Latest topics for the forum "GridSQL - General"

Direct update / insert to nodes

felixng — GMT

Our system have constantly heavy insert (and some updates) to the 'big' tables. Can I modify the data directly on the nodes? The tables I need to operate does not use any sequence or auto-increment keys.

Selling : Becker Mexico 7948 Automotive GPS Receiver .... $3,000

thalestim — GMT

The goal of Wise Buy® is to combine all the facets necessary for someone who wants to buy electronic gadgets , Car Audio, Car Alarms , Car Security , Becker Cascade, Becker Cascade Pro, Becker Mexico, Becker Indianapolis Pro, Becker Grand Prix, Becker iPod remote kit , Bazooka Bass Tubes, Active Bass Tubes, Passive Bass Tubes, Genesis Profile Amplifiers, Genesis Series 3 Amplifiers, Rainbow Speakers, Rockford Fosgate, Focal Speakers , AudioControl Processors, Parrot Hands Free Car Kits, Bluetooth Car Kits, iPod integration, MB Quart speakers, DAB radio, Roberts Radio, Wi-Fi radio, valve amplifiers, Grado Headphones at cheap prices, quality products, and efficient service.

You can always count on getting the latest models and styles from Wise Buy . WB has a full line of GSM Unlocked phones , Laptops and Tablet Pc .®

We also have all types of accessories to match your gadgets .

Our goal is to ensure customer satisfaction.
This is achieved by allowing the customer to browse our easy-to-navigate blog and making the purchasing process safe and simple.

We offer the most competitive prices for wireless technology while still maintaining excellent service standards.®

All orders will be handled with the utmost care and precision by our customer-oriented employees, on hand to assist you during operation hours.
You can feel secure in knowing that all transactions are fully protected through the use of digital encryption technology.

Shopping for electronic gadgets & Accessories is made easy with wisebuy.com® !

Becker Car Radio®

BECKER CAR RADIO MEXICO FULLSTEREO R107 SL W123 (USED) .......... $700
Ferrari Becker Europa Radio_Cassette Player_Amplifier (USED)............ $500
MERCEDES BECKER EUROPA AND CASSETTE 280 250 SE 450 SL (USED) ....... $650
MERCEDES PORSCHE BECKER EUROPA TG SLIM FACE RADIO TUNER ....... $850
1961-1971 Mercedes W111 Becker Europa TR radio Working ........... $1200
Mercedes 300b, c, d Adenauer Radio_ Becker Mexico Radio ............ $1200
ORIGINAL Becker Radio for Porsche 356A .............. $350
ORIGINAL PORSCHE 356 STONE GUARDS FOR HELLA BOSCH HEADLIGHTS LAMPS ...... $850
MERCEDES 111 220 250 280 300 SE BECKER GRAND PRIX RADIO (USED )............ $1400
Porsche Cayenne 04 AM/FM/CD Factory Harman Becker Radio (USED ) ........... $450
Becker Mexico 7948 Automotive GPS Receiver ................... $3000

JL Audio HD750/1..........$350
JL Audio 12W7-3.........$340
JL Audio CLS112RG-W7......$470
JL Audio Slash v2 Series 10001v2...$500
JL Audio Slash v2 Series 5001v2..$220
JVC KD-NX5000 DVD receiver....$370
JVC KW-NX7000BT (KW-NX7000 with KS-BTA200)......$500
JVC KW-NX7000 with KS-BTA200 and KS-U30 Included...$650
JVC KW-NX7000 R/B.........$550
JVC KW-AVX720..........$320

Pioneer AVIC Z130BT.....$600
Pioneer AVIC-X710BT........$400
Pioneer AVH-P3100DVD.....$350
Pioneer AVH-P4100DVD....$310
Pioneer AVIC-Z110BT....$700
Pioneer AVIC-Z120BT...$710
Pioneer AVIC-X910BT.....$450
Pioneer AVIC-X920BT......$650
Pioneer AVIC-U310BT......$300
Pioneer AVH-P7800DVD...$550
Pioneer AVH P4900DVD .$450
Pioneer AVIC-Z1 - … $500
Pioneer AVIC Z2 - .$600
Pioneer AVIC-N4 - $800

Clarion NX700..........$600
Clarion VX709 + Clarion NP509 Package.......$580
Clarion NZ409 (NZ-409)......$300
Clarion NX509 (NX-509)$470
Clarion NX409 (NX-409)....$400
Clarion VX709 (VX-709)......$430
Clarion DRZ9255 Car Audio CD/CD-R/CD-RW Receiver...$320
Clarion VX709 + NP509 + CCB509 Package.....$470

