[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15034081#comment-15034081
]
T Jake Luciani commented on CASSANDRA-10592:
[~aweisberg] if you could squash and rebase
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15034273#comment-15034273
]
Ariel Weisberg commented on CASSANDRA-10592:
Squashed and rebased. Tests are running.
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15033840#comment-15033840
]
T Jake Luciani commented on CASSANDRA-10592:
Hmm why does this OOM on a read command when
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15034026#comment-15034026
]
Sebastian Estevez commented on CASSANDRA-10592:
---
The stress insert=1 is just to prepare
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15032766#comment-15032766
]
Sebastian Estevez commented on CASSANDRA-10592:
---
Confirmed, the error is gone. The
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15031814#comment-15031814
]
Sebastian Estevez commented on CASSANDRA-10592:
---
I'll test it today. Thanks!
>
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022687#comment-15022687
]
T Jake Luciani commented on CASSANDRA-10592:
I just want reallocate to not reallocate if
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022218#comment-15022218
]
T Jake Luciani commented on CASSANDRA-10592:
[~sebastian.este...@datastax.com] can you
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022358#comment-15022358
]
Ariel Weisberg commented on CASSANDRA-10592:
bq. The new property seems like something no
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022709#comment-15022709
]
Ariel Weisberg commented on CASSANDRA-10592:
OK done.
> IllegalArgumentException in
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022730#comment-15022730
]
T Jake Luciani commented on CASSANDRA-10592:
Thanks +1, I'll commit once the tests re-run
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022408#comment-15022408
]
Ariel Weisberg commented on CASSANDRA-10592:
Huh, so actually it's mostly taken care of.
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022604#comment-15022604
]
Ariel Weisberg commented on CASSANDRA-10592:
bq. You call doFlush(0) in a few places in
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15022435#comment-15022435
]
T Jake Luciani commented on CASSANDRA-10592:
You call doFlush(0) in a few places in the
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15004523#comment-15004523
]
Joshua McKenzie commented on CASSANDRA-10592:
-
[~tjake] to review 3.0 branch since
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14992388#comment-14992388
]
Ariel Weisberg commented on CASSANDRA-10592:
[~benedict] The utests look good. The dtests
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14989414#comment-14989414
]
Benedict commented on CASSANDRA-10592:
--
Patch LGTM.
> IllegalArgumentException in
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14990161#comment-14990161
]
Ariel Weisberg commented on CASSANDRA-10592:
Squashed and a new branch for the merge to
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987739#comment-14987739
]
Ariel Weisberg commented on CASSANDRA-10592:
At what threshold should we switch to
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987927#comment-14987927
]
Ariel Weisberg commented on CASSANDRA-10592:
It's not max object size. The test increases
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987730#comment-14987730
]
Ariel Weisberg commented on CASSANDRA-10592:
bq. I think increasing by 50% of the current
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987717#comment-14987717
]
Benedict commented on CASSANDRA-10592:
--
* Could we keep the testing code isolated from the
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987816#comment-14987816
]
Ariel Weisberg commented on CASSANDRA-10592:
Just an FYI, switching from doubling to 50%
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987874#comment-14987874
]
Benedict commented on CASSANDRA-10592:
--
Size of data GC'd should not affect OOM. More likely the
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14987734#comment-14987734
]
Benedict commented on CASSANDRA-10592:
--
The amount of wasted memory is smaller. With > 1Gb
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985418#comment-14985418
]
Benedict commented on CASSANDRA-10592:
--
We should probably switch to a consistent {+ 1/2}
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985673#comment-14985673
]
Benedict commented on CASSANDRA-10592:
--
That sounds pretty reasonable to me, yeah.
>
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985646#comment-14985646
]
Ariel Weisberg commented on CASSANDRA-10592:
Maybe we should use the same strategy that
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14985833#comment-14985833
]
Ariel Weisberg commented on CASSANDRA-10592:
Pushed an update with what I am thinking.
>
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14979309#comment-14979309
]
Ariel Weisberg commented on CASSANDRA-10592:
[~sebastian.este...@datastax.com] I having a
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14976650#comment-14976650
]
Ariel Weisberg commented on CASSANDRA-10592:
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14976570#comment-14976570
]
Sebastian Estevez commented on CASSANDRA-10592:
---
By the way now I'm seeing it on 3.0
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14976569#comment-14976569
]
Sebastian Estevez commented on CASSANDRA-10592:
---
Right sorry about that, let me correct
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14974894#comment-14974894
]
Ariel Weisberg commented on CASSANDRA-10592:
[~sebastian.este...@datastax.com] I was able
[
https://issues.apache.org/jira/browse/CASSANDRA-10592?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14974720#comment-14974720
]
Ariel Weisberg commented on CASSANDRA-10592:
This looks like Integer overflow in DOB
35 matches
Mail list logo
