[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16739393#comment-16739393
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~dpavlov] [~isapego] [~ptupitsyn] thank you for your reviews! I have merged
this change.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16739386#comment-16739386
]
Ignite TC Bot commented on IGNITE-10732:
{panel:title=-- Run :: All: Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}ZooKeeper (Discovery) 2{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=2760675]]
{color:#d04437}Cache (Restarts) 1{color} [[tests
2|https://ci.ignite.apache.org/viewLog.html?buildId=2760697]]
* IgniteCacheRestartTestSuite:
GridCachePartitionedNodeRestartTest.testRestartWithPutFourNodesNoBackups - 0,0%
fails in last 525 master runs.
{panel}
[TeamCity *-- Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=2760738buildTypeId=IgniteTests24Java8_RunAll]
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16739389#comment-16739389
]
Ignite TC Bot commented on IGNITE-10732:
{panel:title=-- Run :: All: Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}ZooKeeper (Discovery) 2{color} [[tests 0 TIMEOUT , Exit Code
|https://ci.ignite.apache.org/viewLog.html?buildId=2760675]]
* GridCommandHandlerTest.testKillHangingRemoteTransactions (last started)
{color:#d04437}Continuous Query 2{color} [[tests 0 TIMEOUT , Out Of Memory
Error , Exit Code |https://ci.ignite.apache.org/viewLog.html?buildId=2760647]]
{color:#d04437}Cache (Restarts) 1{color} [[tests
2|https://ci.ignite.apache.org/viewLog.html?buildId=2760697]]
* IgniteCacheRestartTestSuite:
GridCachePartitionedNodeRestartTest.testRestartWithPutFourNodesNoBackups - 0,0%
fails in last 525 master runs.
{panel}
[TeamCity *-- Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=2760738buildTypeId=IgniteTests24Java8_RunAll]
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16738474#comment-16738474
]
Pavel Tupitsyn commented on IGNITE-10732:
-
[~ilyak] please revert changes from {{OrdinalIgnoreCase}} to {{Ordinal}}, this
breaks compatibility.
Otherwise .NET part looks good to me.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16738472#comment-16738472
]
Ignite TC Bot commented on IGNITE-10732:
{panel:title=-- Run :: All: Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}Cache (Restarts) 1{color} [[tests
4|https://ci.ignite.apache.org/viewLog.html?buildId=2752954]]
* IgniteCacheRestartTestSuite:
GridCacheReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups -
0,0% fails in last 521 master runs.
* IgniteCacheRestartTestSuite:
GridCachePartitionedNodeRestartTest.testRestartWithTxTwoNodesOneBackup - 0,0%
fails in last 521 master runs.
{color:#d04437}Cache (Restarts) 2{color} [[tests
3|https://ci.ignite.apache.org/viewLog.html?buildId=2752955]]
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutTenNodesTwoBackups
- 0,0% fails in last 565 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups
- 0,0% fails in last 565 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithTxPutAllTenNodesTwoBackups
- 0,0% fails in last 565 master runs.
{color:#d04437}Data Structures{color} [[tests
4|https://ci.ignite.apache.org/viewLog.html?buildId=2752969]]
* IgniteCacheDataStructuresSelfTestSuite:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithTxPutAllTenNodesTwoBackups
- 0,0% fails in last 560 master runs.
* IgniteCacheDataStructuresSelfTestSuite:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutSixNodesTwoBackups
- 0,0% fails in last 560 master runs.
* IgniteCacheDataStructuresSelfTestSuite:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutTenNodesTwoBackups
- 0,0% fails in last 560 master runs.
* IgniteCacheDataStructuresSelfTestSuite:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups
- 0,0% fails in last 560 master runs.