Kenwood eXcelon DNX9140 (DNX-9140)....$600
Kenwood KVT-514..........$370
Kenwood DDX714 (DDX-714).......$520
Kenwood DDX714 + Kenwood KNA-G510 Package.....650
Kenwood KVT-617DVD with 7 video screen...$250
Kenwood KVT-819DVD...$450
Kenwood DDX616 (DDX-616).....$230

Alpine IVA-D105...$270
Alpine IVA-D310 7 LCD monitor...$320
Alpine IVA-W205....$300

Mobile Phones®

Nokia Lumia 800 ........................ $400
Nokia N9 .................................... $320

Apple iPhone 4 S 16GB (Factory Unlocked) - Black .......... $480
Apple iPhone 4 S 32GB (Factory Unlocked) - Black .......... $530
Apple iPhone 4 S 64GB (Factory Unlocked) - Black ........... $600
Apple iPhone 4 16GB (Factory Unlocked) - Black ........... $430
Apple iPhone 4 16GB (Factory Unlocked) - White ........... $430
Apple iPhone 4 32GB (Factory Unlocked) - Black ............ $480
Apple iPhone 4 32GB (Factory Unlocked) - White ............ $480

Dell Venue ....... $230
Dell Streak ....... $300
Dell Streak (3G 850/1900/2100 MHz) ..... $370
Dell Venue Pro ...... $440
Dell Venue Pro (3G 850/1900 MHz) ...... $450

HTC Desire A8181 ....... $280
HTC HD7 ....... $289
HTC Desire Z A7275 (3G 850/1900 MHz) ....... $300
HTC EVO 3D (3G 850/900/2100) ............ $400
HTC Incredible S ........... $380
HTC Desire HD A9192 (3G 850/1900 MHz) ...... $350
HTC Sensation 4G Z710a (3G 850/1900 MHz) ...... $420
HTC Sensation XL ( white ) ..................... .... $480
HTC Sensation Xe ...................... $350

LG P990h Optimus 2X (3G 850/1900 MHz) ........ $480
LG P920 Optimus 3D ......... $450
LG Quantum C900B (3G 850/1900/2100 MHz) .... $370

Samsung Google Nexus S i9020T ......... $300
Samsung Google Nexus S (3G 850/1900/2100) .... $310
Samsung Galaxy S II i9100 16GB Black ......... $410
Samsung Galaxy S II i9100 16GB White ........ $450
Samsung Galaxy S II i9100 32GB ........... $500
Samsung Infuse 4G ATT Unlocked (3G 850/1900/2100) ..... $430
Samsung Galaxy Nexus GT-i9250 ......... $470

BlackBerry Bold 9790 ...................................... $450
BlackBerry Bold Touch 9900 (Unlocked) ......... $420
BlackBerry Bold Touch 9930 ....... $390
BlackBerry Torch 9800 ..... $300
BlackBerry Bold 9780 (Unlocked 3G 850/1900/2100 MHz) .... $300

Tablet PC®

Apple iPad2 - 16GB (WiFi Only) .... $360
Apple iPad2 - 16GB (WiFi+3G) ..... $400
Apple iPad2 - 32GB (WiFi Only) ... $430
Apple iPad2 - 32GB (WiFi+3G) ...... $500
Apple iPad2 - 64GB (WiFi Only) ... $500
Apple iPad2 - 64GB (WiFi+3G) ..... $650

Samsung Series 7 XE700 11.6-Inch Slate (64 GB, Win 7 Pro) ...... $800
Samsung Series 7 XE700 11.6-Inch Slate (128 GB, Win 7 Pro) .... $990
Samsung Galaxy Tab 10.1 (16GB Wi-Fi Version) ..... $360
Samsung Galaxy Tab 10.1 (32GB Wi-Fi Version) ..... $400
Samsung Galaxy Tab 10.1 16GB (WIFI + 3G Version) (3G 850/900/1900/2100) ..... $450
Samsung Galaxy Tab 10.1 32GB (WIFI + 3G Version) ......... $500
Samsung Galaxy Tab 10.1 64GB (WIFI + 3G Version) ......... $600
Samsung P1000 Galaxy Tab ........ $300
Samsung P1000 Galaxy Tab AT&T Unlocked Version (3G 850) ....... $500

Many people have not ever heard the term “unlocked cell phones,” and would wonder what such a thing is. Those in the know, however, know that an unlocked cell phone is one that isn’t tied to any specific cell phone service carrier. When you have an unlocked cell phone, you’re free to take it to any service carrier that best suits your needs without having to buy one of their phones.

