Menu
▾
▴
postgres-xc-general — General info and messages
You can subscribe to this list here.
| 2010 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(2) |
Jun
|
Jul
|
Aug
(6) |
Sep
|
Oct
(19) |
Nov
(1) |
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2011 |
Jan
(12) |
Feb
(1) |
Mar
(4) |
Apr
(4) |
May
(32) |
Jun
(12) |
Jul
(11) |
Aug
(1) |
Sep
(6) |
Oct
(3) |
Nov
|
Dec
(10) |
| 2012 |
Jan
(11) |
Feb
(1) |
Mar
(3) |
Apr
(25) |
May
(53) |
Jun
(38) |
Jul
(103) |
Aug
(54) |
Sep
(31) |
Oct
(66) |
Nov
(77) |
Dec
(20) |
| 2013 |
Jan
(91) |
Feb
(86) |
Mar
(103) |
Apr
(107) |
May
(25) |
Jun
(37) |
Jul
(17) |
Aug
(59) |
Sep
(38) |
Oct
(78) |
Nov
(29) |
Dec
(15) |
| 2014 |
Jan
(23) |
Feb
(82) |
Mar
(118) |
Apr
(101) |
May
(103) |
Jun
(45) |
Jul
(6) |
Aug
(10) |
Sep
|
Oct
(32) |
Nov
|
Dec
(9) |
| 2015 |
Jan
(3) |
Feb
(5) |
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
(9) |
Aug
(4) |
Sep
(3) |
Oct
|
Nov
|
Dec
|
| 2016 |
Jan
(3) |
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2017 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
(3) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2018 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(4) |
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| S | M | T | W | T | F | S |
|---|---|---|---|---|---|---|
|
|
|
1
|
2
|
3
|
4
(1) |
5
(5) |
|
6
(1) |
7
(3) |
8
(2) |
9
(4) |
10
(6) |
11
(2) |
12
|
|
13
|
14
(1) |
15
(2) |
16
(2) |
17
(6) |
18
|
19
(1) |
|
20
|
21
(4) |
22
(4) |
23
(5) |
24
(4) |
25
(1) |
26
(2) |
|
27
(2) |
28
(2) |
29
(6) |
30
(11) |
31
(14) |
|
|
|
From: Ashutosh B. <ash...@en...> - 2013-01-22 04:25:15
|
On Tue, Jan 22, 2013 at 3:41 AM, Filip Rembiałkowski <fil...@gm...> wrote: > Hi PG-XC authors. > > I am reading the PGCon2012 Tutorial presentation > [ http://postgresxc.wikia.com/wiki/File:20120515_PGXC_Tutorial_global_1.pdf ] > > Thank you for this document - it really covers many things. > > On page 44 you write about data distribution, and using distributed > tables versus replicated tables. > You use the term "Statement based replication" - and this alerted me instantly. > statement based replication simply means that data is replicated using SQL statements as opposed to WALs etc. These statements are generated at the coordinator, while planning the query. Postgres-XC is intelligent enough to understand which parts of original DML can be evaluated at the datanodes and which are not and it generates the statements by feeding values for the parts that need evaluation at coordinator (like volatile functions). So the problems like pgpool do not occur in Postgres-XC. > What _exactly_ did you mean by this? > > As you probably know, pgpool project uses exactly same term, and for > them it means sending same statement to many parallel nodes. > This approach is proven to be flawed by design - at least when you use > any volatile functions when writing, it makes replicas inconsistent. > > Thanks, > Filip > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. SALE $99.99 this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122412 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general -- Best Wishes, Ashutosh Bapat EntepriseDB Corporation The Enterprise Postgres Company |
|
From: Michael P. <mic...@gm...> - 2013-01-21 23:14:26
|
On Tue, Jan 22, 2013 at 7:11 AM, Filip Rembiałkowski < fil...@gm...> wrote: > Hi PG-XC authors. > > I am reading the PGCon2012 Tutorial presentation > [ > http://postgresxc.wikia.com/wiki/File:20120515_PGXC_Tutorial_global_1.pdf] > > Thank you for this document - it really covers many things. > > On page 44 you write about data distribution, and using distributed > tables versus replicated tables. > You use the term "Statement based replication" - and this alerted me > instantly. > > What _exactly_ did you mean by this? > A query string is generated on Coordinator and used to communicate with the remote nodes. For replicated tables, the same query is used on all the nodes if SQL is a DML, on one node if it is a SELECT. As you probably know, pgpool project uses exactly same term, and for > them it means sending same statement to many parallel nodes. > This approach is proven to be flawed by design - at least when you use > any volatile functions when writing, it makes replicas inconsistent. > The term might be the same, but XC is integrated directly with the Postgres core, while pgpool is a different instance running on top of Postgres servers. In consequence, the query planner in XC can use the concept of query clause shippability. This means that some part of the query can be shipped to remote nodes (meaning that the function or clause is managed on remote Datanodes) or non-shippable (in this case clause is compiled on Coordinator). A volatile function is always considered as non-shippable as it plays by definition with the database data, so in the case of XC its value is computed on Coordinator once and what is sent to Datanodes is: 1) A query of the type "SELECT * FROM tab WHERE val = $1" (It might have been something similar to "SELECT * from tab WHERE val = nextval('seq')" on Coordinator, $1 is replaced by the value computed on Coordinator) 2) The value computed. Make a try with an XC with a typical example of this type: CREATE TABLE tab (a serial) DISTRIBUTE BY REPLICATION; INSERT INTO tab VALUES (DEFAULT); Then have a look at the logs of the Datanodes, you will notice that what has been sent down is not the query received by Coordinator, but something like: Received query from Coordinator 'INSERT INTO tab VALUES ($1)' Received values: $1 = '1' Or similar. So no need to worry about data inconsistencies in those cases. -- Michael Paquier http://michael.otacoo.com |
|
From: Filip R. <fil...@gm...> - 2013-01-21 22:12:10
|
Hi PG-XC authors. I am reading the PGCon2012 Tutorial presentation [ http://postgresxc.wikia.com/wiki/File:20120515_PGXC_Tutorial_global_1.pdf ] Thank you for this document - it really covers many things. On page 44 you write about data distribution, and using distributed tables versus replicated tables. You use the term "Statement based replication" - and this alerted me instantly. What _exactly_ did you mean by this? As you probably know, pgpool project uses exactly same term, and for them it means sending same statement to many parallel nodes. This approach is proven to be flawed by design - at least when you use any volatile functions when writing, it makes replicas inconsistent. Thanks, Filip |
|
From: Koichi S. <koi...@gm...> - 2013-01-21 08:42:01
|
