I don't think you need `IDisposable` anymore since you removed all the code of
`Dispose()` below, though I don't think it really hurts anything.
[ Full content available at: https://github.com/apache/geode-native/pull/428 ]
This message was relayed via gitbox.apache.org for
Just want to say I like this section, much easier to read
[ Full content available at: https://github.com/apache/geode-native/pull/428 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Stray comment
[ Full content available at: https://github.com/apache/geode-native/pull/428 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Something like:
```c++
main() {
...
transactionManager.begin();
try {
for (auto& key : batch) {
auto value = getValueFromExternalSystem(key);
cache.put(key, value);
}
transactionManager.commit();
} catch ( ... ) {
transactionManager.rollback();
}
}
int32_t
[ pull request closed by gesterzhou ]
[ Full content available at: https://github.com/apache/geode/pull/3040 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
I'm glad you mentioned this because I was used to the JUnit4CacheTestCase
behavior of disabling shutdown of the DistributedSystem when the cache is
closed. The new DistributedRule doesn't appear to do this.
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message
Fair point!
[ Full content available at: https://github.com/apache/geode/pull/3049 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Sure, that seems like a simple enough thing to fix.
[ Full content available at: https://github.com/apache/geode/pull/3049 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Yes - even socket.close() can block when there are network problems. That's
why we usually spin off AsyncCloser threads to close sockets.
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
That was left in the root `build.gradle` because that's where it had
effectively been before.
Old style: publication was shoved into a `subprojects {` configuration, and
`askpass` was outside that configuration, implicitly applied to the root
project that called the `apply from:
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
[ pull request closed by galen-pivotal ]
[ Full content available at: https://github.com/apache/geode/pull/3026 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
* New method added to get the number of members for each role.
* Calculating unique addresses for number of members caused an issue
when all VMs are in the same physical machine.
[ Full content available at: https://github.com/apache/geode-benchmarks/pull/33
]
This message was
Introducing a new language to the Geode project raises some questions that we
need to answer as a community before adopting this proposal:
- When is it appropriate to use Kotlin? When should we prefer Java?
- Are there a sufficient number of committers with Kotlin experience to
maintain the
Yes, definitely would be moving a lot of this stuff out of controller code
[ Full content available at: https://github.com/apache/geode/pull/3049 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
I think this and the other modified tests can be renamed now since they no
longer assert that specific comparators are being used. Instead it's just
asserting that we can make comparisons against booleans, different numeric
values and date/time objects
[ Full content available at:
I tried this but Locator isn't AutoClosable & the compiler objected to the
change.
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
That's what the code in Connection was doing & I preserved it. I think the
deal is that in some situations we don't consume anything from the buffer & its
position hasn't advanced. In that case we need to read more data into the
buffer.
[ Full content available at:
[ pull request closed by kirklund ]
[ Full content available at: https://github.com/apache/geode/pull/3045 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Hey @kohlmu-pivotal ,
Could you please rephrase this? I don't fully understand
> I wonder... would it not make more sense to rather treat the REST Controller
> as an endpoint.. rather put the login inside the controller. Why could we not
> have access to the ConfigurationService here?
[
Hey @kohlmu-pivotal ,
Could you please rephrase this? I don't fully understand. Specially, could you
use the phrase "it is **better to** X **instead** of Y"
> I wonder... would it not make more sense to rather treat the REST Controller
> as an endpoint.. rather put the login inside the
I realize that these are test cases so maybe you can ignore this request.
[ Full content available at: https://github.com/apache/geode/pull/3044 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Add IgnoredException to doesNotWaitForPreviousInstanceOfOnlineServer.
While testing for GEODE-6232, the only reproducible problem appears to
be a DistributedSystemDisconnectedException in a background thread due
to doesNotWaitForPreviousInstanceOfOnlineServer disconnecting during
the handling of
[ pull request closed by rhoughton-pivot ]
[ Full content available at: https://github.com/apache/geode/pull/3048 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
[ Full content available at: https://github.com/apache/geode-benchmarks/pull/34
]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
balesh2 opened a new pull request #34: update script parameters in readme
URL: https://github.com/apache/geode-benchmarks/pull/34
This is an automated message from the Apache Git Service.
To respond to the message, please
balesh2 closed pull request #29: Use a dedicated separate profile to work with
AWS.
URL: https://github.com/apache/geode-benchmarks/pull/29
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this
[ pull request closed by balesh2 ]
[ Full content available at: https://github.com/apache/geode-benchmarks/pull/29
]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
fixed - old habits from the days of using JUnit4CacheTestCase. That base class
configures the cache to not disconnect the DistributedSystem.
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
typo: should be `registerFunctionLock`
[ Full content available at: https://github.com/apache/geode/pull/2807 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
…e initialization
Thank you for submitting a contribution to Apache Geode.
In order to streamline the review of the contribution we ask you
to ensure the following steps have been taken:
### For all changes:
- [ ] Is there a JIRA ticket associated with this PR? Is it referenced in the
commit
I checked StressNewTestOpenJDK8 but dunit-hangs.txt and the callstack files are
all empty. There is no actionable info of any sort. Mark and I also ran this
dunit test several 1000 times in GCP over the last couple days without it ever
running into any problems or hanging (just the one time
Even though the callstacks are empty and dunit-hangs.txt shows no tests, the
file geode-core/build/repeatDistributedTest/repeatDistributedTest-progress.txt
does show 683 lines -- half of those are "Starting" and half with one missing
are "Completed" -- so I think there was a dunit test still
[ pull request closed by aditya87 ]
[ Full content available at: https://github.com/apache/geode/pull/3049 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
[ pull request closed by balesh2 ]
[ Full content available at: https://github.com/apache/geode-benchmarks/pull/34
]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
balesh2 closed pull request #34: update script parameters in readme
URL: https://github.com/apache/geode-benchmarks/pull/34
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As this is a foreign
Thank you for submitting a contribution to Apache Geode.
