[jira] [Created] (GEODE-7859) Fix Geode Redis publish/subscribe distributed test
Raymond Ingles created GEODE-7859: - Summary: Fix Geode Redis publish/subscribe distributed test Key: GEODE-7859 URL: https://issues.apache.org/jira/browse/GEODE-7859 Project: Geode Issue Type: Test Components: redis Reporter: Raymond Ingles The Pub/Sub Dunit tests fails intermittently. Timeout has been increased to ensure channel subscriptions complete. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7853) Remove Unnecessary Calls to Logger.isTraceEnabled From InternalDataSerializer
[
https://issues.apache.org/jira/browse/GEODE-7853?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055142#comment-17055142
]
ASF subversion and git services commented on GEODE-7853:
Commit a9a7d15fd32ff9ef557270f5d25cbce35a6ebcd3 in geode's branch
refs/heads/mass-test-run from Juan José Ramos
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=a9a7d15 ]
GEODE-7853: Remove unused field (#4777)
Remove unused field and avoid unnecessary calls to
Logger.isTraceEnabled().
> Remove Unnecessary Calls to Logger.isTraceEnabled From InternalDataSerializer
> -
>
> Key: GEODE-7853
> URL: https://issues.apache.org/jira/browse/GEODE-7853
> Project: Geode
> Issue Type: Bug
> Components: logging, serialization
>Reporter: Juan Ramos
>Assignee: Juan Ramos
>Priority: Major
> Labels: GeodeCommons
> Fix For: 1.13.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> The {{InternalDataSerializer}} class invokes {{Logger.isTraceEnabled()}}
> while
> [initializing|https://github.com/apache/geode/blob/develop/geode-core/src/main/java/org/apache/geode/internal/InternalDataSerializer.java#L249-L251]
> a {{static final}} field that is not used anywhere in the code, remove the
> call and the unused field.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7842) Improve concurrent region creation
[
https://issues.apache.org/jira/browse/GEODE-7842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055136#comment-17055136
]
ASF subversion and git services commented on GEODE-7842:
Commit a5c8e2bc0e41bf591c24e44e1cbfe48c3ca59e7b in geode's branch
refs/heads/mass-test-run from Jens Deppe
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=a5c8e2b ]
GEODE-7842: Improve concurrent region creation (#4767)
> Improve concurrent region creation
> --
>
> Key: GEODE-7842
> URL: https://issues.apache.org/jira/browse/GEODE-7842
> Project: Geode
> Issue Type: Bug
> Components: gfsh
>Reporter: Jens Deppe
>Assignee: Jens Deppe
>Priority: Major
> Fix For: 1.13.0
>
> Time Spent: 50m
> Remaining Estimate: 0h
>
> The Redis adapter has some tests that result in regions being attempted to be
> created concurrently. The region create call uses {{ifNotExists=true}} but
> occasionally I still see {{RegionCreationException}}s as multiple threads are
> in the {{RegionCreateFunction}} concurrently trying to create the same region.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7808) standardize on use of LocatorAddress/HostAddress for connection formation
[ https://issues.apache.org/jira/browse/GEODE-7808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055137#comment-17055137 ] ASF subversion and git services commented on GEODE-7808: Commit 5c2b959e98a6330ecaeddb4d26e11ea29f7a2d7f in geode's branch refs/heads/mass-test-run from Bruce Schuchardt [ https://gitbox.apache.org/repos/asf?p=geode.git;h=5c2b959 ] GEODE-7808: standardize on use of HostAndPort for creating connections (#4765) * Squashed merge of feature/GEODE-7808 removed HostAddress renamed LocatorAddress to HostAndPort modified TcpClient methods to take a HostAndPort argument instead of InetAddress modified SocketCreator to take a HostAndPort argument instead of InetAddress * GEODE-7808 - standardize on use of HostAndPort for connection formation This continues a previous PR that passed and was approved for merge. This commit raises up several methods from SocketCreator into the TcpSocketCreator interface. This is an intermediate commit. A subsequent commit will refactor TcpSocketCreator to separate the client and server methods for creating server-sockets and client connections to server-sockets. * refactored socket-creators to separate concerns ServerSocketCreator holds methods for non-client comms ClientSocketCreator holds methods that clients should use for comms AdvancedSocketCreator holds methods for people who need to get around the limitations of the other two interfaces * adding missing interface * move code out of inner-classes into first-class classes * renaming interfaces and methods to be less confusing * reinstate SocketCreator ip to hostname cache for performance * changes from review comments > standardize on use of LocatorAddress/HostAddress for connection formation > - > > Key: GEODE-7808 > URL: https://issues.apache.org/jira/browse/GEODE-7808 > Project: Geode > Issue Type: Improvement > Components: membership, messaging >Reporter: Bruce J Schuchardt >Assignee: Bruce J Schuchardt >Priority: Major > Fix For: 1.13.0 > > Time Spent: 7.5h > Remaining Estimate: 0h > > We currently use InetAddress and InetSocketAddress in many places to identify > locators, servers and peers. Some work has been done in the past couple of > years to reduce the use of these in order to accommodate changes in IP > addresses due to various causes. The class LocatorAddress was created to > help with this and it is able to hold a host name without resolving it until > that resolution is needed to form a tcp/ip connection. > These days we are seeing more and more movement into cloud computing and the > need to accommodate IP address changes is becoming a bigger issue. To that > end we would like to change our primary client/server and WAN communication > interfaces to stop taking InetAddresses and InetSocketAddresses as arguments > and, instead, take something like a LocatorAddress that can hold an > unresolved hostname that our communication implementations will resolve when > needed. > To that end we should also remove the hostname->inetaddress cache in > SocketCreator and rely on the operating system's DNS cache. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7828) Convert backing store for Redis Hashes and Sets to single regions
[
https://issues.apache.org/jira/browse/GEODE-7828?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055143#comment-17055143
]
ASF subversion and git services commented on GEODE-7828:
Commit f0982cdedf1b122f734919d0647434cca33ec254 in geode's branch
refs/heads/mass-test-run from Nabarun Nag
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=f0982cd ]
Revert "GEODE-7828: Convert backing store for Redis Hashes and Sets to single
regions (#4745)" (#4780)
This reverts commit 2f6bf013368df5a4b5efe68162a4953f9a88bbf2.