You can't afford to miss it ,

1. Complete accessories (Well packed and sealed in original company box)
2. Unlocked /sim free.
3. Brand new (original manufacturer) box - no copies
4. All phones have English language as default
5. All material (software, manual)- car chargers - home chargers - usb data
cables -holsters/belt clips -wireless headsets(bluetooth) - leather and non-leather carrying cases
-batteries. if you are interested, forward your questions and enquires via email with your order and shipping details. we give
1 year warranty for every product sold out to our costumers .

Wise Buy Trading Company says® .....
Stop wasting.start saving
we provide 24x7 email support ? so what are you waiting for ?
email now : wisebuycoy@live.com
Call us : 00447035936654
(Mon-Fri , 9AM-6PM EST)
(SAT : 12Noon-5PM EST)

Deploying a "special" geographically dispersed gridsql

fnikator — GMT

Hi all

I'm considering a non-std kind of gridsql deployment, seeking info on if its possible.

I have a number of geographically dispersed sites. I would like to have a gridsql child node on each site, with a single central master node for all the child nodes.

Each site should be able to load its child node data independently (preferably though libpq), and I'm hoping that if i choose my partition key well, little to no information will need to pass to the central master node. The applications loading the child nodes with data will only be loading data with the partition keys allocated to that particular node (the apps will know what the partition key is for the child node its loading)

The idea is that the master node will ONLY EVER be required to read data from the node DB's, not update or modify them (unless its using really low bandwidth to modify metadata tables or whatever).
I expect that the master node will create tables on all the child nodes (which is not a big deal, and should be low bandwidth operation).

Does this sound feasibly? if code mods in gridsql are required, what sort of level of effort am I looking at to do this?

delete from table a using a.x=b.y is hanging in grid

kenn.minear — GMT

Hi all,

We are using pgplus 8.3.2 with grid 1.1.0.9

we have 2 tables in grid

1.[b] test_1 [/b] count:8067374 (having 10 columns)
2.[b] tempi [/b] count:317655 (having 1 column)

I am using the following query in grid

[quote]delete from test_1 where x in (select y from tempi);[/quote]

The query is taking too much time and at last hangs.

Kindly look into the issue

Thanks & Regards
Pavani

Transparency in the development of GridSQL

SteveH369 — GMT

GridSQL is meant for heavy processing of lots of data. Typically such amounts of data means that one needs relatively expensive hardware and quite a lot of cost involved in the ETL process. This means that the target users of GridSQL are those that are not able to get their processing done with pure PostgreSQL or PostgreSQL plus or any open source software packages. This again means that you are dealing with (or want to deal with) users who are really experiensed with using open source software. Such users will know that you need to be very careful when judging the maturity of open source software (especially those who have experience with the [url=http://monty-says.blogspot.com/2008/11/oops-we-did-it-again-mysql-51-released.html]early release tradition[/url] of MySQL). With GridSQL I am really having a hard time judging the maturity. The reason for this is the lack (correct me if I am wrong) of public test package, lack of a list of open bugs containing severity, and lack of a roadmap (short term if you do not want to disclose the long-term roadmap). This means that since I have run into some bugs while testing GridSQL, I immediately got scared and basically ran away.

What I would really like, is to have a discussion about how to get the development process to be more transparent, so there is strong documentation of the maturity.

One immediate idea I have in order to increase the transparency is to run all the SQL queries that are used to test PostgreSQL (I guess there should be lots of them) on GridSQL and see how it goes. There should also be a performance comparison between GridSQL and PostgreSQL because really poor performance is a bug. I would expect basically all the PostgreSQL queries to take longer on GridSQL, because of the overhead involved in the parallelization, but this is fine.

Is this a good idea? Any views on the idea of making the GridSQL development process more transparent?

Update: I do not know how I could post on behalf of another user, but by username is David_ not SteveH...

Where did the architectural and other project documentation go?

rstephan — GMT

It use to be located on the EnterpriseDB site and Google still says it is at http://enterprisedb.org/community/projects/gridsql/architecture.do . Where did it go? And there use to be White Papers and other information.

Does GridSQL support stored procedures now?

jabyga — GMT

Does GridSQL support stored procedures now?

Failed to update RowID generator

chiru_r — GMT

Dear all,

I am getting below error while loading data into child table table.
i am useing master and child tables.

Fri, 14 Jan 2011 17:26:17 GMT+05:30 ERROR [DatabaseLoad] -> Error While Loading : org.postgresql.util.PSQLException: ERROR: Can not prepare request: Failed to update RowID generator

Please give situation any bady.

