partition by range or by list constraint check (was Re: literal vs dynamic partition constraint in plan execution)

2019-09-02 Thread Luca Ferrari
I've done a simple test case, and find out that probably the problem I got was due to the partition schema I'm using. I want a table to be partitioned by a timestamp field with a first level partition by year, and a second level by month. Therefore, I did a BY LIST partitioning, but that produces a

Re: Upgrade 96 -> 11

2019-09-02 Thread Adrian Klaver
On 9/1/19 9:03 PM, James Sewell wrote: Hi, I'm in the process of upgrading from 96 -> 11 (on RHEL 7.3) . Both the versions have PostGIS 2.5.1 installed and working. pg_upgrade fails with: pg_restore: [archiver (db)] Error from TOC entry 440; 1259 537086 TABLE tablename databasename pg_resto

Re: partition by range or by list constraint check (was Re: literal vs dynamic partition constraint in plan execution)

2019-09-02 Thread Tom Lane
Luca Ferrari writes: > I've done a simple test case, and find out that probably the problem I > got was due to the partition schema I'm using. > I want a table to be partitioned by a timestamp field with a first > level partition by year, and a second level by month. Therefore, I did > a BY LIST p

Posible off topic ? pgmodeler

2019-09-02 Thread stan
Some very helpful folks pointed me to pgmodeler recently. I realize this list may not be the appropriate forum for discussing this, and would welcome pointers to a more appropriate forum. Having said that, this discussion may be a bit more generic. I have seen a number of open source products

Re: Posible off topic ? pgmodeler

2019-09-02 Thread Susan Hurst
Hi Stan! Don't know if you're interested in pgmodeler specifically or database modelers in general. At work, we use postgresql a lot and have found that DBeaver, which is open source, does a decent job of creating ERD diagrams. Personally, I don't like using DBeaver's interface for executin

Re: Posible off topic ? pgmodeler

2019-09-02 Thread Thiemo Kellner
Quoting stan : What I am trying to do, at the moment is get a complete understanding of their bossiness model, regarding the source code for this project. Thanks for any input on this. To the best of my knowledge Raphael tries to fund its time on the development of pgmodeler with selling

Re: Upgrade 96 -> 11

2019-09-02 Thread James Sewell
On Tue, 3 Sep 2019 at 7:15 am, Adrian Klaver wrote: > On 9/2/19 2:04 PM, James Sewell wrote: > Please reply to list also. > Ccing list. > > > > > > On Mon, 2 Sep 2019 at 11:56 pm, Adrian Klaver > > wrote: > > > > On 9/1/19 9:03 PM, James Sewell wrote: > >

Re: Upgrade 96 -> 11

2019-09-02 Thread Adrian Klaver
On 9/2/19 2:20 PM, James Sewell wrote: So this is a separate cluster from the one you used pg_upgrade on? In that case yes it was seperate In other words when doing pg_upgrade you could not get a working cluster, correct? Pg_upgrade does not complete - but as it hasn’t starte

Re: Upgrade 96 -> 11

2019-09-02 Thread Adrian Klaver
On 9/2/19 2:45 PM, James Sewell wrote: On Tue, 3 Sep 2019 at 7:41 am, Adrian Klaver > wrote: On 9/2/19 2:20 PM, James Sewell wrote: > >     So this is a separate cluster from the one you used pg_upgrade on? > > > In that case

Re: Upgrade 96 -> 11

2019-09-02 Thread James Sewell
On Tue, 3 Sep 2019 at 7:41 am, Adrian Klaver wrote: > On 9/2/19 2:20 PM, James Sewell wrote: > > > > > So this is a separate cluster from the one you used pg_upgrade on? > > > > > > In that case yes it was seperate > > > > > > > > In other words when doing pg_upgrade you could not get a w

Re: Upgrade 96 -> 11

2019-09-02 Thread Adrian Klaver
On 9/2/19 4:57 PM, James Sewell wrote: I am going to assume then that it has to do with this: "LINE 39:     "location_pt" "public"."geography"(Point,4283), " What is the above pointing to? This needs the PostGIS types and tables to work - they don't exist as they were not cr

Re: Upgrade 96 -> 11

2019-09-02 Thread James Sewell
> > > > > I am going to assume then that it has to do with this: > > "LINE 39: "location_pt" "public"."geography"(Point,4283), " > > > > What is the above pointing to? > > > > > > This needs the PostGIS types and tables to work - they don't exist as > > they were not created with CR

Re: Upgrade 96 -> 11

2019-09-02 Thread James Sewell
On Tue, 3 Sep 2019 at 07:55, Adrian Klaver wrote: > On 9/2/19 2:45 PM, James Sewell wrote: > > > > > > On Tue, 3 Sep 2019 at 7:41 am, Adrian Klaver > > wrote: > > > > On 9/2/19 2:20 PM, James Sewell wrote: > > > > > > > > So this is a separate

Re: Upgrade 96 -> 11

2019-09-02 Thread Adrian Klaver
On 9/2/19 5:20 PM, James Sewell wrote: What is the pg_upgrade command you are using? pg_upgrade --link --username postgres Where are you in being able to?: https://www.postgresql.org/docs/11/pgupgrade.html 16. Reverting to old cluster ... If the --link option was used, the data fil

Re: Upgrade 96 -> 11

2019-09-02 Thread James Sewell
On Tue, 3 Sep 2019 at 10:34, Adrian Klaver wrote: > On 9/2/19 5:20 PM, James Sewell wrote: > > > > > What is the pg_upgrade command you are using? > > > > > > pg_upgrade --link --username postgres > > Where are you in being able to?: > https://www.postgresql.org/docs/11/pgupgrade.html > > 16.

Re: Upgrade 96 -> 11

2019-09-02 Thread Adrian Klaver
On 9/2/19 5:38 PM, James Sewell wrote: On Tue, 3 Sep 2019 at 10:34, Adrian Klaver > wrote: On 9/2/19 5:20 PM, James Sewell wrote: > >     What is the pg_upgrade command you are using? > > > pg_upgrade --link --username postgres