> Convert backing store for Redis Hashes and Sets to single regions
> -
>
> Key: GEODE-7828
> URL: https://issues.apache.org/jira/browse/GEODE-7828
> Project: Geode
> Issue Type: Improvement
> Components: redis
>Reporter: Jens Deppe
>Assignee: Jens Deppe
>Priority: Major
> Fix For: 1.13.0
>
> Time Spent: 1.5h
> Remaining Estimate: 0h
>
> Currently the Redis adapter creates separate regions for each new hash or
> set. This can be a significant performance impact when many hashes are being
> created. In addition, hashes are conventionally often named with colons (':')
> in their name - for example {{user:1000}}. This does not work as region names
> cannot contain colons.
> The work here will create a single region for hashes and a single region for
> sets.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7841) Add benchmarking scripts for Geode-Redis module
[ https://issues.apache.org/jira/browse/GEODE-7841?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055141#comment-17055141 ] ASF subversion and git services commented on GEODE-7841: Commit b57456e2cd9f179cf018adaf676a9b78bdc98270 in geode's branch refs/heads/mass-test-run from Jens Deppe [ https://gitbox.apache.org/repos/asf?p=geode.git;h=b57456e ] GEODE-7841 Add benchmarking scripts for Geode-Redis module (#4763) > Add benchmarking scripts for Geode-Redis module > --- > > Key: GEODE-7841 > URL: https://issues.apache.org/jira/browse/GEODE-7841 > Project: Geode > Issue Type: New Feature > Components: benchmarks, redis >Reporter: Raymond Ingles >Assignee: Jens Deppe >Priority: Major > Time Spent: 0.5h > Remaining Estimate: 0h > > Add the following scripts to the Geode-Redis module, in the "performanceTest" > directory: > * aggregator.sh runs the actual tests > * benchmark.sh handles validating the redis server (e.g. starting and > stopping Geode Redis if needed), then runs aggregator.sh > * shacompare.sh takes two git SHAs as arguments and runs benchmark.sh on > each of them, generating separate output files for each. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7808) standardize on use of LocatorAddress/HostAddress for connection formation
[ https://issues.apache.org/jira/browse/GEODE-7808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055138#comment-17055138 ] ASF subversion and git services commented on GEODE-7808: Commit 5c2b959e98a6330ecaeddb4d26e11ea29f7a2d7f in geode's branch refs/heads/mass-test-run from Bruce Schuchardt [ https://gitbox.apache.org/repos/asf?p=geode.git;h=5c2b959 ] GEODE-7808: standardize on use of HostAndPort for creating connections (#4765) * Squashed merge of feature/GEODE-7808 removed HostAddress renamed LocatorAddress to HostAndPort modified TcpClient methods to take a HostAndPort argument instead of InetAddress modified SocketCreator to take a HostAndPort argument instead of InetAddress * GEODE-7808 - standardize on use of HostAndPort for connection formation This continues a previous PR that passed and was approved for merge. This commit raises up several methods from SocketCreator into the TcpSocketCreator interface. This is an intermediate commit. A subsequent commit will refactor TcpSocketCreator to separate the client and server methods for creating server-sockets and client connections to server-sockets. * refactored socket-creators to separate concerns ServerSocketCreator holds methods for non-client comms ClientSocketCreator holds methods that clients should use for comms AdvancedSocketCreator holds methods for people who need to get around the limitations of the other two interfaces * adding missing interface * move code out of inner-classes into first-class classes * renaming interfaces and methods to be less confusing * reinstate SocketCreator ip to hostname cache for performance * changes from review comments > standardize on use of LocatorAddress/HostAddress for connection formation > - > > Key: GEODE-7808 > URL: https://issues.apache.org/jira/browse/GEODE-7808 > Project: Geode > Issue Type: Improvement > Components: membership, messaging >Reporter: Bruce J Schuchardt >Assignee: Bruce J Schuchardt >Priority: Major > Fix For: 1.13.0 > > Time Spent: 7.5h > Remaining Estimate: 0h > > We currently use InetAddress and InetSocketAddress in many places to identify > locators, servers and peers. Some work has been done in the past couple of > years to reduce the use of these in order to accommodate changes in IP > addresses due to various causes. The class LocatorAddress was created to > help with this and it is able to hold a host name without resolving it until > that resolution is needed to form a tcp/ip connection. > These days we are seeing more and more movement into cloud computing and the > need to accommodate IP address changes is becoming a bigger issue. To that > end we would like to change our primary client/server and WAN communication > interfaces to stop taking InetAddresses and InetSocketAddresses as arguments > and, instead, take something like a LocatorAddress that can hold an > unresolved hostname that our communication implementations will resolve when > needed. > To that end we should also remove the hostname->inetaddress cache in > SocketCreator and rely on the operating system's DNS cache. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7849) change release urls from www.apache.org/dist/ to downloads.apache.org
[ https://issues.apache.org/jira/browse/GEODE-7849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055144#comment-17055144 ] ASF subversion and git services commented on GEODE-7849: Commit d5adfbe45c05b5de3fc21e7d90f80389d8508585 in geode's branch refs/heads/mass-test-run from Owen Nichols [ https://gitbox.apache.org/repos/asf?p=geode.git;h=d5adfbe ] GEODE-7849: change release urls from www.apache.org/dist/ to downloads.apache.org (#4770) > change release urls from www.apache.org/dist/ to downloads.apache.org > - > > Key: GEODE-7849 > URL: https://issues.apache.org/jira/browse/GEODE-7849 > Project: Geode > Issue Type: Improvement > Components: release, website >Reporter: Owen Nichols >Assignee: Owen Nichols >Priority: Major > Fix For: 1.13.0 > > Time Spent: 50m > Remaining Estimate: 0h > > As requested by Apache on Mar 4, 2020 -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7840) Redis pub/sub does not handle binary messages correctly
