| From: | Czarek <keracz(at)protonmail(dot)com> |
|---|---|
| To: | "pgsql-performance(at)lists(dot)postgresql(dot)org" <pgsql-performance(at)lists(dot)postgresql(dot)org> |
| Subject: | Problems with Multixacts LWLocks |
| Date: | 2020-07-29 14:18:44 |
| Message-ID: | OajTzsuYhm-LUQDPPn_2uujez2BKe_A-d7O65jNqjBafGFI3-2poTR6z7MPvlyryYWjE0OIvyqCw29b9QA526CU3SP5QgXGAGCeHy-6LNYk=@protonmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-performance |
Hi,
I have problem with my Postgres production server.
Sometimes there a lot of Multiexact* LWlocks:
MultiXactOffsetControlLock
MultiXactMemberControlLock
that significantly impact on users queries performance.
How to find what is the reason of Multiexact* LWlocks?
Hardware:
110 CPU
620 GB RAM
CentOS 7.4.1708
kernel 3.10.0-693.5.2.el7.x86_64
Database:
size 4,5 TB
server_version 9.6
checkpoint_timeout 1800 s
effective_cache_size 449280 MB
maintenance_work_mem 4096 MB
max_connections 440
shared_buffers 149760 MB
work_mem 190 MB
Any ideas?
Best regards
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Burhan Akbulut | 2020-08-11 20:52:53 | Hstore index for full text search |
| Previous Message | Julian Wolf | 2020-07-29 06:17:06 | Re: Too few rows expected by Planner on partitioned tables |