Thanks in Advance
RC

Hello Friend

erwin_smith — GMT

I am new at your site. But I like to join this community. There are lot of topic for reply but I like provide you a new topic.You should enjoy this topic and get information about it. Thanks
===========

View Issue

null — GMT

Hi, all

I created a view as
[code]create view v_test as
select *
from (select * from test ) a left join (select * from test2 ) b
on a.id = b.id
[/code]

When I selects from this view, the error comes out.

[color=red]ERROR: Failed to parse view statement: select * from (select * from test ) a left join (select * from test2 ) b on a.id = b.i[/color]

I checked the xsysviews table, the view statement missed some letters.

I corrected the view statement and selected from the view again, but I still got the same error.

I tried restart the server, then I got a new error.

[color=red]ERROR: Node 3 has aborted execution, cause is: org.postgresql.util.PSQLException : ERROR: each UNION query must have the same number of columns
Position: 363
QUERY: SELECT "TMPTT3_3"."id" as "id", "TMPTT3_3"."name" as "name", "TMPTT3_3"."password" as "password", "TMPTT3_3"."rq" as "rq", "TMPTT3_3"."id" as "EXPRESSION1", "TMPTT3_3"."name" as "EXPRESSION2", "TMPTT3_3"."password" as "EXPRESSION3", "TMPTT3_3"."rq" as "EXPRESSION4" FROM "TMPTT3_3" INNER JOIN "TMPTT5_2" ON ("TMPTT3_3"."id" = "TMPTT5_2"."id") UNI...[/color]

Does the gridsql cache the view statement?

Does gridsql support full join now?

Thanks.

Regards,
David

Left Join porblem

null — GMT

Hi, all

[code]select *
(select * from test ) a
left join
(select * from test2) b
on a.id = b.id[/code]

[code]select count(*)
from
(select * from test ) a
left join
(select * from test2) b
on a.id = b.id[/code]

table a id: 1,2,4
table b id: 1,2,5

This select returns 3 rows, but the count returns 12. My gridsql is setting up with 4 nodes.

But I tried nesting the select, it was working.

[code]select count(*)
from
(select *
from (select * from test ) a left join (select * from test2) b on a.id = b.id)[/code]

The simple one without nesting select was also working.

[code]select count(*)
test a
left join
test2 b
on a.id = b.id[/code]

Is this caused by nesting select? Do you have any ideas about this?

Thanks.

Regards,
David

Alias Problem

null — GMT

Hi, all

When I'm using the alias for union function, I meet a problem. The nesting alias doesn't work.

[code]select id2 as id3, name2 name3, password2 password3, rq2 rq3
from
(
select id id2, name name2, password password2, rq rq2
from test
) a[/code]

[color=red]ERROR: Node 2 has aborted execution, cause is: org.postgresql.util.PSQLException : ERROR: column a.id2 does not exist
Position: 8
QUERY: SELECT "a"."id2" as "id3", "a"."name2" as "name3", "a"."password2" as "password3", "a"."rq2" as "rq3" FROM "test" "a"[/color]

It seems some optimization removes the first level of select.

Do you have any ideas about this?

Thanks.

Regards,
David

Insert Very Slow

null — GMT

Hi all,

I have installed two postgres instances on one server with different ports. And I set up the gridsql with these two nodes.

I inserted some data into the gridsql server, but it was very slow, about 100rows per second. I tried to insert directly into the node database, it was much faster, about 20,000 rows per second. I was using a java application, jdbc, batchexcute for both two tests.

I also tried select from the node and the gridsql. The gridsql is faster as we expected.

Could you please help me to figure out the insert problem? Thanks.

Regards,
David

how to config the agent ?

xu_zhao — GMT

Here is two node in my gridsql conifg:
1) coordinator : 172.29.0.30
2) agent: 172.29.0.31
This is my coordinator pc(172.29.0.30)'s gridsql.conf

###########################################################################
# Copyright (c) 2010 EnterpriseDB Corporation
#
# gridsql.config
#
# GridSQL configuration file
###########################################################################

###
### Server settings
###

xdb.port=6453
xdb.maxconnections=10

###
### Node & JDBC Pool configuration
###

### Set defaults for all nodes and MetaData database.
### These can be overriden.

xdb.default.dbusername=gridsql
xdb.default.dbpassword=gridsql

xdb.default.dbport=5432

### Connection thread defaults for each node
### Note that these are pooled, so the number of clients connected
### to GridSQL can be greater than pool size.