[ https://issues.apache.org/jira/browse/GEODE-7840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055140#comment-17055140 ] ASF subversion and git services commented on GEODE-7840: Commit 838e0704ed03b0008b4b62f10269b7eed1bd7e18 in geode's branch refs/heads/mass-test-run from Jens Deppe [ https://gitbox.apache.org/repos/asf?p=geode.git;h=838e070 ] GEODE-7840: Redis pub/sub does not handle binary messages correctly (#4762) > Redis pub/sub does not handle binary messages correctly > --- > > Key: GEODE-7840 > URL: https://issues.apache.org/jira/browse/GEODE-7840 > Project: Geode > Issue Type: Bug > Components: redis >Reporter: Jens Deppe >Assignee: Jens Deppe >Priority: Major > Time Spent: 0.5h > Remaining Estimate: 0h > > The new pub/sub functionality currently only considers published messages to > be actual strings and does not account for them being binary data. This > results in messages being corrupted since they end up being UTF-8 encoded. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7808) standardize on use of LocatorAddress/HostAddress for connection formation
[ https://issues.apache.org/jira/browse/GEODE-7808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055139#comment-17055139 ] ASF subversion and git services commented on GEODE-7808: Commit 5c2b959e98a6330ecaeddb4d26e11ea29f7a2d7f in geode's branch refs/heads/mass-test-run from Bruce Schuchardt [ https://gitbox.apache.org/repos/asf?p=geode.git;h=5c2b959 ] GEODE-7808: standardize on use of HostAndPort for creating connections (#4765) * Squashed merge of feature/GEODE-7808 removed HostAddress renamed LocatorAddress to HostAndPort modified TcpClient methods to take a HostAndPort argument instead of InetAddress modified SocketCreator to take a HostAndPort argument instead of InetAddress * GEODE-7808 - standardize on use of HostAndPort for connection formation This continues a previous PR that passed and was approved for merge. This commit raises up several methods from SocketCreator into the TcpSocketCreator interface. This is an intermediate commit. A subsequent commit will refactor TcpSocketCreator to separate the client and server methods for creating server-sockets and client connections to server-sockets. * refactored socket-creators to separate concerns ServerSocketCreator holds methods for non-client comms ClientSocketCreator holds methods that clients should use for comms AdvancedSocketCreator holds methods for people who need to get around the limitations of the other two interfaces * adding missing interface * move code out of inner-classes into first-class classes * renaming interfaces and methods to be less confusing * reinstate SocketCreator ip to hostname cache for performance * changes from review comments > standardize on use of LocatorAddress/HostAddress for connection formation > - > > Key: GEODE-7808 > URL: https://issues.apache.org/jira/browse/GEODE-7808 > Project: Geode > Issue Type: Improvement > Components: membership, messaging >Reporter: Bruce J Schuchardt >Assignee: Bruce J Schuchardt >Priority: Major > Fix For: 1.13.0 > > Time Spent: 7.5h > Remaining Estimate: 0h > > We currently use InetAddress and InetSocketAddress in many places to identify > locators, servers and peers. Some work has been done in the past couple of > years to reduce the use of these in order to accommodate changes in IP > addresses due to various causes. The class LocatorAddress was created to > help with this and it is able to hold a host name without resolving it until > that resolution is needed to form a tcp/ip connection. > These days we are seeing more and more movement into cloud computing and the > need to accommodate IP address changes is becoming a bigger issue. To that > end we would like to change our primary client/server and WAN communication > interfaces to stop taking InetAddresses and InetSocketAddresses as arguments > and, instead, take something like a LocatorAddress that can hold an > unresolved hostname that our communication implementations will resolve when > needed. > To that end we should also remove the hostname->inetaddress cache in > SocketCreator and rely on the operating system's DNS cache. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7808) standardize on use of LocatorAddress/HostAddress for connection formation
[ https://issues.apache.org/jira/browse/GEODE-7808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055148#comment-17055148 ] ASF subversion and git services commented on GEODE-7808: Commit 5c2b959e98a6330ecaeddb4d26e11ea29f7a2d7f in geode's branch refs/heads/mass-test-run from Bruce Schuchardt [ https://gitbox.apache.org/repos/asf?p=geode.git;h=5c2b959 ] GEODE-7808: standardize on use of HostAndPort for creating connections (#4765) * Squashed merge of feature/GEODE-7808 removed HostAddress renamed LocatorAddress to HostAndPort modified TcpClient methods to take a HostAndPort argument instead of InetAddress modified SocketCreator to take a HostAndPort argument instead of InetAddress * GEODE-7808 - standardize on use of HostAndPort for connection formation This continues a previous PR that passed and was approved for merge. This commit raises up several methods from SocketCreator into the TcpSocketCreator interface. This is an intermediate commit. A subsequent commit will refactor TcpSocketCreator to separate the client and server methods for