Re: [GENERAL] Increased I/O / Writes

2016-05-23 Thread Lucas Possamai
On 24 May 2016 at 09:40, Kevin Grittner wrote: > On Mon, May 23, 2016 at 4:32 PM, Lucas Possamai > wrote: > > > The spikes were happening because, after the reboot, the HUGE PAGES were > > enabled. > > > > After disabling them, all got back to normal. > > Since you said earlier you were on Postg

Re: [GENERAL] Increased I/O / Writes

2016-05-23 Thread Kevin Grittner
On Mon, May 23, 2016 at 4:32 PM, Lucas Possamai wrote: > The spikes were happening because, after the reboot, the HUGE PAGES were > enabled. > > After disabling them, all got back to normal. Since you said earlier you were on PostgreSQL version 9.2, you must be talking about transparent huge pag

Re: [GENERAL] Increased I/O / Writes

2016-05-23 Thread Lucas Possamai
Just an update about this... As you know, the server was rebooted. The spikes were happening because, after the reboot, the HUGE PAGES were enabled. After disabling them, all got back to normal. Cheers

Re: [GENERAL] Increased I/O / Writes

2016-05-17 Thread John R Pierce
On 5/17/2016 3:56 PM, Lucas Possamai wrote: *FATAL*: sorry, too many clients already I believe that error means you've exceeded max_connections. query pg_stat_activity and see how many connections are in the various 'states', like.. select state, count(*) from pg_stat_activity group

Re: [GENERAL] Increased I/O / Writes

2016-05-17 Thread Lucas Possamai
Following https://www.keithf4.com/a-large-database-does-not-mean-large-shared_buffers/ and http://pgtune.leopard.in.ua/ and https://wiki.postgresql.org/wiki/Tuning_Your_PostgreSQL_Server I changed the shared_buffer from 51GB to 35GB. Now, I'm getting spikes every 15 minutes. *FATAL*: sorry, too

Re: [GENERAL] Increased I/O / Writes

2016-05-17 Thread Kevin Grittner
On Tue, May 17, 2016 at 12:11 AM, Lucas Possamai wrote: > This is my postgresql.conf at the moment: > > shared_buffer(51605MB) + > effective_cache_size(96760MB) + > work_mem(32MB) + > max_connections(200) > > = 148397.08 MB You are comparing some very dissimilar settings. effective_cache_size do

Re: [GENERAL] Increased I/O / Writes

2016-05-17 Thread Johnny Morano
, 17. Mai 2016 06:12 To: Sergey Konoplev Cc: Postgres General Subject: Re: [GENERAL] Increased I/O / Writes This is my postgresql.conf at the moment: shared_buffer(51605MB) + effective_cache_size(96760MB) + work_mem(32MB) + max_connections(200) = 148397.08 MB My server has 128GB of RAM So, I&#

Re: [GENERAL] Increased I/O / Writes

2016-05-16 Thread Lucas Possamai
This is my postgresql.conf at the moment: shared_buffer(51605MB) + effective_cache_size(96760MB) + work_mem(32MB) + max_connections(200) *= 148397.08 MB* My server has 128GB of RAM So, I'm using more RAM that I have. (not good at all) I'm glad that it wasn't me who put those confs in there :)

Re: [GENERAL] Increased I/O / Writes

2016-05-10 Thread Sergey Konoplev
On Sun, May 8, 2016 at 7:07 PM, drum.lu...@gmail.com wrote: > Is there a way to detect the queries that are causing that? > > I can use pg_stat_statements to get the most usage queries, but I was > wondering how can I find the queries that are causing that much IO? Take a look at this tool: htt

Re: [GENERAL] Increased I/O / Writes

2016-05-10 Thread Melvin Davidson
On Tue, May 10, 2016 at 1:31 PM, Joshua D. Drake wrote: > On 05/10/2016 09:50 AM, Igor Neyman wrote: > > Please, if anyone can share anything.. Thanks a lot! >> >> Lucas >> >> So, what’s wrong with using pg_stat_statements? >> >> It has a set of colu

Re: [GENERAL] Increased I/O / Writes

2016-05-10 Thread Joshua D. Drake
On 05/10/2016 09:50 AM, Igor Neyman wrote: Please, if anyone can share anything.. Thanks a lot! Lucas So, what’s wrong with using pg_stat_statements? It has a set of columns pertaining to IO. You could use iotop to determine which postgres pid is eating the IO, then use statement logging

Re: [GENERAL] Increased I/O / Writes

2016-05-10 Thread Igor Neyman
From: pgsql-general-ow...@postgresql.org [mailto:pgsql-general-ow...@postgresql.org] On Behalf Of drum.lu...@gmail.com Sent: Sunday, May 08, 2016 10:07 PM To: Postgres General Subject: [GENERAL] Increased I/O / Writes Hi all, I've recently an increase IO for writes on my DB. I'm