Hello, Please visit http://postgresxc.wikia.com/wiki/Postgres-XC_Wiki. I added a couple of link and files uploaded in this page. I will find other useful materials and will upload them. Good luck. ---------- Koichi Suzuki 2013/1/21 Ashutosh Bapat <ash...@en...>: > Hi Raju, > There is a quite substantial (considering the age of this product) > material for Postgres-XC available at various places. > > Take a look at http://postgres-xc.sourceforge.net/, that's project homepage. > > The product documentation can be found at > > http://postgres-xc.sourceforge.net/docs/ > > You will find more material at > http://sourceforge.net/projects/postgres-xc/files/?source=navbar > > Hope that's helpful. > > On Sat, Jan 19, 2013 at 4:13 AM, Raju Angani <an...@gm...> wrote: >> Hi Ashutosh, >> >> This morning I came accross postgres-xc. >> I'm very much interested in learning about postgres-xc, could you please >> send me some pointers(pdf, details notes) >> or even better if you could point me to tutorial(implementation, >> maintenance, troubleshooting). >> >> I wanted to know the internals, I'm familiar with oracel RAC, and MSSQL >> HADR >> >> Appreciate your help! >> >> Thank you >> Raju > > > > -- > Best Wishes, > Ashutosh Bapat > EntepriseDB Corporation > The Enterprise Postgres Company > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. SALE $99.99 this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122412 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general |
|
From: Ashutosh B. <ash...@en...> - 2013-01-21 04:25:27
|
Hi Raju, There is a quite substantial (considering the age of this product) material for Postgres-XC available at various places. Take a look at http://postgres-xc.sourceforge.net/, that's project homepage. The product documentation can be found at http://postgres-xc.sourceforge.net/docs/ You will find more material at http://sourceforge.net/projects/postgres-xc/files/?source=navbar Hope that's helpful. On Sat, Jan 19, 2013 at 4:13 AM, Raju Angani <an...@gm...> wrote: > Hi Ashutosh, > > This morning I came accross postgres-xc. > I'm very much interested in learning about postgres-xc, could you please > send me some pointers(pdf, details notes) > or even better if you could point me to tutorial(implementation, > maintenance, troubleshooting). > > I wanted to know the internals, I'm familiar with oracel RAC, and MSSQL > HADR > > Appreciate your help! > > Thank you > Raju -- Best Wishes, Ashutosh Bapat EntepriseDB Corporation The Enterprise Postgres Company |
|
From: Koichi S. <koi...@gm...> - 2013-01-19 01:50:08
|
2013/1/18 Filip Rembiałkowski <fil...@gm...>: > What do you exactly mean by saying that triggers are not supported? > > Is it that in the context of single datanode, they will still work, > but the changes are not replicated over the whole XC cluster? Making > the cluster inconsistent? Not that simple. Depending upon trigger functions, some of them cannoot simply be invoked at datanode. The updated rows may have to be transfered to the coordinator for such trigger frunctions, if trigger functions are volatile or stable. Internally, we've tried several times to support triggers but we found restrictions which are not simple to describe. This time, we're trying not to bring any retrictions, though some of them may need further optimization. > > If the triggers are not supported, how come you passed all the > postgres regression tests? They're skipped in the regression. Regards; --- Koichi > > Thanks > > > > > > > On Thu, Jan 17, 2013 at 12:12 AM, Ashutosh Bapat > <ash...@en...> wrote: >> On Thu, Jan 17, 2013 at 9:26 AM, Filip Rembiałkowski >> <fil...@gm...> wrote: >>> Huh, these are really interesting questions :-) >>> >>> Let me ask them instead as a proof that userbase is interested. >>> >>> 1) Is there support for triggers in Postres-XC? Are there any >>> limitations? Known issues? >>> >> >> Triggers will be supported in next release; we are working on that. >> >>> 2) Does it make sense to run datanodes over distant links (think WAN >>> or intercontinental) ? >>> >> >> No, at least not in next few versions. XC being a scalability >> solution, WAN or intercontinental links do not help; they slow down >> things. As of now, XC is susceptible to network failures/glitches. >> Using WAN or intercontinental links might cause network failures or >> glitches, which will cause XC to behave erratically. We haven't tested >> XC in such environment, so can't describe exact problems it will face. >> >>> Thanks && Regards, >>> Filip Rembiałkowski >>> >>> >>> On Wed, Jan 16, 2013 at 9:15 PM, Michael Paquier <mi...@ot...> wrote: >>> >>>> In order to get answers to your questions, I highly recommend that you >>>> send your messages to the Postgres-XC mailing lists, in your case >>>> pos...@li.... >>> >>> ------------------------------------------------------------------------------ >>> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, >>> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current >>> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft >>> MVPs and experts. ON SALE this month only -- learn more at: >>> http://p.sf.net/sfu/learnmore_122712 >>> _______________________________________________ >>> Postgres-xc-general mailing list >>> Pos...@li... >>> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general >> >> >> >> -- >> Best Wishes, >> Ashutosh Bapat >> EntepriseDB Corporation >> The Enterprise Postgres Company > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general |
|
From: Filip R. <fil...@gm...> - 2013-01-17 15:23:29
|
What do you exactly mean by saying that triggers are not supported? Is it that in the context of single datanode, they will still work, but the changes are not replicated over the whole XC cluster? Making the cluster inconsistent? If the triggers are not supported, how come you passed all the postgres regression tests? Thanks On Thu, Jan 17, 2013 at 12:12 AM, Ashutosh Bapat <ash...@en...> wrote: > On Thu, Jan 17, 2013 at 9:26 AM, Filip Rembiałkowski > <fil...@gm...> wrote: >> Huh, these are really interesting questions :-) >> >> Let me ask them instead as a proof that userbase is interested. >> >> 1) Is there support for triggers in Postres-XC? Are there any >> limitations? Known issues? >> > > Triggers will be supported in next release; we are working on that. > >> 2) Does it make sense to run datanodes over distant links (think WAN >> or intercontinental) ? >> > > No, at least not in next few versions. XC being a scalability > solution, WAN or intercontinental links do not help; they slow down > things. As of now, XC is susceptible to network failures/glitches. > Using WAN or intercontinental links might cause network failures or > glitches, which will cause XC to behave erratically. We haven't tested > XC in such environment, so can't describe exact problems it will face. > >> Thanks && Regards, >> Filip Rembiałkowski >> >> >> On Wed, Jan 16, 2013 at 9:15 PM, Michael Paquier <mi...@ot...> wrote: >> >>> In order to get answers to your questions, I highly recommend that you >>> send your messages to the Postgres-XC mailing lists, in your case >>> pos...@li.... >> >> ------------------------------------------------------------------------------ >> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, >> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current >> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft >> MVPs and experts. ON SALE this month only -- learn more at: >> http://p.sf.net/sfu/learnmore_122712 >> _______________________________________________ >> Postgres-xc-general mailing list >> Pos...@li... >> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general > > > > -- > Best Wishes, > Ashutosh Bapat > EntepriseDB Corporation > The Enterprise Postgres Company |