{panel}
[TeamCity *-- Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=2752995buildTypeId=IgniteTests24Java8_RunAll]
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16738246#comment-16738246
]
Dmitriy Pavlov commented on IGNITE-10732:
-
[~ilyak], it is OK for me to keep warning, but I still think odd warning may
confuse users.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16738094#comment-16738094
]
Igor Sapego commented on IGNITE-10732:
--
[~ilyak],
Instead of
{code:cpp}
if (i->find(fileEncParam) == 0)
{code}
Use
{code:cpp}
if (i->find(fileEncParam) != std::string::npos)
{code}
Otherwise seems OK, though I'd like to see relevant C++ TC suits run results.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16738072#comment-16738072
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~vveider] fixed indentation.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16738077#comment-16738077
]
Ignite TC Bot commented on IGNITE-10732:
{panel:title=-- Run :: All: Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}Cache (Restarts) 1{color} [[tests
2|https://ci.ignite.apache.org/viewLog.html?buildId=2678099]]
* IgniteCacheRestartTestSuite:
GridCacheReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups -
0,0% fails in last 521 master runs.
{color:#d04437}Cache (Restarts) 2{color} [[tests
5|https://ci.ignite.apache.org/viewLog.html?buildId=2678100]]
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutTenNodesTwoBackups
- 0,0% fails in last 564 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicNodeRestartTest.testRestartWithPutTenNodesTwoBackups - 0,0%
fails in last 564 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicNodeRestartTest.testRestartWithPutFourNodesNoBackups - 0,0%
fails in last 532 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups
- 0,0% fails in last 564 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithTxPutAllTenNodesTwoBackups
- 0,0% fails in last 564 master runs.
{color:#d04437}ZooKeeper (Discovery) 4{color} [[tests
1|https://ci.ignite.apache.org/viewLog.html?buildId=2678135]]
* ZookeeperDiscoverySpiTestSuite4:
GridCacheReplicatedAtomicMultiNodeFullApiSelfTest.testTtlNoTxOldEntry - 0,0%
fails in last 568 master runs.
{panel}
[TeamCity *-- Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=2678140buildTypeId=IgniteTests24Java8_RunAll]
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 20m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16737898#comment-16737898
]
Peter Ivanov commented on IGNITE-10732:
---
[~ilyak], I'd change indentation of shell part of PR (it should be 4 spaces, as
everywhere in *.sh), otherwise looks pretty solid.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16736499#comment-16736499
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~dpavlov] Even if encoding is consistent between nodes, it might lead to data
corruption when Unicode strings are encoded to 8-bit encoding and then
de-encoded. Some characters will turn into ?'s as they're not representable in
a given 8-bit charset. Therefore, we should keep the warning. They may still
run but they need to be aware. We have quite a few warnings anyway which are
printed even with default configuration.
Maybe we should also check real inconsistency in the cluster, reject nodes
which have file.encoding which is not consistent to existing ones. I think this
will demand another ticket.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16731323#comment-16731323
]
Dmitriy Pavlov commented on IGNITE-10732:
-
[~ilyak] I left only 1 comment in the PR and I'm concerned about a case when
user set non-UTF-8 encoding cluster-wide but set it for all nodes without
mistakes. Will user get a warning in this case? I believe user may potentially
want to have another encoding.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730735#comment-16730735
]
Ilya Kasnacheev commented on IGNITE-10732:
--
Blockers are probably due to old master.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730733#comment-16730733
]
Ignite TC Bot commented on IGNITE-10732:
{panel:title=-- Run :: All: Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}ZooKeeper (Discovery) 4{color} [[tests
1|https://ci.ignite.apache.org/viewLog.html?buildId=2678135]]
* ZookeeperDiscoverySpiTestSuite4:
GridCacheReplicatedAtomicMultiNodeFullApiSelfTest.testTtlNoTxOldEntry - 0,0%
fails in last 422 master runs.
{color:#d04437}Cache (Restarts) 1{color} [[tests
2|https://ci.ignite.apache.org/viewLog.html?buildId=2678099]]
* IgniteCacheRestartTestSuite:
GridCacheReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups -
0,0% fails in last 401 master runs.
{color:#d04437}Cache (Restarts) 2{color} [[tests
5|https://ci.ignite.apache.org/viewLog.html?buildId=2678100]]
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutTenNodesTwoBackups
- 0,0% fails in last 426 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicNodeRestartTest.testRestartWithPutTenNodesTwoBackups - 0,0%
fails in last 426 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicNodeRestartTest.testRestartWithPutFourNodesNoBackups - 0,0%
fails in last 258 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithPutFourNodesNoBackups
- 0,0% fails in last 426 master runs.
