Re: [PATCHES] [HACKERS] Including Snapshot Info with Indexes

2008-01-16 Thread Gokulakannan Somasundaram
Hi, I did some more bug fixes and performance updates especially for select count(1) queries. Thanks, Gokul. patchfile.tar.gz Description: GNU Zip compressed data ---(end of broadcast)--- TIP 5: don't forget to increase your free space map set

[PATCHES] Proposed patch for renaming constraints along with indexes

2008-01-16 Thread Tom Lane
As per my suggestion just now for resolving bug #3854. I'm thinking in terms of applying this for 8.3, but not back-patching --- though I suppose depending on how seriously you take the bug, a case could be made for back-patching. I don't offhand see anyplace that this would need to be documented

Re: [PATCHES] [HACKERS] SSL over Unix-domain sockets

2008-01-16 Thread Bruce Momjian
Bruce Momjian wrote: > Tom Lane wrote: > > Conclusions: > > > > * SSL, even without real authentication, is *way* too expensive to > > enable by default. > > > > * The extra cost of going across a local TCP connection is measurable, > > but it's insignificant compared to the cost of turning on SS

Re: [PATCHES] postgresql in FreeBSD jails: proposal

2008-01-16 Thread Marc G. Fournier
[EMAIL PROTECTED] (Mischa Sandberg) writes: >Unfortunately, with multiple jails running PG servers and (due to app >limitations) all servers having same PGPORT, you get the situation that >when jail#2 (,jail#3,...) server comes up, it: >- detects that there is a shm seg with ipc key 5432001 >- che

[PATCHES] Compile of Test example is put into win32.mak of libpq.

2008-01-16 Thread Hiroshi Saito
Hi. First: I needed other programs, in order to test what always built independent libpq.dll. Therefore, environment was considered variously. Then, I thought that a sample program was helpful. They are written to the document. Next: I understood that test example could not compile by MinGW (