|
From: Koichi S. <koi...@gm...> - 2013-01-17 14:44:36
|
Well, asynchronous replica is not that simple in XC if it is log-shipping and hot standby. Hot standby implements different means to provide consistent MVCC view and we need corresponding means, which could be a bit complicated than current GTM technique. Communication between coordinator/datanode via WAN will be a bottleneck. Why do you need to install servers in different locations? For disaster? For locality and management? We should expect such usecase in the future. If very few data is shared among subcluster, I think we can design a feasible architecture. Regards; ---------- Koichi Suzuki 2013/1/17 Nikhil Sontakke <ni...@st...>: > Hi Filip, > > >> 1) Is there support for triggers in Postres-XC? Are there any >> limitations? Known issues? >> > > I believe the support for this feature is being worked upon. But not > present in the current source base as of now. > >> 2) Does it make sense to run datanodes over distant links (think WAN >> or intercontinental) ? >> > > You can definitely place your datanodes wherever you want them. > However note that queries will be routed via the coordinator to them > and results will be brought back from datanodes to the coordinator. So > distant datanodes might not perform that well here. > > What might make more sense would be is to have async replicas (again > synchronous replicas might not help here) of these datanodes on these > distant links. > > So probably a tighter PGXC cluster with distance datanode async replicas. YMMV > > Regards, > Nikhils > > >> Thanks && Regards, >> Filip Rembiałkowski >> >> >> On Wed, Jan 16, 2013 at 9:15 PM, Michael Paquier <mi...@ot...> wrote: >> >>> In order to get answers to your questions, I highly recommend that you >>> send your messages to the Postgres-XC mailing lists, in your case >>> pos...@li.... >> >> ------------------------------------------------------------------------------ >> Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, >> MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current >> with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft >> MVPs and experts. ON SALE this month only -- learn more at: >> http://p.sf.net/sfu/learnmore_122712 >> _______________________________________________ >> Postgres-xc-general mailing list >> Pos...@li... >> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general > > > > -- > StormDB - http://www.stormdb.com > The Database Cloud > Postgres-XC Support and Service > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general |
|
From: Ashutosh B. <ash...@en...> - 2013-01-17 06:12:30
|
On Thu, Jan 17, 2013 at 9:26 AM, Filip Rembiałkowski <fil...@gm...> wrote: > Huh, these are really interesting questions :-) > > Let me ask them instead as a proof that userbase is interested. > > 1) Is there support for triggers in Postres-XC? Are there any > limitations? Known issues? > Triggers will be supported in next release; we are working on that. > 2) Does it make sense to run datanodes over distant links (think WAN > or intercontinental) ? > No, at least not in next few versions. XC being a scalability solution, WAN or intercontinental links do not help; they slow down things. As of now, XC is susceptible to network failures/glitches. Using WAN or intercontinental links might cause network failures or glitches, which will cause XC to behave erratically. We haven't tested XC in such environment, so can't describe exact problems it will face. > Thanks && Regards, > Filip Rembiałkowski > > > On Wed, Jan 16, 2013 at 9:15 PM, Michael Paquier <mi...@ot...> wrote: > >> In order to get answers to your questions, I highly recommend that you >> send your messages to the Postgres-XC mailing lists, in your case >> pos...@li.... > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general -- Best Wishes, Ashutosh Bapat EntepriseDB Corporation The Enterprise Postgres Company |
|
From: Nikhil S. <ni...@st...> - 2013-01-17 06:10:07
|
Hi Filip, > 1) Is there support for triggers in Postres-XC? Are there any > limitations? Known issues? > I believe the support for this feature is being worked upon. But not present in the current source base as of now. > 2) Does it make sense to run datanodes over distant links (think WAN > or intercontinental) ? > You can definitely place your datanodes wherever you want them. However note that queries will be routed via the coordinator to them and results will be brought back from datanodes to the coordinator. So distant datanodes might not perform that well here. What might make more sense would be is to have async replicas (again synchronous replicas might not help here) of these datanodes on these distant links. So probably a tighter PGXC cluster with distance datanode async replicas. YMMV Regards, Nikhils > Thanks && Regards, > Filip Rembiałkowski > > > On Wed, Jan 16, 2013 at 9:15 PM, Michael Paquier <mi...@ot...> wrote: > >> In order to get answers to your questions, I highly recommend that you >> send your messages to the Postgres-XC mailing lists, in your case >> pos...@li.... > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general -- StormDB - http://www.stormdb.com The Database Cloud Postgres-XC Support and Service |
|
From: Filip R. <fil...@gm...> - 2013-01-17 03:56:45
|
Huh, these are really interesting questions :-) Let me ask them instead as a proof that userbase is interested. 1) Is there support for triggers in Postres-XC? Are there any limitations? Known issues? 2) Does it make sense to run datanodes over distant links (think WAN or intercontinental) ? Thanks && Regards, Filip Rembiałkowski On Wed, Jan 16, 2013 at 9:15 PM, Michael Paquier <mi...@ot...> wrote: > In order to get answers to your questions, I highly recommend that you > send your messages to the Postgres-XC mailing lists, in your case > pos...@li.... |