* IgniteCacheRestartTestSuite2:
IgniteCacheAtomicReplicatedNodeRestartSelfTest.testRestartWithTxPutAllTenNodesTwoBackups
- 0,0% fails in last 426 master runs.
{panel}
[TeamCity *-- Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=2678140buildTypeId=IgniteTests24Java8_RunAll]
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730688#comment-16730688
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~isapego] please review C++ part of commit.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730689#comment-16730689
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~vveider] please review shell/bat part of commit.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730687#comment-16730687
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~ptupitsyn] please review C# part of commit
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730686#comment-16730686
]
Ilya Kasnacheev commented on IGNITE-10732:
--
[~dpavlov] please review proposed fix
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16730685#comment-16730685
]
Ignite TC Bot commented on IGNITE-10732:
{panel:title=-- Run :: All: Possible
Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}Cache 7{color} [[tests
1|https://ci.ignite.apache.org/viewLog.html?buildId=2614888]]
* IgniteCacheTestSuite7:
TxRollbackAsyncWithPersistenceTest.testSynchronousRollback - 0,0% fails in last
406 master runs.
{panel}
[TeamCity *-- Run :: All*
Results|https://ci.ignite.apache.org/viewLog.html?buildId=2614918buildTypeId=IgniteTests24Java8_RunAll]
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows < - > Windows and Linux < - > Linux will get correct result but
> Windows < - > Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16726792#comment-16726792
]
ASF GitHub Bot commented on IGNITE-10732:
-
GitHub user alamar opened a pull request:
https://github.com/apache/ignite/pull/5725
IGNITE-10732 Force -Dfile.encoding=UTF-8
in linux/windows batch runners, C# and C++.
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/gridgain/apache-ignite ignite-10732
Alternatively you can review and apply these changes as the patch at:
https://github.com/apache/ignite/pull/5725.patch
To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:
This closes #5725
commit 180abe4ad6b54068075affc6d0d5acff0ccd7936
Author: Ilya Kasnacheev
Date: 2018-12-21T14:52:21Z
IGNITE-10732 Force -Dfile.encoding=UTF-8 in linux/windows batch runners, C#
and C++.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Affects Versions: 2.4
>Reporter: Ilya Kasnacheev
>Priority: Critical
> Labels: windows
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows <-> Windows and Linux <-> Linux will get correct result but Windows
> <-> Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
[jira] [Commented] (IGNITE-10732) Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between nodes
[
https://issues.apache.org/jira/browse/IGNITE-10732?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16724237#comment-16724237
]
Ilya Kasnacheev commented on IGNITE-10732:
--
Seems to be broken in 2.4, was pegged to be UTF-8 before.
> Incorrect file.encoding leads to inconsistent SqlFieldsQuery results between
> nodes
> --
>
> Key: IGNITE-10732
> URL: https://issues.apache.org/jira/browse/IGNITE-10732
> Project: Ignite
> Issue Type: Bug
> Components: sql
>Reporter: Ilya Kasnacheev
>Assignee: Ilya Kasnacheev
>Priority: Critical
>
> When doing
> {code}
> cache.query(new SqlFieldsQuery("SELECT _key FROM Cache"))
> {code}
> resulting Unicode values may be different when coming from Windows or Linux
> node.
> Linux nodes will mostly use UTF-8 but Windows nodes will use local Cp
> encoding to encode query results, as bizzare as it may sound.
> Windows <-> Windows and Linux <-> Linux will get correct result but Windows
> <-> Linux will get broken strings.
> Note that if cluster has Windows and Linux nodes and cache is REPLICATED,
> results will be different for subsequent queries!
> There is a workaround for this: set -Dfile.encoding=UTF-8 JVM arg on Windows.
> There is probably an underlying problem in H2 but since non-UTF-8
> file.encoding is dangerous (it affects String.getBytes()) I think we should
> peg it to UTF-8.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