xdb.default.threads.pool.initsize=5
xdb.default.threads.pool.maxsize=10

### Connectivity for MetaData database

xdb.metadata.database=XDBSYS
xdb.metadata.dbhost=172.29.0.30

### The number of nodes in cluster

xdb.nodecount=2

### The hosts of the underlying databases

xdb.node.1.dbhost=172.29.0.30
xdb.node.2.dbhost=172.29.0.31
#xdb.node.3.dbhost=127.0.0.1
#xdb.node.4.dbhost=127.0.0.1

### Designate coordinator node number
### In practice, the coordinator node should be the node where
### GridSQL is running.

xdb.coordinator.node=1

###
### The next few sections are required when wanting to run agents on the nodes.
### Uncomment them and modify to communicate with agents
###

### Only for agent version
### Port for node's SocketCommunicator

#xdb.node.1.port=6455
#xdb.node.1.host=192.168.123.100
xdb.node.1.host=172.29.0.30
xdb.node.1.port=6455
xdb.node.2.host=172.29.0.31
xdb.node.2.port=6455
#xdb.node.3.port=6455
#xdb.node.3.host=192.168.123.102
#xdb.node.4.port=6455
#xdb.node.4.host=192.168.123.103

### Designate coordinator node
### In practice, the coordinator node should be the node where
### GridSQL is running.

#xdb.coordinator.host=192.168.123.100
xdb.coordinator.host=172.29.0.30
xdb.coordinator.port=6454

# Specify protocol types.
# Can use local connection between coordinator and node 1,
# since they are the same system

#xdb.connector.0.1=0
#xdb.connector.1.0=0

###
### Logging Settings
###

### The log4j library is used.
### More info at http://logging.apache.org/log4j/docs/

# rootLogger. Log warnings and errors.
log4j.rootLogger=WARN, console

# Define other characteristics for console log
log4j.appender.console=org.apache.log4j.RollingFileAppender
log4j.appender.console.maxFileSize=500KB
log4j.appender.console.maxBackupIndex=10
log4j.appender.console.layout=org.apache.log4j.PatternLayout
log4j.appender.console.layout.ConversionPattern=%d{ISO8601} - %-5p %m%n
log4j.appender.console.File=/usr/local/gridsql-2.0/log/console.log

# Log Server messages to the console logger
log4j.logger.Server=ALL, console

# Query logger.
# This logs all queries sent to the database.
log4j.logger.query=INFO, QUERY
log4j.appender.QUERY=org.apache.log4j.RollingFileAppender
log4j.appender.QUERY.File=/usr/local/gridsql-2.0/log/query.log
log4j.appender.QUERY.maxFileSize=500KB
log4j.appender.QUERY.maxBackupIndex=10
log4j.appender.QUERY.layout=org.apache.log4j.PatternLayout
log4j.appender.QUERY.layout.ConversionPattern=%d{ISO8601} - %m%n

# Activity logger.
# This logs grid activity
log4j.logger.activity=INFO, activity
log4j.appender.activity=org.apache.log4j.RollingFileAppender
log4j.appender.activity.File=/usr/local/gridsql-2.0/log/activity.log
log4j.appender.activity.maxFileSize=10MB
log4j.appender.activity.maxBackupIndex=10
log4j.appender.activity.layout=org.apache.log4j.PatternLayout
log4j.appender.activity.layout.ConversionPattern=%d{ISO8601} - %m%n

# Uncomment this if you would like other SQL commands other
# than SELECT to be logged in the query logger as well
# (e.g. INSERT, UPDATE, DELETE).

#log4j.logger.command=INFO, QUERY

# A separate "long query" log may be defined to separately log queries
# that appear to be be taking a long time.
# Specify the threshold in seconds at which queries will show up in the
# long query log.
xdb.longQuerySeconds=300

log4j.logger.longquery=INFO, LONGQUERY
log4j.appender.LONGQUERY=org.apache.log4j.RollingFileAppender
log4j.appender.LONGQUERY.File=/usr/local/gridsql-2.0/log/longqry.log
log4j.appender.LONGQUERY.maxFileSize=500KB
log4j.appender.LONGQUERY.maxBackupIndex=10
log4j.appender.LONGQUERY.layout=org.apache.log4j.PatternLayout
log4j.appender.LONGQUERY.layout.ConversionPattern=%d{ISO8601} - %m%n