creating server-sockets and client connections to server-sockets. * refactored socket-creators to separate concerns ServerSocketCreator holds methods for non-client comms ClientSocketCreator holds methods that clients should use for comms AdvancedSocketCreator holds methods for people who need to get around the limitations of the other two interfaces * adding missing interface * move code out of inner-classes into first-class classes * renaming interfaces and methods to be less confusing * reinstate SocketCreator ip to hostname cache for performance * changes from review comments > standardize on use of LocatorAddress/HostAddress for connection formation > - > > Key: GEODE-7808 > URL: https://issues.apache.org/jira/browse/GEODE-7808 > Project: Geode > Issue Type: Improvement > Components: membership, messaging >Reporter: Bruce J Schuchardt >Assignee: Bruce J Schuchardt >Priority: Major > Fix For: 1.13.0 > > Time Spent: 7.5h > Remaining Estimate: 0h > > We currently use InetAddress and InetSocketAddress in many places to identify > locators, servers and peers. Some work has been done in the past couple of > years to reduce the use of these in order to accommodate changes in IP > addresses due to various causes. The class LocatorAddress was created to > help with this and it is able to hold a host name without resolving it until > that resolution is needed to form a tcp/ip connection. > These days we are seeing more and more movement into cloud computing and the > need to accommodate IP address changes is becoming a bigger issue. To that > end we would like to change our primary client/server and WAN communication > interfaces to stop taking InetAddresses and InetSocketAddresses as arguments > and, instead, take something like a LocatorAddress that can hold an > unresolved hostname that our communication implementations will resolve when > needed. > To that end we should also remove the hostname->inetaddress cache in > SocketCreator and rely on the operating system's DNS cache. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7808) standardize on use of LocatorAddress/HostAddress for connection formation
[ https://issues.apache.org/jira/browse/GEODE-7808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055149#comment-17055149 ] ASF subversion and git services commented on GEODE-7808: Commit 5c2b959e98a6330ecaeddb4d26e11ea29f7a2d7f in geode's branch refs/heads/mass-test-run from Bruce Schuchardt [ https://gitbox.apache.org/repos/asf?p=geode.git;h=5c2b959 ] GEODE-7808: standardize on use of HostAndPort for creating connections (#4765) * Squashed merge of feature/GEODE-7808 removed HostAddress renamed LocatorAddress to HostAndPort modified TcpClient methods to take a HostAndPort argument instead of InetAddress modified SocketCreator to take a HostAndPort argument instead of InetAddress * GEODE-7808 - standardize on use of HostAndPort for connection formation This continues a previous PR that passed and was approved for merge. This commit raises up several methods from SocketCreator into the TcpSocketCreator interface. This is an intermediate commit. A subsequent commit will refactor TcpSocketCreator to separate the client and server methods for creating server-sockets and client connections to server-sockets. * refactored socket-creators to separate concerns ServerSocketCreator holds methods for non-client comms ClientSocketCreator holds methods that clients should use for comms AdvancedSocketCreator holds methods for people who need to get around the limitations of the other two interfaces * adding missing interface * move code out of inner-classes into first-class classes * renaming interfaces and methods to be less confusing * reinstate SocketCreator ip to hostname cache for performance * changes from review comments > standardize on use of LocatorAddress/HostAddress for connection formation > - > > Key: GEODE-7808 > URL: https://issues.apache.org/jira/browse/GEODE-7808 > Project: Geode > Issue Type: Improvement > Components: membership, messaging >Reporter: Bruce J Schuchardt >Assignee: Bruce J Schuchardt >Priority: Major > Fix For: 1.13.0 > > Time Spent: 7.5h > Remaining Estimate: 0h > > We currently use InetAddress and InetSocketAddress in many places to identify > locators, servers and peers. Some work has been done in the past couple of > years to reduce the use of these in order to accommodate changes in IP > addresses due to various causes. The class LocatorAddress was created to > help with this and it is able to hold a host name without resolving it until > that resolution is needed to form a tcp/ip connection. > These days we are seeing more and more movement into cloud computing and the > need to accommodate IP address changes is becoming a bigger issue. To that > end we would like to change our primary client/server and WAN communication > interfaces to stop taking InetAddresses and InetSocketAddresses as arguments > and, instead, take something like a LocatorAddress that can hold an > unresolved hostname that our communication implementations will resolve when > needed. > To that end we should also remove the hostname->inetaddress cache in > SocketCreator and rely on the operating system's DNS cache. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7853) Remove Unnecessary Calls to Logger.isTraceEnabled From InternalDataSerializer
[
https://issues.apache.org/jira/browse/GEODE-7853?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055153#comment-17055153
]
ASF subversion and git services commented on GEODE-7853:
Commit a9a7d15fd32ff9ef557270f5d25cbce35a6ebcd3 in geode's branch
refs/heads/mass-test-run from Juan José Ramos
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=a9a7d15 ]
GEODE-7853: Remove unused field (#4777)
Remove unused field and avoid unnecessary calls to
Logger.isTraceEnabled().