In order to streamline the review of the contribution we ask you
to ensure the following steps have been taken:
### For all changes:
- [x] Is there a JIRA ticket associated with this PR? Is it referenced in the
commit message?
- [x] Has
@metatype I agree.
Maybe the Kotlin experiment has a more limited scope like a non-critical Geode
module, this way the effect is contained. Even though, I wholeheartedly believe
that some of the concepts that Kotlin embraces by default, like immutability
and non-null, could benefit
@metatype I agree.
Maybe the Kotlin experiment has a more limited scope like a non-critical Geode
module, this way the effect is contained. Even though, I wholeheartedly believe
that some of the concepts that Kotlin embraces by default, like immutability
and non-null, could benefit
@mhansonp I hit the '+' on the StressNewTestOpenJDK8 but couldn't get it to
rerun the test that way. I pushed an empty commit to retrigger all test jobs. I
also ran `./gradlew repeatDistributedTest --tests
PersistentPartitionedRegionRegressionTest -Prepeat=50` locally without any
issues. I'm
[ pull request closed by bschuchardt ]
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
Signed-off-by: Matthew Reddington
[ Full content available at: https://github.com/apache/geode-native/pull/429 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
I checked StressNewTestOpenJDK8 but dunit-hangs.txt and the callstack files are
all empty. There is no actionable info of any sort. Mark and I also ran this
dunit test several 1000 times in GCP over the last couple days without it ever
running into any problems or hanging.
[ Full content
@ mhansonp is there anything interesting that you can find in the results of
StressNewTestOpenJDK8? I downloaded the artifacts, but the dunit-hangs.txt
shows no tests were run and the callstack files are empty. It looks like the
gradle target hung but I don't think it actually started any dunit
> @mhansonp is there anything interesting that you can find in the results of
> StressNewTestOpenJDK8? I downloaded the artifacts, but the dunit-hangs.txt
> shows no tests were run and the callstack files are empty. It looks like the
> gradle target hung but I don't think it actually started
`// TODO error handling`
[ Full content available at: https://github.com/apache/geode-native/pull/408 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
@mhansonp is there anything interesting that you can find in the results of
StressNewTestOpenJDK8? I downloaded the artifacts, but the dunit-hangs.txt
shows no tests were run and the callstack files are empty. It looks like the
gradle target hung but I don't think it actually started any dunit
@pdxcodemonkey and I followed the chain up and the comment looks like it was
there to explain the `break;` statement. The errors are not handled at a
higher level.
[ Full content available at: https://github.com/apache/geode-native/pull/408 ]
This message was relayed via gitbox.apache.org for
nabarunnag closed pull request #33: GEODE-6243: New method to get number of
members.
URL: https://github.com/apache/geode-benchmarks/pull/33
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As
[ pull request closed by nabarunnag ]
[ Full content available at: https://github.com/apache/geode-benchmarks/pull/33
]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
[ pull request closed by gesterzhou ]
[ Full content available at: https://github.com/apache/geode/pull/3037 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
I'm leaving it but adding a comment. It took me a while to find this debug
setting & I don't want to have to do that again.
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
I removed use of CacheFactory.getAnyInstance().
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
[ pull request closed by upthewaterspout ]
[ Full content available at: https://github.com/apache/geode-benchmarks/pull/28
]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
upthewaterspout closed pull request #28: Adding additional metrics of standard
error
URL: https://github.com/apache/geode-benchmarks/pull/28
This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:
As
good catch!
[ Full content available at: https://github.com/apache/geode/pull/2807 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
> @pdxrunner is this PR going to see any more effort or should it be closed?
Thanks for pinging me on this. I should have closed it out when I superseded it
with https://github.com/apache/geode/pull/2846
[ Full content available at: https://github.com/apache/geode/pull/2825 ]
This message was
sounds good.
[ Full content available at: https://github.com/apache/geode/pull/3036 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
[ pull request closed by pdxrunner ]
[ Full content available at: https://github.com/apache/geode/pull/2825 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
@Petahhh the intent of comment is to portrait that I prefer that the REST
endpoint is "DUMB" and is only responsible invoking a service call.
```ConfigurationService.createRegion("name","type",attributes)```
The fact that we use a function call to create is a detail that we have leaked.
One
- demonstrates commit and rollback of transactions
- handles exceptions
[ Full content available at: https://github.com/apache/geode-native/pull/430 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
This part of fix will resolve the root cause which caused the revert.
[ Full content available at: https://github.com/apache/geode/pull/3044 ]
This message was relayed via gitbox.apache.org for
notifications@geode.apache.org
I personally would be ok with those changes being requested, let's see if
others have your reservations. Seems like it wouldn't be too crazy to throw
some error and then rollback in something slightly more realistic.
[ Full content available at: https://github.com/apache/geode-native/pull/430
63 matches
Mail list logo
