Re: WAL records - Mailing list pgsql-hackers

From Tomas Vondra
Subject Re: WAL records
Date
Msg-id 20190926184739.3c6xmp4fakthfr5o@development
Whole thread Raw
In response to WAL records  (Looserof7 <looserof7@gmail.com>)
List pgsql-hackers
On Thu, Sep 26, 2019 at 11:14:53AM +0530, Looserof7 wrote:
>I have started to learn postgresql. While going through the WAL dump saw
>the below records,
>
>rmgr: Heap2       len (rec/tot):     60/    60, tx:          0, lsn:
>4F/CFF1F0F8, prev 4F/CFF1EB70, desc: CLEAN remxid 0, blkref #0: rel
>1663/16385/1259 blk 1301
>
>rmgr: Heap2       len (rec/tot):     60/    60, tx:          0, lsn:
>4F/CF26F808, prev 4F/CF26F7C0, desc: CLEAN remxid 0, blkref #0: rel
>1663/16385/3312732 blk 2634
>
>rmgr: Heap2       len (rec/tot):     56/    56, tx:   79718109, lsn:
>4F/CF96D970, prev 4F/CF96D930, desc: CLEAN remxid 0, blkref #0: rel
>1663/16385/209012 blk 6621
>
>
>Wanted to know more about these,
>
>      1. Are "CLEAN" vacuum records ? If not, what are those ?
>      2. Why is the transaction id "0" ?
>      3. What does these records do ?
>
>Postgresql version : 10.6

It's an information for the replica (hot-standby) with information about
recent data removed from the table by vacuum, so that the standby may
abort user queries that'd need the data etc. See heap_xlog_clean()
function in heapam.c.

regards

-- 
Tomas Vondra                  http://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services



pgsql-hackers by date:

Previous
From: Tomas Vondra
Date:
Subject: Re: PATCH: logical_work_mem and logical streaming of largein-progress transactions
Next
From: Tomas Vondra
Date:
Subject: Re: WIP: BRIN multi-range indexes