> Remove Unnecessary Calls to Logger.isTraceEnabled From InternalDataSerializer
> -
>
> Key: GEODE-7853
> URL: https://issues.apache.org/jira/browse/GEODE-7853
> Project: Geode
> Issue Type: Bug
> Components: logging, serialization
>Reporter: Juan Ramos
>Assignee: Juan Ramos
>Priority: Major
> Labels: GeodeCommons
> Fix For: 1.13.0
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> The {{InternalDataSerializer}} class invokes {{Logger.isTraceEnabled()}}
> while
> [initializing|https://github.com/apache/geode/blob/develop/geode-core/src/main/java/org/apache/geode/internal/InternalDataSerializer.java#L249-L251]
> a {{static final}} field that is not used anywhere in the code, remove the
> call and the unused field.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7842) Improve concurrent region creation
[
https://issues.apache.org/jira/browse/GEODE-7842?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055147#comment-17055147
]
ASF subversion and git services commented on GEODE-7842:
Commit a5c8e2bc0e41bf591c24e44e1cbfe48c3ca59e7b in geode's branch
refs/heads/mass-test-run from Jens Deppe
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=a5c8e2b ]
GEODE-7842: Improve concurrent region creation (#4767)
> Improve concurrent region creation
> --
>
> Key: GEODE-7842
> URL: https://issues.apache.org/jira/browse/GEODE-7842
> Project: Geode
> Issue Type: Bug
> Components: gfsh
>Reporter: Jens Deppe
>Assignee: Jens Deppe
>Priority: Major
> Fix For: 1.13.0
>
> Time Spent: 50m
> Remaining Estimate: 0h
>
> The Redis adapter has some tests that result in regions being attempted to be
> created concurrently. The region create call uses {{ifNotExists=true}} but
> occasionally I still see {{RegionCreationException}}s as multiple threads are
> in the {{RegionCreateFunction}} concurrently trying to create the same region.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7808) standardize on use of LocatorAddress/HostAddress for connection formation
[ https://issues.apache.org/jira/browse/GEODE-7808?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055150#comment-17055150 ] ASF subversion and git services commented on GEODE-7808: Commit 5c2b959e98a6330ecaeddb4d26e11ea29f7a2d7f in geode's branch refs/heads/mass-test-run from Bruce Schuchardt [ https://gitbox.apache.org/repos/asf?p=geode.git;h=5c2b959 ] GEODE-7808: standardize on use of HostAndPort for creating connections (#4765) * Squashed merge of feature/GEODE-7808 removed HostAddress renamed LocatorAddress to HostAndPort modified TcpClient methods to take a HostAndPort argument instead of InetAddress modified SocketCreator to take a HostAndPort argument instead of InetAddress * GEODE-7808 - standardize on use of HostAndPort for connection formation This continues a previous PR that passed and was approved for merge. This commit raises up several methods from SocketCreator into the TcpSocketCreator interface. This is an intermediate commit. A subsequent commit will refactor TcpSocketCreator to separate the client and server methods for creating server-sockets and client connections to server-sockets. * refactored socket-creators to separate concerns ServerSocketCreator holds methods for non-client comms ClientSocketCreator holds methods that clients should use for comms AdvancedSocketCreator holds methods for people who need to get around the limitations of the other two interfaces * adding missing interface * move code out of inner-classes into first-class classes * renaming interfaces and methods to be less confusing * reinstate SocketCreator ip to hostname cache for performance * changes from review comments > standardize on use of LocatorAddress/HostAddress for connection formation > - > > Key: GEODE-7808 > URL: https://issues.apache.org/jira/browse/GEODE-7808 > Project: Geode > Issue Type: Improvement > Components: membership, messaging >Reporter: Bruce J Schuchardt >Assignee: Bruce J Schuchardt >Priority: Major > Fix For: 1.13.0 > > Time Spent: 7.5h > Remaining Estimate: 0h > > We currently use InetAddress and InetSocketAddress in many places to identify > locators, servers and peers. Some work has been done in the past couple of > years to reduce the use of these in order to accommodate changes in IP > addresses due to various causes. The class LocatorAddress was created to > help with this and it is able to hold a host name without resolving it until > that resolution is needed to form a tcp/ip connection. > These days we are seeing more and more movement into cloud computing and the > need to accommodate IP address changes is becoming a bigger issue. To that > end we would like to change our primary client/server and WAN communication > interfaces to stop taking InetAddresses and InetSocketAddresses as arguments > and, instead, take something like a LocatorAddress that can hold an > unresolved hostname that our communication implementations will resolve when > needed. > To that end we should also remove the hostname->inetaddress cache in > SocketCreator and rely on the operating system's DNS cache. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7840) Redis pub/sub does not handle binary messages correctly
