[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16224256#comment-16224256
]
stack commented on HBASE-15071:
---
Assigned [~chenheng] because did a load of work on this one.
> Cleanup
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16224253#comment-16224253
]
stack commented on HBASE-15071:
---
It looks like the bulk of the patch attached here went in with
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16224219#comment-16224219
]
stack commented on HBASE-15071:
---
HBASE-18770 removes bypass from all MasterObserver operations except
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16208760#comment-16208760
]
Duo Zhang commented on HBASE-15071:
---
Got it sir.
> Cleanup bypass semantic in MasterCoprocessorHost
>
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16208755#comment-16208755
]
stack commented on HBASE-15071:
---
[~Apache9] I was going to have a go after HBASE-19007 sir.
> Cleanup
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16208747#comment-16208747
]
Duo Zhang commented on HBASE-15071:
---
Any progress here? After changing the title of HBASE-18770 I
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16157052#comment-16157052
]
Chia-Ping Tsai commented on HBASE-15071:
Any progress? RegionCoprocessorHost has similar issue.
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15085961#comment-15085961
]
stack commented on HBASE-15071:
---
[~apurtell]
bq. The coprocessor says it has taken care of everything,
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15085980#comment-15085980
]
Matteo Bertozzi commented on HBASE-15071:
-
I don't think the problem is procedure.
the client is
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15086047#comment-15086047
]
Andrew Purtell commented on HBASE-15071:
Bypass should only be called from a CP that knows what
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15086053#comment-15086053
]
Andrew Purtell commented on HBASE-15071:
bq. So we should make all "pre" actions respect bypass
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15086058#comment-15086058
]
Andrew Purtell commented on HBASE-15071:
bq. the problem here is. what bypass mean for things
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15086729#comment-15086729
]
Heng Chen commented on HBASE-15071:
---
Currently, the usages of bypass in MasterCoprocessorHost:
* we
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15083635#comment-15083635
]
Matteo Bertozzi commented on HBASE-15071:
-
the bypass() documentation says
{quote}Call to
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15084610#comment-15084610
]
Heng Chen commented on HBASE-15071:
---
Sounds reasonable. Let me take it, OK?
> Cleanup bypass semantic
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15084985#comment-15084985
]
Andrew Purtell commented on HBASE-15071:
Interrupts could work as long as they're not thrown back
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15085071#comment-15085071
]
Andrew Purtell commented on HBASE-15071:
bq. I don't think it is a good way, even an exception to
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15084959#comment-15084959
]
Heng Chen commented on HBASE-15071:
---
{quote}
A bunch of methods in MasterCoprocessorHost on the 'pre'
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15085064#comment-15085064
]
Heng Chen commented on HBASE-15071:
---
{quote}
Interrupts could work as long as they're not thrown back
[
https://issues.apache.org/jira/browse/HBASE-15071?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15085087#comment-15085087
]
Heng Chen commented on HBASE-15071:
---
{quote}
The point of bypass is to skip further core code
20 matches
Mail list logo