|
From: Koichi S. <koi...@gm...> - 2013-01-16 04:56:36
|
You should be careful about max_connections and max_prepared_transactions. They should at least be (max_connections at coordinator) * (number of coordinators) respectively. Regards; ---------- Koichi Suzuki 2013/1/16 Ashutosh Bapat <ash...@en...>: > For datanodes, you can configure it much like PostgreSQL. But > coordinators are a bit tricky. > > If you have distributed your tables such that most of your queries are > being run on datanode, you can do with medium sized work mem. You > don't need much buffer space, except for catalog queries. Now, how do > we detect, how much of the query is being run on the datanode? In the > explain outputs of plans steps with labels "Data node scan" are > executed on datanodes, whereas any other steps are run at coordinator. > If you see a lot of steps being executed on the coordinator, you need > more memory and CPU resources at the coordinator. > > On Wed, Jan 16, 2013 at 4:49 AM, Michael Paquier > <mic...@gm...> wrote: >> >> >> On Wed, Jan 16, 2013 at 12:41 AM, Arni Sumarlidason >> <Arn...@md...> wrote: >>> >>> Good Morning Everyone, >>> >>> >>> >>> We are setting up a 25 node postgres-xc cluster, does anyone have suggests >>> on how much disk space a coordinator uses with its local database? >> >> Not that much, Coordinator contains only catalog information. I let at your >> appreciation the amount of space you think you need, but not gigs for sure. >> >>> Also hints regarding the memory section in postgresql.conf would be >>> awesome! >> >> I would do the settings of shared_buffers, work_mem, maintenance_work_mem >> and effective cache size on Datanodes as on normal PostgreSQL servers. >> For Coordinators, you don't need that much memory except if you have sort >> operations happening there, in this case adapt work_mem. You can also set up >> shared_buffers to put all the catalog information on-cache, but I do not >> think it is that much a win. >> >> Thanks, >> -- >> Michael Paquier >> http://michael.otacoo.com >> ------------------------------------------------------------------------------ >> Master SQL Server Development, Administration, T-SQL, SSAS, SSIS, SSRS >> and more. Get SQL Server skills now (including 2012) with LearnDevNow - >> 200+ hours of step-by-step video tutorials by Microsoft MVPs and experts. >> SALE $99.99 this month only - learn more at: >> http://p.sf.net/sfu/learnmore_122512 >> _______________________________________________ >> Postgres-xc-general mailing list >> Pos...@li... >> https://lists.sourceforge.net/lists/listinfo/postgres-xc-general >> > > > > -- > Best Wishes, > Ashutosh Bapat > EntepriseDB Corporation > The Enterprise Postgres Company > > ------------------------------------------------------------------------------ > Master Java SE, Java EE, Eclipse, Spring, Hibernate, JavaScript, jQuery > and much more. Keep your Java skills current with LearnJavaNow - > 200+ hours of step-by-step video tutorials by Java experts. > SALE $49.99 this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122612 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general |
|
From: Ashutosh B. <ash...@en...> - 2013-01-16 04:32:41
|
For datanodes, you can configure it much like PostgreSQL. But coordinators are a bit tricky. If you have distributed your tables such that most of your queries are being run on datanode, you can do with medium sized work mem. You don't need much buffer space, except for catalog queries. Now, how do we detect, how much of the query is being run on the datanode? In the explain outputs of plans steps with labels "Data node scan" are executed on datanodes, whereas any other steps are run at coordinator. If you see a lot of steps being executed on the coordinator, you need more memory and CPU resources at the coordinator. On Wed, Jan 16, 2013 at 4:49 AM, Michael Paquier <mic...@gm...> wrote: > > > On Wed, Jan 16, 2013 at 12:41 AM, Arni Sumarlidason > <Arn...@md...> wrote: >> >> Good Morning Everyone, >> >> >> >> We are setting up a 25 node postgres-xc cluster, does anyone have suggests >> on how much disk space a coordinator uses with its local database? > > Not that much, Coordinator contains only catalog information. I let at your > appreciation the amount of space you think you need, but not gigs for sure. > >> Also hints regarding the memory section in postgresql.conf would be >> awesome! > > I would do the settings of shared_buffers, work_mem, maintenance_work_mem > and effective cache size on Datanodes as on normal PostgreSQL servers. > For Coordinators, you don't need that much memory except if you have sort > operations happening there, in this case adapt work_mem. You can also set up > shared_buffers to put all the catalog information on-cache, but I do not > think it is that much a win. > > Thanks, > -- > Michael Paquier > http://michael.otacoo.com > ------------------------------------------------------------------------------ > Master SQL Server Development, Administration, T-SQL, SSAS, SSIS, SSRS > and more. Get SQL Server skills now (including 2012) with LearnDevNow - > 200+ hours of step-by-step video tutorials by Microsoft MVPs and experts. > SALE $99.99 this month only - learn more at: > http://p.sf.net/sfu/learnmore_122512 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general > -- Best Wishes, Ashutosh Bapat EntepriseDB Corporation The Enterprise Postgres Company |
|
From: Michael P. <mic...@gm...> - 2013-01-15 23:19:26
|
On Wed, Jan 16, 2013 at 12:41 AM, Arni Sumarlidason < Arn...@md...> wrote: > Good Morning Everyone,**** > > ** ** > > We are setting up a 25 node postgres-xc cluster, does anyone have suggests > on how much disk space a coordinator uses with its local database? > Not that much, Coordinator contains only catalog information. I let at your appreciation the amount of space you think you need, but not gigs for sure. **** > > Also hints regarding the memory section in postgresql.conf would be > awesome! > I would do the settings of shared_buffers, work_mem, maintenance_work_mem and effective cache size on Datanodes as on normal PostgreSQL servers. For Coordinators, you don't need that much memory except if you have sort operations happening there, in this case adapt work_mem. You can also set up shared_buffers to put all the catalog information on-cache, but I do not think it is that much a win. Thanks, -- Michael Paquier http://michael.otacoo.com |
|
From: Arni S. <Arn...@md...> - 2013-01-15 15:41:16
|