[ https://issues.apache.org/jira/browse/GEODE-7840?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055151#comment-17055151 ] ASF subversion and git services commented on GEODE-7840: Commit 838e0704ed03b0008b4b62f10269b7eed1bd7e18 in geode's branch refs/heads/mass-test-run from Jens Deppe [ https://gitbox.apache.org/repos/asf?p=geode.git;h=838e070 ] GEODE-7840: Redis pub/sub does not handle binary messages correctly (#4762) > Redis pub/sub does not handle binary messages correctly > --- > > Key: GEODE-7840 > URL: https://issues.apache.org/jira/browse/GEODE-7840 > Project: Geode > Issue Type: Bug > Components: redis >Reporter: Jens Deppe >Assignee: Jens Deppe >Priority: Major > Time Spent: 0.5h > Remaining Estimate: 0h > > The new pub/sub functionality currently only considers published messages to > be actual strings and does not account for them being binary data. This > results in messages being corrupted since they end up being UTF-8 encoded. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7828) Convert backing store for Redis Hashes and Sets to single regions
[
https://issues.apache.org/jira/browse/GEODE-7828?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055154#comment-17055154
]
ASF subversion and git services commented on GEODE-7828:
Commit f0982cdedf1b122f734919d0647434cca33ec254 in geode's branch
refs/heads/mass-test-run from Nabarun Nag
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=f0982cd ]
Revert "GEODE-7828: Convert backing store for Redis Hashes and Sets to single
regions (#4745)" (#4780)
This reverts commit 2f6bf013368df5a4b5efe68162a4953f9a88bbf2.
> Convert backing store for Redis Hashes and Sets to single regions
> -
>
> Key: GEODE-7828
> URL: https://issues.apache.org/jira/browse/GEODE-7828
> Project: Geode
> Issue Type: Improvement
> Components: redis
>Reporter: Jens Deppe
>Assignee: Jens Deppe
>Priority: Major
> Fix For: 1.13.0
>
> Time Spent: 1.5h
> Remaining Estimate: 0h
>
> Currently the Redis adapter creates separate regions for each new hash or
> set. This can be a significant performance impact when many hashes are being
> created. In addition, hashes are conventionally often named with colons (':')
> in their name - for example {{user:1000}}. This does not work as region names
> cannot contain colons.
> The work here will create a single region for hashes and a single region for
> sets.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7851) Pulse should support OAuth2 authorization code flow
[ https://issues.apache.org/jira/browse/GEODE-7851?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055156#comment-17055156 ] ASF subversion and git services commented on GEODE-7851: Commit 69ade72db301d2177b15f422342d198420021b73 in geode's branch refs/heads/mass-test-run from Jinmei Liao [ https://gitbox.apache.org/repos/asf?p=geode.git;h=69ade72 ] GEODE-7851: configure pulse security using java code instead of using xml (#4772) Co-authored-by: Joris Melchior > Pulse should support OAuth2 authorization code flow > --- > > Key: GEODE-7851 > URL: https://issues.apache.org/jira/browse/GEODE-7851 > Project: Geode > Issue Type: New Feature > Components: pulse >Reporter: Jinmei Liao >Priority: Major > Time Spent: 1h 20m > Remaining Estimate: 0h > > Instead of using username/password to log in to pulse, pulse should redirect > to a configured authentication provider to get access token to login. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7849) change release urls from www.apache.org/dist/ to downloads.apache.org
[ https://issues.apache.org/jira/browse/GEODE-7849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055155#comment-17055155 ] ASF subversion and git services commented on GEODE-7849: Commit d5adfbe45c05b5de3fc21e7d90f80389d8508585 in geode's branch refs/heads/mass-test-run from Owen Nichols [ https://gitbox.apache.org/repos/asf?p=geode.git;h=d5adfbe ] GEODE-7849: change release urls from www.apache.org/dist/ to downloads.apache.org (#4770) > change release urls from www.apache.org/dist/ to downloads.apache.org > - > > Key: GEODE-7849 > URL: https://issues.apache.org/jira/browse/GEODE-7849 > Project: Geode > Issue Type: Improvement > Components: release, website >Reporter: Owen Nichols >Assignee: Owen Nichols >Priority: Major > Fix For: 1.13.0 > > Time Spent: 50m > Remaining Estimate: 0h > > As requested by Apache on Mar 4, 2020 -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7841) Add benchmarking scripts for Geode-Redis module
