Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-10-02 Thread Daniel Gustafsson
> On 13 Sep 2017, at 15:01, Tomas Vondra wrote: > > On 09/13/2017 07:53 AM, Andrey Borodin wrote: >>> * I see there are conditions like this: >>> >>>if(xlogreader->blocks[nblock].forknum == MAIN_FORKNUM) >>> >>> Why is it enough to restrict the block-tracking

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-09-13 Thread Tomas Vondra
On 09/13/2017 07:53 AM, Andrey Borodin wrote: >> * I see there are conditions like this: >> >>    if(xlogreader->blocks[nblock].forknum == MAIN_FORKNUM) >> >> Why is it enough to restrict the block-tracking code to main fork? >> Aren't we interested in all relation forks? > fsm, vm and others

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-09-13 Thread Andrey Borodin
Hi! Thank you for your interest and experiment results. > 13 сент. 2017 г., в 15:43, Ants Aasma написал(а): > > On Thu, Aug 31, 2017 at 9:02 AM, Andrey Borodin wrote: >> When we have accumulated diff blocknumbers for most of segments we can >>

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-09-13 Thread Ants Aasma
On Thu, Aug 31, 2017 at 9:02 AM, Andrey Borodin wrote: > When we have accumulated diff blocknumbers for most of segments we can > significantly speed up method of WAL scanning. If we have blocknumbers for > all segments we can skip WAL scanning at all. Have you measured

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-09-12 Thread Andrey Borodin
Hi Tomas! Thank you for looking into that patch. > 8 сент. 2017 г., в 1:53, Tomas Vondra > написал(а): > > A few more comments: > > * The patch defines wal_switch_hook, but it's never called. That call was missing, that's a bug, thanks for spotting that out. > *

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-09-07 Thread Tomas Vondra
Hi, On 09/01/2017 08:13 AM, Andrey Borodin wrote: > Thank you for your reply, Michael! Your comments are valuable, especially in the world of backups. > >> 31 авг. 2017 г., в 19:44, Michael Paquier написал(а): >> Such things are not Postgres-C like. > Will be fixed. >

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-09-01 Thread Andrey Borodin
Thank you for your reply, Michael! Your comments are valuable, especially in the world of backups. > 31 авг. 2017 г., в 19:44, Michael Paquier > написал(а): > Such things are not Postgres-C like. Will be fixed. > I don't understand what xlog_begin_insert_hook() is

Re: [HACKERS] Hooks to track changed pages for backup purposes

2017-08-31 Thread Michael Paquier
On Thu, Aug 31, 2017 at 3:02 PM, Andrey Borodin wrote: > Here is the patch with hooks that I consider sufficient for implementation of > incremental backup with pages tracking as extension. > > Recently I was posting these things to the thread "Adding hook in BufferSync >

[HACKERS] Hooks to track changed pages for backup purposes

2017-08-31 Thread Andrey Borodin
Hi hackers! Here is the patch with hooks that I consider sufficient for implementation of incremental backup with pages tracking as extension. Recently I was posting these things to the thread "Adding hook in BufferSync for backup purposes" [0], but here I start separate thread since Subj