Good Morning Everyone, We are setting up a 25 node postgres-xc cluster, does anyone have suggests on how much disk space a coordinator uses with its local database? Also hints regarding the memory section in postgresql.conf would be awesome! Thank you everyone for your assistance over the past couple of weeks!! Arni Sumarlidason | Software Engineer, Information Technology MDA | 820 West Diamond Ave | Gaithersburg, MD | USA O: 240-833-8200 D: 240-833-8318 M: 256-393-2803 arn...@md...<mailto:arn...@md...>| http://www.mdaus.com<http://www.mdaus.com/> |
|
From: Filip R. <fil...@gm...> - 2013-01-14 21:08:03
|
subscribe |
|
From: Michael P. <mic...@gm...> - 2013-01-11 01:07:25
|
On Fri, Jan 11, 2013 at 10:01 AM, Koichi Suzuki <koi...@gm...>wrote: > Thanks Michael for the reply. > > Just for an info. > > In the next major release, I expect we will have node addition/removal > to the cluster on the fly in the next major release. The code is not > available at master yet though. > Just another information. Stuff has been designed, and this stuff is more like a set of tools that you can use to write your own scripts to customize node addition and removal in the cluster on the fly. -- Michael Paquier http://michael.otacoo.com |
|
From: Koichi S. <koi...@gm...> - 2013-01-11 01:01:48
|
Thanks Michael for the reply. Just for an info. In the next major release, I expect we will have node addition/removal to the cluster on the fly in the next major release. The code is not available at master yet though. Regards; ---------- Koichi Suzuki 2013/1/11 Michael Paquier <mic...@gm...>: > > > On Fri, Jan 11, 2013 at 6:41 AM, Adam McManus <ada...@gm...> > wrote: >> >> I've seen a few posts about the future ideas for adding a node to an >> existing cluster, but I'm curious if there are any best practices now? I've >> got a two-node cluster and would like to add a third. Any tips, gotchas, or >> info would be greatly appreciated. > > It is recommended to take a pg_dump of your cluster then rebuild a new > cluster with 3 nodes and pg_restore it for the time being. Note also that > the team is currently working on some tools necessary to help in doing > dynamic node addition like: > - the possibility to lock DDL and utilities to make all the node's schema > remain constant during a node addition > - the possibility to run pg_restore on a new node with any issues > > However, if you really want to try... You could also try to do the following > if you are sure that during the node addition there will be no DDL run (this > is really important to keep the schema consistent): > - initdb a new node > - Take a dump of a Datanode schema that will be used for the new Datanode, > or a Coordinator schema for a new Coordinator. This is important as there > are differences between the objects saved in Da and Co catalogs, like views > and sequences are only created on Coordinators. > - Restore the dump to the new node to make its schema data consistent with > the other nodes. Please note that there might be issues around this area > like the following like new node might complain that there is no Datanode > registered when issuing the restore, so the restore will fail. There might > be some ways to trick the newly-added node by running the restore in > single-user mode. Honestly I never tested myself. > - Then issue all the necessary CREATE NODE commands on all the Coordinators > with pgxc_pool_reload(). This consists in adding the new node info in the > fomer Coordinators, and adding all the other node information on the new > node if it is a Coordinator. > -- > Michael Paquier > http://michael.otacoo.com > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general > |
|
From: Michael P. <mic...@gm...> - 2013-01-10 23:53:50
|
On Fri, Jan 11, 2013 at 6:41 AM, Adam McManus <ada...@gm...>wrote: > I've seen a few posts about the future ideas for adding a node to an > existing cluster, but I'm curious if there are any best practices now? I've > got a two-node cluster and would like to add a third. Any tips, gotchas, or > info would be greatly appreciated. It is recommended to take a pg_dump of your cluster then rebuild a new cluster with 3 nodes and pg_restore it for the time being. Note also that the team is currently working on some tools necessary to help in doing dynamic node addition like: - the possibility to lock DDL and utilities to make all the node's schema remain constant during a node addition - the possibility to run pg_restore on a new node with any issues However, if you really want to try... You could also try to do the following if you are sure that during the node addition there will be no DDL run (this is really important to keep the schema consistent): - initdb a new node - Take a dump of a Datanode schema that will be used for the new Datanode, or a Coordinator schema for a new Coordinator. This is important as there are differences between the objects saved in Da and Co catalogs, like views and sequences are only created on Coordinators. - Restore the dump to the new node to make its schema data consistent with the other nodes. Please note that there might be issues around this area like the following like new node might complain that there is no Datanode registered when issuing the restore, so the restore will fail. There might be some ways to trick the newly-added node by running the restore in single-user mode. Honestly I never tested myself. - Then issue all the necessary CREATE NODE commands on all the Coordinators with pgxc_pool_reload(). This consists in adding the new node info in the fomer Coordinators, and adding all the other node information on the new node if it is a Coordinator. -- Michael Paquier http://michael.otacoo.com |
|
From: Adam M. <ada...@gm...> - 2013-01-10 21:41:26
|
I've seen a few posts about the future ideas for adding a node to an existing cluster, but I'm curious if there are any best practices now? I've got a two-node cluster and would like to add a third. Any tips, gotchas, or info would be greatly appreciated. Thanks, -- Adam McManus |
|
From: Andrei M. <and...@gm...> - 2013-01-10 20:01:01
|