[ https://issues.apache.org/jira/browse/GEODE-7841?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055152#comment-17055152 ] ASF subversion and git services commented on GEODE-7841: Commit b57456e2cd9f179cf018adaf676a9b78bdc98270 in geode's branch refs/heads/mass-test-run from Jens Deppe [ https://gitbox.apache.org/repos/asf?p=geode.git;h=b57456e ] GEODE-7841 Add benchmarking scripts for Geode-Redis module (#4763) > Add benchmarking scripts for Geode-Redis module > --- > > Key: GEODE-7841 > URL: https://issues.apache.org/jira/browse/GEODE-7841 > Project: Geode > Issue Type: New Feature > Components: benchmarks, redis >Reporter: Raymond Ingles >Assignee: Jens Deppe >Priority: Major > Time Spent: 0.5h > Remaining Estimate: 0h > > Add the following scripts to the Geode-Redis module, in the "performanceTest" > directory: > * aggregator.sh runs the actual tests > * benchmark.sh handles validating the redis server (e.g. starting and > stopping Geode Redis if needed), then runs aggregator.sh > * shacompare.sh takes two git SHAs as arguments and runs benchmark.sh on > each of them, generating separate output files for each. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7851) Pulse should support OAuth2 authorization code flow
[ https://issues.apache.org/jira/browse/GEODE-7851?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055145#comment-17055145 ] ASF subversion and git services commented on GEODE-7851: Commit 69ade72db301d2177b15f422342d198420021b73 in geode's branch refs/heads/mass-test-run from Jinmei Liao [ https://gitbox.apache.org/repos/asf?p=geode.git;h=69ade72 ] GEODE-7851: configure pulse security using java code instead of using xml (#4772) Co-authored-by: Joris Melchior > Pulse should support OAuth2 authorization code flow > --- > > Key: GEODE-7851 > URL: https://issues.apache.org/jira/browse/GEODE-7851 > Project: Geode > Issue Type: New Feature > Components: pulse >Reporter: Jinmei Liao >Priority: Major > Time Spent: 1h 20m > Remaining Estimate: 0h > > Instead of using username/password to log in to pulse, pulse should redirect > to a configured authentication provider to get access token to login. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (GEODE-7860) Indexed query benchmarks fail with EOFException
Donal Evans created GEODE-7860:
--
Summary: Indexed query benchmarks fail with EOFException
Key: GEODE-7860
URL: https://issues.apache.org/jira/browse/GEODE-7860
Project: Geode
Issue Type: Bug
Components: benchmarks, querying
Reporter: Donal Evans
{noformat}
org.apache.geode.benchmark.tests.ReplicatedIndexedQueryBenchmark > run() FAILED
java.util.concurrent.CompletionException: java.lang.RuntimeException:
java.rmi.UnmarshalException: Error unmarshaling return header; nested exception
is:
java.io.EOFException
Caused by:
java.lang.RuntimeException: java.rmi.UnmarshalException: Error
unmarshaling return header; nested exception is:
java.io.EOFException
Caused by:
java.rmi.UnmarshalException: Error unmarshaling return header;
nested exception is:
java.io.EOFException
Caused by:
java.io.EOFException
16 tests completed, 1 failed
org.apache.geode.benchmark.tests.PartitionedIndexedQueryBenchmark > run() FAILED
java.util.concurrent.CompletionException: java.lang.RuntimeException:
java.rmi.UnmarshalException: Error unmarshaling return header; nested exception
is:
java.io.EOFException
Caused by:
java.lang.RuntimeException: java.rmi.UnmarshalException: Error
unmarshaling return header; nested exception is:
java.io.EOFException
Caused by:
java.rmi.UnmarshalException: Error unmarshaling return header;
nested exception is:
java.io.EOFException
Caused by:
java.io.EOFException
16 tests completed, 1 failed
{noformat}
Failure seen in this run:
https://concourse.apachegeode-ci.info/teams/main/pipelines/apache-develop-main/jobs/Benchmark_base/builds/146
but was passing in the following run.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (GEODE-7861) Improve error reporting when a member is unable to contact a locator
Dan Smith created GEODE-7861:
Summary: Improve error reporting when a member is unable to
contact a locator
Key: GEODE-7861
URL: https://issues.apache.org/jira/browse/GEODE-7861
Project: Geode
Issue Type: Improvement
Components: membership
Reporter: Dan Smith
When a member is unable to contact a join the system due to a failure to
contact a locator, we could be a lot more specific, which would help users
debug issues in their environment. We currently print out
{noformat}
Unable to join the distributed system. Operation either timed out, was stopped
or Locator does not exist.
{noformat}
We should include the underlying exception from the last locator we tried to
contact as a cause, so that users can see why it failed (DNS error, ???).