### User defined functions: subdate, adddate

xdb.sqlfunction.subdate.template=DATE({arg1})-INTERVAL '{arg2} days'
xdb.sqlfunction.subdate.returntype=DATE
xdb.sqlfunction.subdate.paramcount=2
xdb.sqlfunction.subdate.arg1=DATE
xdb.sqlfunction.subdate.arg2=ANYNUMBER

xdb.sqlfunction.adddate.template=DATE({arg1})+INTERVAL '{arg2} days'
xdb.sqlfunction.adddate.returntype=DATE
xdb.sqlfunction.adddate.paramcount=2
xdb.sqlfunction.adddate.arg1=ANYDATE
xdb.sqlfunction.adddate.arg2=ANYNUMBER

#### Redwood Style Date Setting
# When set to true DATE translates to TIMESTAMP(0) during DDL, default=false
#xdb.edb_redwood_date=false

----------------------
how to config the agent(172.29.0.31)?

cannot send data to nodes

sri_v — GMT

Hi,

I am

Question about connection pool settings when using gs-loader.sh

ranga.gopalan@sorrisotech.com — GMT

Hi,

I have the following setup for my GridSQL setup with 2 nodes (both pointing to the same Postgres server)

[code]
xdb.maxconnections=200
xdb.default.threads.pool.initsize=5
xdb.default.threads.pool.maxsize=15
xdb.nodecount=2
[/code]

I have not modified the jdbc pool parameters and assume that the values will be picked up based upon the thread pool settings above.

I have some batch processing that kicks off and invokes several gs-loader.sh processes in parallel to load data into the database. I noticed that sometimes I am running out of connections - I noticed that after my run there were upto 200 connections taken up on postgres by the GridSQL. (Assume that the batch processing kicked off about a 100 gs-loader processes in parallel).

Questions:

1. Is the number of connections taken up on Postgres not limited by the xdb.default.threads.pool.maxsize value in case of gs-loader processes?

- I would have expected the behavior to be that a maximum of 15 x 2 (nodes) = 30 connections would be taken up and perhaps the batch loading would get slowed down. However, it appears that each gs-loader.sh client seems to take up its own connection, which is not cleanly closed up after the gs-loader exits.

2. The connections on Postgres don't seem to time out (as I would expect after 10 minutes based upon the documented pool timeout default)

I can understand the need to raise the total number of connections on Postgres (I've set it to 500 now) - but it still does not explain the high number of connections or why they don't time out.

Can you please explain?

Thanks,

Ranga

port out of range error

sivakumar_k — GMT

I am getting the following exception when i ran the following command:

gs-server.sh -d xtest

Also note that i dont have gs-agent running on any nodes.

2010-08-26 11:06:24,242 - INFO Coordinator: Node 1 is connected
2010-08-26 11:06:24,242 - INFO Coordinator: Node 1 is connected
2010-08-26 11:06:24,248 - ERROR Catching throwable:
java.lang.IllegalArgumentException: port out of range:-1
at java.net.InetSocketAddress.(InetSocketAddress.java:83)
at com.edb.gridsql.communication.SocketConnector$SendingThread.send(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.run(Unknown Source)
at java.lang.Thread.run(Thread.java:619)
2010-08-26 11:06:24,249 - ERROR Catching throwable:
java.lang.IllegalArgumentException: port out of range:-1
at java.net.InetSocketAddress.(InetSocketAddress.java:83)
at com.edb.gridsql.communication.SocketConnector$SendingThread.send(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.SocketConnector$SendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.run(Unknown Source)
at java.lang.Thread.run(Thread.java:619)
2010-08-26 11:06:24,249 - ERROR Catching throwable:
java.lang.IllegalArgumentException: port out of range:-1
at java.net.InetSocketAddress.(InetSocketAddress.java:83)
at com.edb.gridsql.communication.SocketConnector$SendingThread.send(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.SocketConnector$SendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.run(Unknown Source)
at java.lang.Thread.run(Thread.java:619)
2010-08-26 11:06:24,250 - ERROR Can not deliver message, sending MSG_ABORT back to Agent
2010-08-26 11:06:24,250 - ERROR Can not deliver message, sending MSG_ABORT back to Agent
2010-08-26 11:06:24,250 - WARN Message was not consumed: NodeMessage[Type:MSG_ABORT, From:3, To: [0], Session: null, Request: 0]
2010-08-26 11:06:24,250 - WARN Message was not consumed: NodeMessage[Type:MSG_ABORT, From:3, To: [0], Session: null, Request: 0]
2010-08-26 11:06:24,250 - ERROR Catching throwable:
java.lang.IllegalArgumentException: port out of range:-1
at java.net.InetSocketAddress.(InetSocketAddress.java:83)
at com.edb.gridsql.communication.SocketConnector$SendingThread.send(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.run(Unknown Source)
at java.lang.Thread.run(Thread.java:619)
2010-08-26 11:06:24,250 - ERROR Catching throwable:
java.lang.IllegalArgumentException: port out of range:-1
at java.net.InetSocketAddress.(InetSocketAddress.java:83)
at com.edb.gridsql.communication.SocketConnector$SendingThread.send(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.SocketConnector$SendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.run(Unknown Source)
at java.lang.Thread.run(Thread.java:619)
2010-08-26 11:06:24,251 - ERROR Catching throwable:
java.lang.IllegalArgumentException: port out of range:-1
at java.net.InetSocketAddress.(InetSocketAddress.java:83)
at com.edb.gridsql.communication.SocketConnector$SendingThread.send(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.SocketConnector$SendingThread.sendFailed(Unknown Source)
at com.edb.gridsql.communication.AbstractConnector$AbstractSendingThread.run(Unknown Source)
at java.lang.Thread.run(Thread.java:619)
2010-08-26 11:06:24,251 - ERROR Can not deliver message, sending MSG_ABORT back to Agent
2010-08-26 11:06:24,251 - ERROR Can not deliver message, sending MSG_ABORT back to Agent
2010-08-26 11:06:24,251 - WARN Message was not consumed: NodeMessage[Type:MSG_ABORT, From:2, To: [0], Session: null, Request: 0]
2010-08-26 11:06:24,251 - WARN Message was not consumed: NodeMessage[Type:MSG_ABORT, From:2, To: [0], Session: null, Request: 0]
2010-08-26 11:06:24,255 - INFO Node 1: thread pool is ready
2010-08-26 11:06:24,255 - INFO Node 1: thread pool is ready
2010-08-26 11:06:24,255 - INFO Node 1: Connection to Coordinator is completed
2010-08-26 11:06:24,255 - INFO Node 1: Connection to Coordinator is completed
2010-08-26 11:06:24,283 - INFO Node 1: connection pool for database xtest is ready
2010-08-26 11:06:24,283 - INFO Node 1: connection pool for database xtest is ready