I guess you are using last released Postgres-XC code, correct? The patch may be not compatible with that version, it was only tested with git HEAD at the time of writing the tutorial. I would suggest to check out current Postgres-XC code. Or let us some time to investigate the issue and provide a fix. 2013/1/10 Arni Sumarlidason <Arn...@md...> > Thank you for everyone’s help last night.. I resolved the library issue > with your help.**** > > ** ** > > However, I’m getting a segment fault now, please see the following image,* > *** > > http://sumarlidason.com/perm/2013.01.10/pgxc.png**** > > ** ** > > Again, thank you for your time,**** > > ** ** > > *From:* Jim Mlodgenski [mailto:ji...@gm...] > *Sent:* Wednesday, January 09, 2013 7:47 PM > *To:* Arni Sumarlidason > *Cc:* pos...@li... > *Subject:* Re: [Postgres-xc-general] xc&gis**** > > ** ** > > > > On Wednesday, January 9, 2013, Arni Sumarlidason wrote:**** > > Good Evening,**** > > **** > > I followed the tutorial you provided ( thank you again ) and I encounter > the following error while trying to CREATE EXTENSION:**** > > http://sumarlidason.com/perm/2013.01.09/xcgis.PNG<https://console.mxlogic.com/redir/?zC763hOYqejhOrv7cIcECzAQsCM0t6Rtyv93qlxO-bv5n4XEIZt7HCzD0Y1as5wo4zt-hojuv78I9CzATsS8Ollb3D0kh4-ndEzC4NNJ5BYse7ffcL6SkPobZ8Qg6BIblFcz7W5M-NszYfzFVFtd40vpEzaxKvxYY1NJ4SCrpop73zhO-UrPVjP> > **** > > **** > > I thought maybe permissions issue, so I chown’ed the entire /usr/local/lib > folder, to no avail. Any guidance would be appreciated, I included my build > details below[1].**** > > Does it show as missing when you run an ldd on the geo library? > /usr/local/lib may not be in the library path. Try setting LD_LIBRARY_PATH > to include the directory **** > > **** > > Thank you so much,**** > > **** > > *Arni Sumarlidason | Web Developer, Information Technology***** > > MDA | 820 West Diamond Ave | Gaithersburg, MD | USA**** > > O: 240-833-8200 D: 240-833-8318 M: 256-393-2803**** > > arn...@md...| http://www.mdaus.com<https://console.mxlogic.com/redir/?zC763hOYqejhOrv7cIcECzAQsCM0iG6JQVv3t-hojuv78I9CzATsS8Ollb3D0kh4-ndEzC4NNJ5BYse7ffcL6SkPobZ8Qg6BIblFcz7W5M-NszYfzFVFtd40vpEzaxKvxYY1NJcSCrpop73zhO-Urd9yn> > **** > > **** > > **** > > **** > > [1]**** > > ## update the system**** > > yum -y update**** > > **** > > ## install dev tools**** > > yum -y groupinstall "Development tools"**** > > **** > > ## Install the following packages**** > > yum -y install \**** > > wget-1.12-1.4.el6.x86_64 \**** > > ncurses-devel-5.7-3.20090208.el6.x86_64 \**** > > readline-devel-6.0-4.el6.x86_64 \**** > > zlib-devel-1.2.3-27.el6.x86_64 \**** > > libxml2-devel-2.7.6-8.el6_3.4.x86_64 \**** > > pcre-devel-7.8-4.el6.x86_64 \**** > > compat-readline5-5.2-17.1.el6.x86_64 \**** > > ruby-libs-1.8.7.352-7.el6_2.x86_64 \**** > > ruby-1.8.7.352-7.el6_2.x86_64 \**** > > ruby-devel-1.8.7.352-7.el6_2.x86_64 \**** > > xerces-c-3.0.1-20.el6.x86_64 \**** > > libidn-devel-1.18-2.el6.x86_64 \**** > > libcurl-devel-7.19.7-26.el6_2.4.x86_64 \**** > > xerces-c-devel-3.0.1-20.el6.x86_64 \**** > > expat-devel-2.0.1-11.el6_2.x86_64 \**** > > python-devel-2.6.6-29.el6_3.3.x86_64 \**** > > gmp-devel-4.3.1-7.el6_2.2.x86_64 \**** > > ppl-devel-0.10.2-11.el6.x86_64 \**** > > screen-4.0.3-16.el6.x86_64 \**** > > nano-2.0.9-7.el6.x86_64 \**** > > man**** > > **** > > ## Disable SELINX**** > > echo -e "SELINUX=disabled\nSELINUXTYPE=targeted" > /etc/selinux/config**** > > **** > > **** > > ## Turn off & Disable firewalls**** > > service iptables stop**** > > service ip6tables stop**** > > chkconfig iptables off**** > > chkconfig ip6tables off**** > > **** > > ## Extract all source files**** > > tar xzvf swig-2.0.9.tar.gz**** > > tar xzvf proj-4.8.0.tar.gz**** > > tar jxvf postgis-2.0.2.tar.gz**** > > tar xzvf pgxc-v1.0.1.tar.gz**** > > tar xzvf geos-3.3.6.tar.bz2**** > > tar xzvf gdal-1.9.2.tar.gz**** > > tar xzvf FileGDB_API_1_3-64.tar.gz**** > > **** > > **** > > ## Build swig**** > > cd swig-2.0.9**** > > ./configure**** > > make**** > > make install**** > > cd ..**** > > **** > > ## Build Proj4**** > > cd proj-4.8.0**** > > ./configure**** > > make**** > > make install**** > > cd ..**** > > **** > > ## Build Geos**** > > cd geos-3.3.6**** > > ./configure --enable-ruby**** > > **** > > ## Configure for GEOS ends like this:**** > > ## Swig: true**** > > ## Python bindings: false**** > > ## Ruby bindings: true**** > > ## PHP bindings: false**** > > **** > > **** > > make**** > > make install**** > > cd ..**** > > **** > > ## Build Postgres-XC**** > > cd pgxc**** > > patch -p1 < /root/Downloads/postgis_xc.patch**** > > ./configure --prefix=/opt/PostgresXC**** > > make**** > > make install**** > > cd ..**** > > **** > > ## setup LD to point to FileGDB_API's library files**** > > echo /root/Downloads/FileGDB_API/lib > /etc/ld.so.conf.d/fgdb.conf**** > > **** > > **** > > ## Building: GDAL**** > > cd gdal-1.9.2**** > > ./configure \**** > > --with-pgconfig=/opt/PostgresXC/bin/pg_config \**** > > --with-libz=internal --with-curl=/usr/bin/curl-config > --with-fgdb=/root/Downloads/FileGDB_API \**** > > --with-png=internal --with-libtiff=internal --with-geotiff=internal \* > *** > > --with-jpeg=internal --with-gif=internal --with-geos=yes \**** > > --with-sqlite3=/usr/local/lib \**** > > --with-threads=yes --with-poppler=yes --with-python --with-liblzma=yes > \**** > > --with-xerces=yes --with-expat=yes**** > > make**** > > ## Get Coffee..............**** > > make install**** > > cd ..**** > > **** > > ## Copy over missing source data**** > > cp -R /root/Downloads/pgxc/src/include/gtm > /opt/PostgresXC/include/postgresql/server/gtm**** > > **** > > ## Building: PostGIS**** > > cd postgis-2.0.2**** > > ./configure --with-pgconfig=/opt/PostgresXC/bin/pg_config**** > > make**** > > make install**** > > cd ..**** > > > > ------------------------------------------------------------------------------ > Master Visual Studio, SharePoint, SQL, ASP.NET, C# 2012, HTML5, CSS, > MVC, Windows 8 Apps, JavaScript and much more. Keep your skills current > with LearnDevNow - 3,200 step-by-step video tutorials by Microsoft > MVPs and experts. ON SALE this month only -- learn more at: > http://p.sf.net/sfu/learnmore_122712 > _______________________________________________ > Postgres-xc-general mailing list > Pos...@li... > https://lists.sourceforge.net/lists/listinfo/postgres-xc-general > > -- Andrei Martsinchyk StormDB - http://www.stormdb.com The Database Cloud |
|
From: Arni S. <Arn...@md...> - 2013-01-10 14:43:35
|
