wiki, to no avail). I'm on Ubuntu server
8.04.1 and Postgres 8.3.7.

Thanks and regards,

Will Temperley