My gridsql.config file is as follows:

###
### Server settings
###

xdb.port=6453
xdb.maxconnections=10

###
### Node & JDBC Pool configuration
###

### Set defaults for all nodes and MetaData database.
### These can be overriden.

xdb.default.dbusername=gridsql
xdb.default.dbpassword=gridsql

xdb.default.dbport=5432

### Connection thread defaults for each node
### Note that these are pooled, so the number of clients connected
### to GridSQL can be greater than pool size.

xdb.default.threads.pool.initsize=5
xdb.default.threads.pool.maxsize=10
### Connectivity for MetaData database

xdb.metadata.database=XDBSYS
xdb.metadata.dbhost=127.0.0.1
xdb.metadata.dbport=5432

## JDBC Driver
#
xdb.default.jdbcdriver=org.postgresql.Driver
xdb.default.jdbcstring=jdbc:postgresql://{dbhost}:{dbport}/{database}

### The number of nodes in cluster

xdb.nodecount=3

### The hosts of the underlying databases

xdb.node.1.dbhost=216.205.106.242
xdb.node.2.dbhost=216.205.106.230
xdb.node.3.dbhost=216.205.109.142

### Designate coordinator node number
### In practice, the coordinator node should be the node where
### GridSQL is running.

xdb.coordinator.node=1

### Only for agent version
### Port for node's SocketCommunicator

#xdb.node.1.port=6455
#xdb.node.1.host=216.205.106.242
#xdb.node.2.port=6455
#xdb.node.2.host=216.205.106.230
#xdb.node.3.port=6455
#xdb.node.3.host=216.205.109.142
#xdb.node.4.port=6455
#xdb.node.4.host=192.168.123.103

### Designate coordinator node
### In practice, the coordinator node should be the node where
### GridSQL is running.

xdb.coordinator.host=216.205.106.242
xdb.coordinator.port=6454

# Specify protocol types.
# Can use local connection between coordinator and node 1,
# since they are the same system

xdb.connector.0.1=0
xdb.connector.1.0=0

GridSQL 2.0 Released

Mason_S — GMT

GridSQL 2.0 is immediately available for download from sourceforge.net at:

http://sourceforge.net/projects/gridsql/files

From the release notes:

This software requires the Java Runtime Environment 5,
version 1.5.0_12 or later. Earlier versions of the JRE may
result in memory leaks that over time result in an OutOfMemory
exception.

GridSQL has also been successfully tested with Java 6 update 18,
and is recommended.

If GridSQL was not installed via an installer that included a
JRE, please go to http://java.sun.com/javase/downloads/index.jsp
to download a Java Runtime Environment

After installing, please make sure that your PATH environment
variable refers to the JRE's bin directory.

