[jira] [Updated] (PHOENIX-5256) Remove queryserver related scripts/files as the former has its own repo
[ https://issues.apache.org/jira/browse/PHOENIX-5256?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Lars Hofhansl updated PHOENIX-5256: --- Fix Version/s: (was: 5.1.0) (was: 4.15.0) 5.1.1 4.15.1 > Remove queryserver related scripts/files as the former has its own repo > --- > > Key: PHOENIX-5256 > URL: https://issues.apache.org/jira/browse/PHOENIX-5256 > Project: Phoenix > Issue Type: Improvement >Affects Versions: 5.0.0, 4.14.2 >Reporter: Swaroopa Kadam >Assignee: Swaroopa Kadam >Priority: Blocker > Labels: newbie > Fix For: 4.15.1, 5.1.1 > > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (PHOENIX-5490) Queries with leading part of the primary key doing a parallel scan causes missing data
Amarnath Ramamoorthi created PHOENIX-5490:
-
Summary: Queries with leading part of the primary key doing a
parallel scan causes missing data
Key: PHOENIX-5490
URL: https://issues.apache.org/jira/browse/PHOENIX-5490
Project: Phoenix
Issue Type: Bug
Reporter: Amarnath Ramamoorthi
Table properties
* saltbucket_size=10
* compression=snappy
* normalization_enabled=true
PRIMARY KEY ("id_1", "id_2", "id_3", "id_4", "id_5", "id_6", "id")
Partial primary key lookup is not working as expected.
On using only one key "id_1" = 60268375; should return 2 rows
{code:java|title=Parallel scan shows 1 rows}
0: jdbc:phoenix:labs-test-namenode-lv-101,labs> explain SELECT "id_1", "id_2",
"id_3", "id_4", "id_5", "id_6", "id" FROM "foo"."BIG_TABLE" WHERE "id_1" =
60268375;
+--+
| PLAN
|
+--+
| CLIENT 8-CHUNK 0 ROWS 0 BYTES PARALLEL 8-WAY ROUND ROBIN RANGE SCAN OVER
foo.BIG_TABLE [0,60268375] - [9,60268375] |
| SERVER FILTER BY FIRST KEY ONLY
|
+--+
2 rows selected (0.047 seconds)
0: jdbc:phoenix:labs-test-namenode-lv-101,labs> SELECT "id_1", "id_2", "id_3",
"id_4", "id_5", "id_6", "id" FROM "foo"."BIG_TABLE" WHERE "id_1" = 60268375;
+---+-+++++-+
| id_1 |id_2| id_3 | id_4| id_5 | id_6 | id |
+---+-+++++-+
| 60268375 | 1306309959 | 4 | 365224321 | 359338114 | 523433837 |
1052445537 |
+---+-+++++-+
1 row selected (3.24 seconds)
{code}
{code:java|title=Serial scan shows 2 rows}
0: jdbc:phoenix:labs-test-namenode-lv-101,labs> explain SELECT "id_1", "id_2",
"id_3", "id_4", "id_5", "id_6", "id" FROM "foo"."BIG_TABLE" WHERE "id_1" =
60268375 limit 10;
+---+
| PLAN
|
+---+
| CLIENT 10-CHUNK SERIAL 10-WAY ROUND ROBIN RANGE SCAN OVER foo.BIG_TABLE
[0,60268375] - [9,60268375] |
| SERVER FILTER BY FIRST KEY ONLY
|
| SERVER 10 ROW LIMIT
|
| CLIENT 10 ROW LIMIT
|
+---+
0: jdbc:phoenix:labs-test-namenode-lv-101,labs> SELECT "id_1", "id_2", "id_3",
"id_4", "id_5", "id_6", "id" FROM "foo"."BIG_TABLE" WHERE "id_1" = 60268375
limit 10;
+---+-+++++-+
| id_1 |id_2| id_3 | id_4| id_5 | id_6 | id |
+---+-+++++-+
| 60268375 | 1306309960 | 4 | 365224322 | 359338115 | 523433838 |
1052445538 |
| 60268375 | 1306309959 | 4 | 365224321 | 359338114 | 523433837 |
1052445537 |
+---+-+++++-+
2 rows selected (0.12 seconds)
{code}
Now the missing row can be retrieved on using full primary key.