Perhaps also the list of locators we tried to contact.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7857) PR.clear's eventID should not be NULL
[ https://issues.apache.org/jira/browse/GEODE-7857?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055248#comment-17055248 ] ASF subversion and git services commented on GEODE-7857: Commit cc32e722197037d408509ea13b49087ca898455f in geode's branch refs/heads/feature/GEODE-7857 from zhouxh [ https://gitbox.apache.org/repos/asf?p=geode.git;h=cc32e72 ] GEODE-7857: PR.clear's event id should be created and used in BR > PR.clear's eventID should not be NULL > - > > Key: GEODE-7857 > URL: https://issues.apache.org/jira/browse/GEODE-7857 > Project: Geode > Issue Type: Improvement >Reporter: Xiaojian Zhou >Assignee: Xiaojian Zhou >Priority: Major > Labels: GeodeCommons > > In earlier implementation for PR.clear, I set eventid to be null for the time > being. The event id should be the same at both PR and BR level. but not null. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7859) Fix Geode Redis publish/subscribe distributed test
[ https://issues.apache.org/jira/browse/GEODE-7859?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055282#comment-17055282 ] ASF subversion and git services commented on GEODE-7859: Commit ec0c1a23d6c65d22c03f60aadea4ae6097d46767 in geode's branch refs/heads/develop from Ray Ingles [ https://gitbox.apache.org/repos/asf?p=geode.git;h=ec0c1a2 ] GEODE-7859 Increase timeout for channel subscription (#4782) > Fix Geode Redis publish/subscribe distributed test > -- > > Key: GEODE-7859 > URL: https://issues.apache.org/jira/browse/GEODE-7859 > Project: Geode > Issue Type: Test > Components: redis >Reporter: Raymond Ingles >Priority: Major > Time Spent: 20m > Remaining Estimate: 0h > > The Pub/Sub Dunit tests fails intermittently. Timeout has been increased to > ensure channel subscriptions complete. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7854) Update Geode Redis benchmark scripts
[ https://issues.apache.org/jira/browse/GEODE-7854?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055283#comment-17055283 ] ASF subversion and git services commented on GEODE-7854: Commit 9e093c1c69547aa3b4d24a11dd75a78fcf0c6768 in geode's branch refs/heads/develop from Ray Ingles [ https://gitbox.apache.org/repos/asf?p=geode.git;h=9e093c1 ] GEODE-7854 Update Geode Redis benchmark scripts (#4779) > Update Geode Redis benchmark scripts > > > Key: GEODE-7854 > URL: https://issues.apache.org/jira/browse/GEODE-7854 > Project: Geode > Issue Type: New Feature > Components: redis >Reporter: Raymond Ingles >Priority: Major > Time Spent: 20m > Remaining Estimate: 0h > > The benchmark scripts have been updated to display the fastest, slowest, and > ~95%ile response times in addition to the average requests per second from > before. The scripts are documented in the README.md in the geode-redis > directory. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (GEODE-7862) Add Tests for Redis DEL command
Raymond Ingles created GEODE-7862: - Summary: Add Tests for Redis DEL command Key: GEODE-7862 URL: https://issues.apache.org/jira/browse/GEODE-7862 Project: Geode Issue Type: Test Components: redis Reporter: Raymond Ingles Adds tests (including concurrency) for the Redis DEL command, which was untested before. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055600#comment-17055600
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055599#comment-17055599
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055597#comment-17055597
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055601#comment-17055601
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055598#comment-17055598
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055602#comment-17055602
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055604#comment-17055604
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055606#comment-17055606
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055607#comment-17055607
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055608#comment-17055608
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055603#comment-17055603
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055605#comment-17055605
]
ASF subversion and git services commented on GEODE-7727:
Commit c8413592e5573f675c538c63ef9ee9f97a349e73 in geode's branch
refs/heads/develop from Mario Ivanac
[ https://gitbox.apache.org/repos/asf?p=geode.git;h=c841359 ]
GEODE-7727: modify sender thread to detect relese of connection (#4751)
* GEODE-7727: modify sender thread to detect relese of connection
* GEODE-7727: Update solution only for shared connections
* GEODE-7727: added test
* GEODE-7727: update ater comments
* GEODE-7727: update test
* GEODE-7727: fix for async write hanging
* GEODE-7727: Test of region operations in the face of closed connections
Adding a test for what happens to region operations when a connection is closed
out from under the system. This test hangs without the changes to let the
reader thread keep running.
Fix to test
* GEODE-7727: Preventing a double release of the input buffer
The releaseInputBuffer method was not thread safe. If it is called
concurrently, it will end up being released twice, which will add the buffer to
to the buffer pool twice. Later, this could result in two threads using the
same buffer, resulting in corruption of the buffer.
With the changes for GEODE-7727, we made it likely that releaseInputBuffer
would be called concurrently. If a member departs, one thread will call
Connection.close. Connection.close will close the socket and call
releaseInputBuffer. However, closing the socket will wake up the reader thread,
which will also call releaseInputBuffer concurrently.
Making releaseInputBuffer thread safe by introducing a lock.
* GEODE-7727: update after merge
* GEODE-7727: update test name
Co-authored-by: Dan Smith
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Resolved] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mario Ivanac resolved GEODE-7727.
-
Resolution: Fixed
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (GEODE-7727) Geode P2P connection hanging
[
https://issues.apache.org/jira/browse/GEODE-7727?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Mario Ivanac updated GEODE-7727:
Fix Version/s: 1.13.0
> Geode P2P connection hanging
>
>
> Key: GEODE-7727
> URL: https://issues.apache.org/jira/browse/GEODE-7727
> Project: Geode
> Issue Type: Bug
>Reporter: Mario Ivanac
>Assignee: Mario Ivanac
>Priority: Major
> Labels: needs-review, pull-request-available
> Fix For: 1.13.0
>
> Time Spent: 3.5h
> Remaining Estimate: 0h
>
> {color:#172b4d}Geode P2P handshake reader stops listening to it's socket once
> the handshake between 2 peers is established. This seems to be a design
> choice.
> {color}
> {color:#172b4d}The problem is when the connection gets killed (TCP FIN).
> Since nothing is listening on the socket, nothing will get that FIN package
> and close the connection. The connection is left hanging (CLOSE-WAIT state).
> The peers are then unable to establish proper P2P communication later.{color}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
