Tom Lane wrote:
> Hiroshi Inoue <Inoue@tpf.co.jp> writes:
> > I fixed a few bugs caused by SearchSysCacheTuple()
> > recently. There seems to remain more similar bugs.
>
> Yes, I'm sure there are a lot :-(. We have talked about solving this
> by using some form of reference-counting in the cache, but I don't
> believe anyone's put forward a complete proposal. See for example
> the pghackers thread "Another nasty cache problem" around 30-Jan-2000.
>
Those bugs are pretty serious.
Sometimes they caused ERROR with strange messags.
Sometimes database-wide restart has been caused
by segmentation fault error. Database-wide restart
is pretty serious for any dbms.
And possibly they might have done the wrong thing
silently.
Isn't it practical to replace all susipicious Search
SysCacheTuple() by SearchSysCacheTupleCopy() ?
Regards.
Hiroshi Inoue