Re: [UNVERIFIED SENDER] FW: autovac issue with large number of tables

Nasby, Jim <nasbyj@amazon.com>

From: Jim Nasby <nasbyj@amazon.com>
To: "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>
Date: 2020-07-27T18:41:54Z
Lists: pgsql-hackers
Sorry, please ignore this duplicate!

On 7/27/20 1:39 PM, Nasby, Jim wrote:
>
> A database with a very large number of  tables eligible for autovacuum 
> can result in autovacuum workers “stuck” in a tight loop of 
> table_recheck_autovac() constantly reporting nothing to do on the 
> table. This is because a database with a very large number of tables 
> means it takes a while to search the statistics hash to verify that 
> the table still needs to be processed[1]. If a worker spends some time 
> processing a table, when it’s done it can spend a significant amount 
> of time rechecking each table that it identified at launch (I’ve seen 
> a worker in this state for over an hour). A simple work-around in this 
> scenario is to kill the worker; the launcher will quickly fire up a 
> new worker on the same database, and that worker will build a new list 
> of tables.
>
> That’s not a complete solution though… if the database contains a 
> large number of very small tables you can end up in a state where 1 or 
> 2 workers is busy chugging through those small tables so quickly than 
> any additional workers spend all their time in 
> table_recheck_autovac(), because that takes long enough that the 
> additional workers are never able to “leapfrog” the workers that are 
> doing useful work.
>
> PoC patch attached.
>
> 1: top hits from `perf top -p xxx` on an affected worker
>
> Samples: 72K of event 'cycles', Event count (approx.): 17131910436
>
> Overhead Shared Object     Symbol
>
>   42.62% postgres          [.] hash_search_with_hash_value
>
>   10.34% libc-2.17.so      [.] __memcpy_sse2
>
>    6.99% [kernel]          [k] copy_user_enhanced_fast_string
>
>    4.73% libc-2.17.so      [.] _IO_fread
>
>    3.91% postgres          [.] 0x00000000002d6478
>
>    2.95% libc-2.17.so      [.] _IO_getc
>
>    2.44% libc-2.17.so      [.] _IO_file_xsgetn
>
>    1.73% postgres          [.] hash_search
>
>    1.65% [kernel]          [k] find_get_entry
>
>    1.10% postgres          [.] hash_uint32
>
>    0.99% libc-2.17.so      [.] __memcpy_ssse3_back
>

Commits

  1. Speed up rechecking if relation needs to be vacuumed or analyze in autovacuum.