| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Heikki Linnakangas <hlinnakangas(at)vmware(dot)com> |
| Cc: | fabriziomello(at)gmail(dot)com, Robert Haas <robertmhaas(at)gmail(dot)com>, Andres Freund <andres(at)2ndquadrant(dot)com>, Stephen Frost <sfrost(at)snowman(dot)net>, Pgsql Hackers <pgsql-hackers(at)postgresql(dot)org>, Thom Brown <thom(at)linux(dot)com> |
| Subject: | Re: GSoC proposal - "make an unlogged table logged" |
| Date: | 2014-04-03 13:46:37 |
| Message-ID: | 16784.1396532797@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Heikki Linnakangas <hlinnakangas(at)vmware(dot)com> writes:
> No-one's replied yet, but perhaps the worry is that after you've written
> the commit record, you have to go ahead with removing/creating the init
> fork, and that is seen as too risky. If a creat() or unlink() call
> fails, that will have to be a PANIC, and crash recovery will likewise
> have to PANIC if the forks still cannot be removed/created.
> My first thought is that that seems ok.
No, it isn't. No filesystem operation should *ever* be thought to be
guaranteed to succeed.
I also concur with Andres' complaint that this feature is not worth
adding complication to the core transaction commit path for.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Heikki Linnakangas | 2014-04-03 14:14:23 | WAL format and API changes (9.5) |
| Previous Message | Andres Freund | 2014-04-03 13:46:04 | Re: quiet inline configure check misses a step for clang |