This version includes enhancements, performance improvements
and minor bug fixes, including

- Faster, more efficient row shipping which makes cross-node queries and multi-node aggregation execute more quickly
- Faster dumping of data via COPY
- EXPLAIN command is now supported which shows how GridSQL processes queries
- Improved thread connection handling
- Better compatibility with PostgreSQL connectors
- Added support for multiple statements in one request
- Expression based indexes may now be created
- Expanded support for usage of inet, macaddr, and cidr data types

how to restart the gridsql when one node is crash.

null — GMT

With open source GridSQL, how to restart the gridsql when one node is crash.

Paralle data loading

sijie_guo — GMT

Hi,

It is my first time to use GridSQL.
I tried to load 200G data into a parititioned table of a 20-nodes GridSQL cluster using gs-loader. It seems that gs-loader will send data to the coordinator, then the coordinator partitions the data to each agent. So the coordinator is a bottleneck during data loading.

Is there any parallel data loading tools to load the data parallely?

Regards,

Samuel

could not start gridsql

null — GMT

After install the gridsql 2.0 and config it, then start the gs-server.sh , then start gs-dbstart.sh -d test.

always create an error 'can not bring db test online'

Not able retieve records even though data is available in the table

sri_v — GMT

Hi,

We are using GridSQL 1.1.0.9 with postgresplus standard server.
We are having problem in getting records even though data is available in the table.

GridSQL -> select count(*) from test where served_number='779294936';
+----------+
| count(*) |
+----------+
| 0 |
+----------+
1 row(s).

GridSQL -> select served_number from test where served_number like '%779294936%';
+---------------+
| served_number |
+---------------+
| 779294936 |
| 779294936 |
| 779294936 |
| 779294936 |
| 779294936 |
+---------------+
5 row(s).

GridSQL -> select served_number,length(served_number) from test where served_number like '%779294936%';
+------------------------+
| served_number | length |
+------------------------+
| 779294936 | 9 |
| 779294936 | 9 |
| 779294936 | 9 |
| 779294936 | 9 |
| 779294936 | 9 |
+------------------------+

It is working fine if we use trim,but is not required as the above query shows all the records are having same length.

If we create a temp table with create table test1 as select served_number from test;
select count(*) from test1 where served_number='779294936';
+----------+
| count(*) |
+----------+
| 5 |
+----------+

Then it is working fine.
What can be the reson for this problem?

Thanks
Sri

Does the gridSQL server not use the exact same SQl parser as the backend postgres server?

Bill_Fanselow — GMT

Many (documented) commands which work on direct connection to postgres server do not seem to work when connected thru the gridSQL server. For example: SELECT TIMESTAMP WITH TIME ZONE 'epoch' + 982384720 * INTERVAL '1 second';

This works fine when connected directly to the postgres server running behind the gridSQL service, but gives this error when entered into the GridSQL prompt: SQLException: ERROR: Encountered "\'epoch\'" at line 1, column 33.

Does the gridSQL server not use the exact same SQl parser as the backend postgres server?

Add a new node to a coordinator

Wilfried_Hartman — GMT

Hello

I'd like to use GridSQL to execute distributed queries on several PostgreSQL servers.

I ran some test with GridSQL but I still have some questions. My coordinator is currently managing two nodes. These two nodes have been "initialized" (DB schema creation) through the coordinator with some "create table" instructions. In a few months, when I'll decide to add a new node, how will I initialize it ? I can't use 'create table' instructions in the coordinator because the 2 first nodes will already have these tables... So, is there another way to initialize the third node than "manually" filling the metadata database ? Is it possible to synchronize a new node from an existing one ? Is it possible to do it without rebooting gs-server ?

Thank you for your time.

Install GRIDSQL 2.0

sirisha_s — GMT

Hi,

I have installed Gridsql 2.0 . I am trying to test the performance of COPY with respect to Gridsql 1.0.

Please find below the loading time:

grid 1.0
----------
GridSQL -> copy table1 from '/home/postgres/load_test/table1' delimiter ',' CSV;
OK
Response time: 6m 37s 572ms Total time: 6m 37s 572ms

grid 2.0
----------
GridSQL -> copy table1 from '/home/postgres/load_test/table1' delimiter ',' CSV;
OK
Response time: 6m 48s 159ms Total time: 6m 48s 159ms

the file sixe is 9.2 GB and the number of records are 19000000.

I haven't found any significance improvement in the performance of COPY.

Can you please explain me why is it so.

Thanks.