Thank you for everyone's help last night.. I resolved the library issue with your help. However, I'm getting a segment fault now, please see the following image, http://sumarlidason.com/perm/2013.01.10/pgxc.png Again, thank you for your time, From: Jim Mlodgenski [mailto:ji...@gm...] Sent: Wednesday, January 09, 2013 7:47 PM To: Arni Sumarlidason Cc: pos...@li... Subject: Re: [Postgres-xc-general] xc&gis On Wednesday, January 9, 2013, Arni Sumarlidason wrote: Good Evening, I followed the tutorial you provided ( thank you again ) and I encounter the following error while trying to CREATE EXTENSION: http://sumarlidason.com/perm/2013.01.09/xcgis.PNG<https://console.mxlogic.com/redir/?zC763hOYqejhOrv7cIcECzAQsCM0t6Rtyv93qlxO-bv5n4XEIZt7HCzD0Y1as5wo4zt-hojuv78I9CzATsS8Ollb3D0kh4-ndEzC4NNJ5BYse7ffcL6SkPobZ8Qg6BIblFcz7W5M-NszYfzFVFtd40vpEzaxKvxYY1NJ4SCrpop73zhO-UrPVjP> I thought maybe permissions issue, so I chown'ed the entire /usr/local/lib folder, to no avail. Any guidance would be appreciated, I included my build details below[1]. Does it show as missing when you run an ldd on the geo library? /usr/local/lib may not be in the library path. Try setting LD_LIBRARY_PATH to include the directory Thank you so much, Arni Sumarlidason | Web Developer, Information Technology MDA | 820 West Diamond Ave | Gaithersburg, MD | USA O: 240-833-8200 D: 240-833-8318 M: 256-393-2803 arn...@md...<javascript:_e(%7b%7d,%20'cvml',%20'arn...@md...');>| http://www.mdaus.com<https://console.mxlogic.com/redir/?zC763hOYqejhOrv7cIcECzAQsCM0iG6JQVv3t-hojuv78I9CzATsS8Ollb3D0kh4-ndEzC4NNJ5BYse7ffcL6SkPobZ8Qg6BIblFcz7W5M-NszYfzFVFtd40vpEzaxKvxYY1NJcSCrpop73zhO-Urd9yn> [1] ## update the system yum -y update ## install dev tools yum -y groupinstall "Development tools" ## Install the following packages yum -y install \ wget-1.12-1.4.el6.x86_64 \ ncurses-devel-5.7-3.20090208.el6.x86_64 \ readline-devel-6.0-4.el6.x86_64 \ zlib-devel-1.2.3-27.el6.x86_64 \ libxml2-devel-2.7.6-8.el6_3.4.x86_64 \ pcre-devel-7.8-4.el6.x86_64 \ compat-readline5-5.2-17.1.el6.x86_64 \ ruby-libs-1.8.7.352-7.el6_2.x86_64 \ ruby-1.8.7.352-7.el6_2.x86_64 \ ruby-devel-1.8.7.352-7.el6_2.x86_64 \ xerces-c-3.0.1-20.el6.x86_64 \ libidn-devel-1.18-2.el6.x86_64 \ libcurl-devel-7.19.7-26.el6_2.4.x86_64 \ xerces-c-devel-3.0.1-20.el6.x86_64 \ expat-devel-2.0.1-11.el6_2.x86_64 \ python-devel-2.6.6-29.el6_3.3.x86_64 \ gmp-devel-4.3.1-7.el6_2.2.x86_64 \ ppl-devel-0.10.2-11.el6.x86_64 \ screen-4.0.3-16.el6.x86_64 \ nano-2.0.9-7.el6.x86_64 \ man ## Disable SELINX echo -e "SELINUX=disabled\nSELINUXTYPE=targeted" > /etc/selinux/config ## Turn off & Disable firewalls service iptables stop service ip6tables stop chkconfig iptables off chkconfig ip6tables off ## Extract all source files tar xzvf swig-2.0.9.tar.gz tar xzvf proj-4.8.0.tar.gz tar jxvf postgis-2.0.2.tar.gz tar xzvf pgxc-v1.0.1.tar.gz tar xzvf geos-3.3.6.tar.bz2 tar xzvf gdal-1.9.2.tar.gz tar xzvf FileGDB_API_1_3-64.tar.gz ## Build swig cd swig-2.0.9 ./configure make make install cd .. ## Build Proj4 cd proj-4.8.0 ./configure make make install cd .. ## Build Geos cd geos-3.3.6 ./configure --enable-ruby ## Configure for GEOS ends like this: ## Swig: true ## Python bindings: false ## Ruby bindings: true ## PHP bindings: false make make install cd .. ## Build Postgres-XC cd pgxc patch -p1 < /root/Downloads/postgis_xc.patch ./configure --prefix=/opt/PostgresXC make make install cd .. ## setup LD to point to FileGDB_API's library files echo /root/Downloads/FileGDB_API/lib > /etc/ld.so.conf.d/fgdb.conf ## Building: GDAL cd gdal-1.9.2 ./configure \ --with-pgconfig=/opt/PostgresXC/bin/pg_config \ --with-libz=internal --with-curl=/usr/bin/curl-config --with-fgdb=/root/Downloads/FileGDB_API \ --with-png=internal --with-libtiff=internal --with-geotiff=internal \ --with-jpeg=internal --with-gif=internal --with-geos=yes \ --with-sqlite3=/usr/local/lib \ --with-threads=yes --with-poppler=yes --with-python --with-liblzma=yes \ --with-xerces=yes --with-expat=yes make ## Get Coffee.............. make install cd .. ## Copy over missing source data cp -R /root/Downloads/pgxc/src/include/gtm /opt/PostgresXC/include/postgresql/server/gtm ## Building: PostGIS cd postgis-2.0.2 ./configure --with-pgconfig=/opt/PostgresXC/bin/pg_config make make install cd .. |
|
From: Jim M. <ji...@gm...> - 2013-01-10 00:47:26
|
On Wednesday, January 9, 2013, Arni Sumarlidason wrote: > Good Evening,**** > > ** ** > > I followed the tutorial you provided ( thank you again ) and I encounter > the following error while trying to CREATE EXTENSION:**** > > http://sumarlidason.com/perm/2013.01.09/xcgis.PNG**** > > ** ** > > I thought maybe permissions issue, so I chown’ed the entire /usr/local/lib > folder, to no avail. Any guidance would be appreciated, I included my build > details below[1].**** > Does it show as missing when you run an ldd on the geo library? /usr/local/lib may not be in the library path. Try setting LD_LIBRARY_PATH to include the directory > ** ** > > Thank you so much,**** > > ** ** > > *Arni Sumarlidason | Web Developer, Information Technology***** > > MDA | 820 West Diamond Ave | Gaithersburg, MD | USA**** > > O: 240-833-8200 D: 240-833-8318 M: 256-393-2803**** > > arn...@md... <javascript:_e({}, 'cvml', > 'arn...@md...');>| http://www.mdaus.com **** > > ** ** > > ** ** > > ** ** > > [1]**** > > ## update the system**** > > yum -y update**** > > ** ** > > ## install dev tools**** > > yum -y groupinstall "Development tools"**** > > ** ** > > ## Install the following packages**** > > yum -y install \**** > > wget-1.12-1.4.el6.x86_64 \**** > > ncurses-devel-5.7-3.20090208.el6.x86_64 \**** > > readline-devel-6.0-4.el6.x86_64 \**** > > zlib-devel-1.2.3-27.el6.x86_64 \**** > > libxml2-devel-2.7.6-8.el6_3.4.x86_64 \**** > > pcre-devel-7.8-4.el6.x86_64 \**** > > compat-readline5-5.2-17.1.el6.x86_64 \**** > > ruby-libs-1.8.7.352-7.el6_2.x86_64 \**** > > ruby-1.8.7.352-7.el6_2.x86_64 \**** > > ruby-devel-1.8.7.352-7.el6_2.x86_64 \**** > > xerces-c-3.0.1-20.el6.x86_64 \**** > > libidn-devel-1.18-2.el6.x86_64 \**** > > libcurl-devel-7.19.7-26.el6_2.4.x86_64 \**** > > xerces-c-devel-3.0.1-20.el6.x86_64 \**** > > expat-devel-2.0.1-11.el6_2.x86_64 \**** > > python-devel-2.6.6-29.el6_3.3.x86_64 \**** > > gmp-devel-4.3.1-7.el6_2.2.x86_64 \**** > > ppl-devel-0.10.2-11.el6.x86_64 \**** > > screen-4.0.3-16.el6.x86_64 \**** > > nano-2.0.9-7.el6.x86_64 \**** > > man**** > > ** ** > > ## Disable SELINX**** > > echo -e "SELINUX=disabled\nSELINUXTYPE=targeted" > /etc/selinux/config**** > > ** ** > > ** ** > > ## Turn off & Disable firewalls**** > > service iptables stop**** > > service ip6tables stop**** > > chkconfig iptables off**** > > chkconfig ip6tables off**** > > ** ** > > ## Extract all source files**** > > tar xzvf swig-2.0.9.tar.gz**** > > tar xzvf proj-4.8.0.tar.gz**** > > tar jxvf postgis-2.0.2.tar.gz**** > > tar xzvf pgxc-v1.0.1.tar.gz**** > > tar xzvf geos-3.3.6.tar.bz2**** > > tar xzvf gdal-1.9.2.tar.gz**** > > tar xzvf FileGDB_API_1_3-64.tar.gz**** > > ** ** > > ** ** > > ## Build swig**** > > cd swig-2.0.9**** > > ./configure**** > > make**** > > make install**** > > cd ..**** > > ** ** > > ## Build Proj4**** > > cd proj-4.8.0**** > > ./configure**** > > make**** > > make install**** > > cd ..**** > > ** ** > > ## Build Geos**** > > cd geos-3.3.6**** > > ./configure --enable-ruby**** > > ** ** > > ## Configure for GEOS ends like this:**** > > ## Swig: true**** > > ## Python bindings: false**** > > ## Ruby bindings: true**** > > ## PHP bindings: false**** > > ** ** > > ** ** > > make**** > > make install**** > > cd ..