Did you try using NOWAIT instead of SKIP LOCKED to see if the behavior
still shows up?
On Tue, Aug 18, 2020, 8:22 PM Jim Jarvie wrote:
> Thank you for the quick response.
>
> No adjustments of fill factors. Hadn't though of that - I'll investigate
> and try some options to see if I can measure
On Wed, Jul 15, 2020 at 8:24 AM Henrique Montenegro
wrote:
>
>
> On Wed, Jul 15, 2020 at 4:03 AM Sebastian Dressler
> wrote:
>
>> Hi Henrique,
>>
>> On 15. Jul 2020, at 03:13, Henrique Montenegro wrote:
>> [...]
>>
>> ```
>> ssl = o
On Wed, Jul 15, 2020 at 4:03 AM Sebastian Dressler
wrote:
> Hi Henrique,
>
> On 15. Jul 2020, at 03:13, Henrique Montenegro wrote:
> [...]
>
> ```
> ssl = off
> shared_buffers = 160GB # min 128kB
> work_mem = 96GB # min 64kB
&g
ackground_ratio = 10
$ sysctl vm.dirty_expire_centisecs
vm.dirty_expire_centisecs = 3000
These are default values; if what I understood from them is right, it seems
to me that these values should be fine.
On Mon, Jul 13, 2020 at 9:02 PM Henrique Montenegro
wrote:
>
>
> On Mon, Jul 13, 2020 at 8:05 PM
On Mon, Jul 13, 2020 at 8:05 PM Jeff Janes wrote:
> On Mon, Jul 13, 2020 at 10:23 AM Henrique Montenegro
> wrote:
>
> insert into users_no_dups (
>> created_ts,
>> user_id,
>> name,
>> url
>> ) (
>> select
>>
On Mon, Jul 13, 2020 at 12:50 PM Sebastian Dressler
wrote:
> Hi Henrique,
>
> On 13. Jul 2020, at 18:42, Henrique Montenegro wrote:
>
> On Mon, Jul 13, 2020 at 11:20 AM Sebastian Dressler
> wrote:
>
>
>> Running the above loop worked fine for about 12 hours. Each
On Mon, Jul 13, 2020 at 11:20 AM Sebastian Dressler
wrote:
> Hi Henrique,
>
> On 13. Jul 2020, at 16:23, Henrique Montenegro wrote:
>
> [...]
>
> * Insert the data from the `users` table into the `users_no_dups` table
>
> ```
> insert into users_no_dups (
&
On Mon, Jul 13, 2020 at 12:28 PM Michael Lewis wrote:
> Is this an insert only table and perhaps not being picked up by
> autovacuum? If so, try a manual "vacuum analyze" before/after each batch
> run perhaps. You don't mention updates, but also have been adjusting
> fillfactor so I am not not su
Hello list,
I am having issues with performance inserting data in Postgres and would
like
to ask for help figuring out the problem as I ran out of ideas.
I have a process that generates a CSV file with 1 million records in it
every
5 minutes and each file is about 240MB. I need this data to be in