{code:java}
0: jdbc:phoenix:labs-test-namenode-lv-101,labs> select "id_1", "id_2", "id_3",
"id_4", "id_5", "id_6", "id" FROM "foo"."BIG_TABLE" WHERE "id_1" IN (60268375)
AND "id_2" IN (1306309960) AND "id_3" = 4 AND "id_4" = 365224322 AND "id_5" =
359338115 AND "id_6" = 523433838 AND "id" = 1052445538;
+---+-+++++-+
| id_1 |id_2| id_3 | id_4| id_5 | id_6 | id |
+---+-+++++-+
| 60268375 | 1306309960 | 4 | 365224322 | 359338115 | 523433838 |
1052445538 |
+---+-+++++-+
1 row selected (0.07 seconds)
[jira] [Created] (PHOENIX-5489) Expected length of at least 8 bytes, but had 7 (state=22000,code=201)
Ashish created PHOENIX-5489: --- Summary: Expected length of at least 8 bytes, but had 7 (state=22000,code=201) Key: PHOENIX-5489 URL: https://issues.apache.org/jira/browse/PHOENIX-5489 Project: Phoenix Issue Type: Bug Affects Versions: 4.14.1 Reporter: Ashish Fix For: 4.14.1 As part of the process to upgrade phoenix from 4.7 to 4.14. We need to export all tables' data out. We found the following error during export one of the table. This indicates some data is not written to hbase by using phoenix client api or the one of the column data type has been changed. We got below error. I am not sure how to fix this. Need your assistance on this ASAP. java.sql.SQLException: ERROR 201 (22000): Illegal data. Expected length of at least 8 bytes, but had 7 at org.apache.phoenix.mapreduce.PhoenixRecordReader.nextKeyValue(PhoenixRecordReader.java:159) This can be reproduced by running: select messageid from sfe.sfe_trade limit 100; -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (PHOENIX-5489) Expected length of at least 8 bytes, but had 7 (state=22000,code=201)
[ https://issues.apache.org/jira/browse/PHOENIX-5489?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ashish updated PHOENIX-5489: Description: As part of the process to upgrade phoenix from 4.7 to 4.14. We need to export all tables' data out. We found the following error during export one of the table. This indicates some data is not written to hbase by using phoenix client api or the one of the column data type has been changed. We got below error. I am not sure how to fix this. Need your assistance on this ASAP. java.sql.SQLException: ERROR 201 (22000): Illegal data. Expected length of at least 8 bytes, but had 7 at org.apache.phoenix.mapreduce.PhoenixRecordReader.nextKeyValue(PhoenixRecordReader.java:159) This can be reproduced by running: select messageid from TABLE limit 100; was: As part of the process to upgrade phoenix from 4.7 to 4.14. We need to export all tables' data out. We found the following error during export one of the table. This indicates some data is not written to hbase by using phoenix client api or the one of the column data type has been changed. We got below error. I am not sure how to fix this. Need your assistance on this ASAP. java.sql.SQLException: ERROR 201 (22000): Illegal data. Expected length of at least 8 bytes, but had 7 at org.apache.phoenix.mapreduce.PhoenixRecordReader.nextKeyValue(PhoenixRecordReader.java:159) This can be reproduced by running: select messageid from sfe.sfe_trade limit 100; > Expected length of at least 8 bytes, but had 7 (state=22000,code=201) > - > > Key: PHOENIX-5489 > URL: https://issues.apache.org/jira/browse/PHOENIX-5489 > Project: Phoenix > Issue Type: Bug >Affects Versions: 4.14.1 >Reporter: Ashish >Priority: Major > Fix For: 4.14.1 > > > As part of the process to upgrade phoenix from 4.7 to 4.14. We need to export > all tables' data out. We found the following error during export one of the > table. This indicates some data is not written to hbase by using phoenix > client api or the one of the column data type has been changed. We got below > error. I am not sure how to fix this. Need your assistance on this ASAP. > > java.sql.SQLException: ERROR 201 (22000): Illegal data. Expected length of at > least 8 bytes, but had 7 > at > org.apache.phoenix.mapreduce.PhoenixRecordReader.nextKeyValue(PhoenixRecordReader.java:159) > > This can be reproduced by running: > select messageid from TABLE limit 100; -- This message was sent by Atlassian Jira (v8.3.4#803005)