**** > > ** ** > > ## Build Postgres-XC**** > > cd pgxc**** > > patch -p1 < /root/Downloads/postgis_xc.patch**** > > ./configure --prefix=/opt/PostgresXC**** > > make**** > > make install**** > > cd ..**** > > ** ** > > ## setup LD to point to FileGDB_API's library files**** > > echo /root/Downloads/FileGDB_API/lib > /etc/ld.so.conf.d/fgdb.conf**** > > ** ** > > ** ** > > ## Building: GDAL**** > > cd gdal-1.9.2**** > > ./configure \**** > > --with-pgconfig=/opt/PostgresXC/bin/pg_config \**** > > --with-libz=internal --with-curl=/usr/bin/curl-config > --with-fgdb=/root/Downloads/FileGDB_API \**** > > --with-png=internal --with-libtiff=internal --with-geotiff=internal \* > *** > > --with-jpeg=internal --with-gif=internal --with-geos=yes \**** > > --with-sqlite3=/usr/local/lib \**** > > --with-threads=yes --with-poppler=yes --with-python --with-liblzma=yes > \**** > > --with-xerces=yes --with-expat=yes**** > > make**** > > ## Get Coffee..............**** > > make install**** > > cd ..**** > > ** ** > > ## Copy over missing source data**** > > cp -R /root/Downloads/pgxc/src/include/gtm > /opt/PostgresXC/include/postgresql/server/gtm**** > > ** ** > > ## Building: PostGIS**** > > cd postgis-2.0.2**** > > ./configure --with-pgconfig=/opt/PostgresXC/bin/pg_config**** > > make**** > > make install**** > > cd ..**** > |
|
From: Devrim G. <de...@gu...> - 2013-01-10 00:12:44
|
Hi, Are you sure that you ran ls in the directory that XC is looking for? The dir names does not seem to match between left and right. Regards, Arni Sumarlidason <Arn...@md...> wrote: >Good Evening, > >I followed the tutorial you provided ( thank you again ) and I >encounter the following error while trying to CREATE EXTENSION: >http://sumarlidason.com/perm/2013.01.09/xcgis.PNG > >I thought maybe permissions issue, so I chown'ed the entire >/usr/local/lib folder, to no avail. Any guidance would be appreciated, >I included my build details below[1]. > >Thank you so much, > >Arni Sumarlidason | Web Developer, Information Technology >MDA | 820 West Diamond Ave | Gaithersburg, MD | USA >O: 240-833-8200 D: 240-833-8318 M: 256-393-2803 >arn...@md...<mailto:arn...@md...>| >http://www.mdaus.com<http://www.mdaus.com/> > > > >[1] >## update the system >yum -y update > >## install dev tools >yum -y groupinstall "Development tools" > >## Install the following packages >yum -y install \ >wget-1.12-1.4.el6.x86_64 \ >ncurses-devel-5.7-3.20090208.el6.x86_64 \ >readline-devel-6.0-4.el6.x86_64 \ >zlib-devel-1.2.3-27.el6.x86_64 \ >libxml2-devel-2.7.6-8.el6_3.4.x86_64 \ >pcre-devel-7.8-4.el6.x86_64 \ >compat-readline5-5.2-17.1.el6.x86_64 \ >ruby-libs-1.8.7.352-7.el6_2.x86_64 \ >ruby-1.8.7.352-7.el6_2.x86_64 \ >ruby-devel-1.8.7.352-7.el6_2.x86_64 \ >xerces-c-3.0.1-20.el6.x86_64 \ >libidn-devel-1.18-2.el6.x86_64 \ >libcurl-devel-7.19.7-26.el6_2.4.x86_64 \ >xerces-c-devel-3.0.1-20.el6.x86_64 \ >expat-devel-2.0.1-11.el6_2.x86_64 \ >python-devel-2.6.6-29.el6_3.3.x86_64 \ >gmp-devel-4.3.1-7.el6_2.2.x86_64 \ >ppl-devel-0.10.2-11.el6.x86_64 \ >screen-4.0.3-16.el6.x86_64 \ >nano-2.0.9-7.el6.x86_64 \ >man > >## Disable SELINX >echo -e "SELINUX=disabled\nSELINUXTYPE=targeted" > /etc/selinux/config > > >## Turn off & Disable firewalls >service iptables stop >service ip6tables stop >chkconfig iptables off >chkconfig ip6tables off > >## Extract all source files >tar xzvf swig-2.0.9.tar.gz >tar xzvf proj-4.8.0.tar.gz >tar jxvf postgis-2.0.2.tar.gz >tar xzvf pgxc-v1.0.1.tar.gz >tar xzvf geos-3.3.6.tar.bz2 >tar xzvf gdal-1.9.2.tar.gz >tar xzvf FileGDB_API_1_3-64.tar.gz > > >## Build swig >cd swig-2.0.9 >./configure >make >make install >cd .. > >## Build Proj4 >cd proj-4.8.0 >./configure >make >make install >cd .. > >## Build Geos >cd geos-3.3.6 >./configure --enable-ruby > >## Configure for GEOS ends like this: >## Swig: true >## Python bindings: false >## Ruby bindings: true >## PHP bindings: false > > >make >make install >cd .. > >## Build Postgres-XC >cd pgxc >patch -p1 < /root/Downloads/postgis_xc.patch >./configure --prefix=/opt/PostgresXC >make >make install >cd .. > >## setup LD to point to FileGDB_API's library files >echo /root/Downloads/FileGDB_API/lib > /etc/ld.so.conf.d/fgdb.conf > > >## Building: GDAL >cd gdal-1.9.2 >./configure \ > --with-pgconfig=/opt/PostgresXC/bin/pg_config \ >--with-libz=internal --with-curl=/usr/bin/curl-config >--with-fgdb=/root/Downloads/FileGDB_API \ > --with-png=internal --with-libtiff=internal --with-geotiff=internal \ > --with-jpeg=internal --with-gif=internal --with-geos=yes \ > --with-sqlite3=/usr/local/lib \ >--with-threads=yes --with-poppler=yes --with-python --with-liblzma=yes >\ > --with-xerces=yes --with-expat=yes >make >## Get Coffee.............. >make install >cd .. > >## Copy over missing source data >cp -R /root/Downloads/pgxc/src/include/gtm >/opt/PostgresXC/include/postgresql/server/gtm > >## Building: PostGIS >cd postgis-2.0.2 >./configure --with-pgconfig=/opt/PostgresXC/bin/pg_config >make >make install >cd .. > > >------------------------------------------------------------------------ > >------------------------------------------------------------------------------ >Master Java SE, Java EE, Eclipse, Spring, Hibernate, JavaScript, jQuery >and much more. Keep your Java skills current with LearnJavaNow - >200+ hours of step-by-step video tutorials by Java experts. >SALE $49.99 this month only -- learn more at: >http://p.sf.net/sfu/learnmore_122612 > >------------------------------------------------------------------------ > >_______________________________________________ >Postgres-xc-general mailing list >Pos...@li... >https://lists.sourceforge.net/lists/listinfo/postgres-xc-general -- Sent from my Android phone with K-9 Mail. Please excuse my brevity. |
|
From: Michael P. <mic...@gm...> - 2013-01-09 23:55:39
|
On Thu, Jan 10, 2013 at 8:36 AM, Arni Sumarlidason < Arn...@md...> wrote: > Good Evening,**** > > ** ** > > I followed the tutorial you provided ( thank you again ) and I encounter > the following error while trying to CREATE EXTENSION:**** > > http://sumarlidason.com/perm/2013.01.09/xcgis.PNG**** > > ** ** > > I thought maybe permissions issue, so I chown’ed the entire /usr/local/lib > folder, to no avail. Any guidance would be appreciated, I included my build > details below[1]. > I am not sure about the cause of the problem. The authorization looks correct. Does the error show up on Coordinator or Datanode pg_log? -- Michael Paquier http://michael.otacoo.com |
3 messages has been excluded from this view by a project administrator.
