http://queue.acm.org/detail.cfm?id=1594206 .
I am more with Dhruba on this one. Let contrib stay there that are
maintained (I assume a committer owns it) and those that are not
maintained and fall behind will be dropped if they are not done by a
certain deadline before a release. That will weed out the old contribs
but keeps the project toget
hadoop also uses the hostnames. if a host is multi-homed - it's
hostname is a better identifier (which still allows it to use
different nics/ips for actual traffic). it can help in the case the
cluster is migrated for example (all the ips change). one could have
the same hostname resolve to differe
Hi JD,
I have more question in regards to HLog and more concretely the
underlaying SequenceFileLogWriter. This is what is called from HLog
(assuming that is current, you said you changed things to hflush() in
the replication context?):
@Override
public void sync() throws IOException {
thi
On that note, should we make HLog.Writer also implement Syncable to
pass on the hsync() and hflush() calls? That would be in trunk
obviously.
On Fri, Jan 29, 2010 at 2:17 PM, Lars George wrote:
> Hi JD,
>
> I have more question in regards to HLog and more concretely the
> underlaying SequenceFile
@Joy: The info stored in .META. for various regions as well as in the ephemeral
nodes for region servers in zookeeper are both already IP address based. So
doesn't look like multi-homing and/or the other flexibilities you mention were
a design goal as far as I can tell.
Regarding: <<< doesn't t
Review how we handle addresses in HBase
---
Key: HBASE-2174
URL: https://issues.apache.org/jira/browse/HBASE-2174
Project: Hadoop HBase
Issue Type: Improvement
Reporter: Jean-Daniel Cryans
I just created https://issues.apache.org/jira/browse/HBASE-2174
We handle addresses in different ways depending on which part of the
code you're in. We should correct that everywhere by implementing a
solution that also solves what you guys are seeing.
J-D
On Fri, Jan 29, 2010 at 8:33 AM, Kannan
@Kannan - Karthik's mail said the reverse lookup happens in the RS
(not the master). the master simply tried to match the offered
hostname.
i dont know whose reading is right - but if it's the RS - i didn't
understand why that wasn't just the reverse lookup done once at
bootstrap time (which would
The master does another lookup independent of the region server using the
hostname given by the region server:
ServerManager.java, regionServerReport() does:
HServerInfo storedInfo = serversToServerInfo.get(info.getServerName()); //
info.getServerName() is hostname
Which eventually does:
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806434#action_12806434
]
Jean-Daniel Cryans commented on HBASE-2174:
---
One example of weirdness is when the
hmmm .. if the master doesn't need the RS ip address at this point -
seems like it should be able to use the hostname offered by the RS
directly?
On Fri, Jan 29, 2010 at 10:44 AM, Karthik Ranganathan
wrote:
> The master does another lookup independent of the region server using the
> hostname gi
Yup totally - either name or ip would work. Not sure if there is a pro or a con
to choosing either one - but thought it better to use the ip as that always
remains the same (no resolve required) and used to open the sockets.
@jd-cryans: Saw your JIRA update: "One example of weirdness is when the
It would be nice to log all those kind of issues and all your findings
in the Jira to start the discussion up there.
J-D
On Fri, Jan 29, 2010 at 11:19 AM, Karthik Ranganathan
wrote:
> Yup totally - either name or ip would work. Not sure if there is a pro or a
> con to choosing either one - but
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806455#action_12806455
]
Jean-Daniel Cryans commented on HBASE-2174:
---
Another thing people see sometimes i
i meant even if we were using hostnames for RS registration (which i
think has lot of advantages - not necessarily in our environment
though) - the master processing of the heartbeat (or whatever it's
processing) shouldn't require a forward lookup. if it needs the ip
address - it already has that v
Investigate .META. slowdowns when more than 1 store files
-
Key: HBASE-2175
URL: https://issues.apache.org/jira/browse/HBASE-2175
Project: Hadoop HBase
Issue Type: Bug
Repor
On Fri, Jan 29, 2010 at 11:29 AM, Joydeep Sarma wrote:
> i meant even if we were using hostnames for RS registration (which i
> think has lot of advantages - not necessarily in our environment
> though) -
Agreed, we should use hostnames for the advantages it gives listed
earlier in this thread...
On Fri, Jan 29, 2010 at 5:17 AM, Lars George wrote:
> �...@override
> public void sync() throws IOException {
> this.writer.sync();
> if (this.writer_out != null) {
> this.writer_out.sync();
> }
> }
>
> The first sync calls SequenceFile.Writer.sync() which is not at all
> what we w
HRegionInfo reported empty on regions in meta, leading to them being deleted,
although the regions contain data and exist
-
Key: HBASE-2176
URL
[
https://issues.apache.org/jira/browse/HBASE-2176?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Andrei Dragomir updated HBASE-2176:
---
Attachment: 799255.txt
Log forensics on our cluster.
> HRegionInfo reported empty on region
[
https://issues.apache.org/jira/browse/HBASE-2173?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-2173:
-
Status: Patch Available (was: Open)
Adds indexed to the javadoc target.
> New idx javadoc not included with the
[
https://issues.apache.org/jira/browse/HBASE-2173?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-2173:
-
Attachment: HBASE-2173.patch
> New idx javadoc not included with the rest
> -
[
https://issues.apache.org/jira/browse/HBASE-1679?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806543#action_12806543
]
Kannan Muthukkaruppan commented on HBASE-1679:
--
Yup, this is pretty much the i
[
https://issues.apache.org/jira/browse/HBASE-1679?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806544#action_12806544
]
Kannan Muthukkaruppan commented on HBASE-1679:
--
s/asked to restarted/asked to
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806561#action_12806561
]
Kannan Muthukkaruppan commented on HBASE-2174:
--
JD: The cluster coming down wh
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806562#action_12806562
]
Kannan Muthukkaruppan commented on HBASE-2174:
--
To fill in some more details,
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806570#action_12806570
]
ryan rawson commented on HBASE-2174:
im wondering if we should do something radical and
[
https://issues.apache.org/jira/browse/HBASE-1679?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
stack updated HBASE-1679:
-
Priority: Critical (was: Major)
Fix Version/s: 0.20.4
Upped priority and marking as a fix for 0.20.4 as
[
https://issues.apache.org/jira/browse/HBASE-1679?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806579#action_12806579
]
Kannan Muthukkaruppan commented on HBASE-1679:
--
Yes, having some of these key
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806580#action_12806580
]
Joydeep Sen Sarma commented on HBASE-2174:
--
yeah - hostnames are much more flexibl
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806608#action_12806608
]
Karthik Ranganathan commented on HBASE-2174:
@Joydeep/Ryan - good point about E
[
https://issues.apache.org/jira/browse/HBASE-2174?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806611#action_12806611
]
Kannan Muthukkaruppan commented on HBASE-2174:
--
Also, in '.META.', the region
[
https://issues.apache.org/jira/browse/HBASE-1246?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806619#action_12806619
]
Jean-Daniel Cryans commented on HBASE-1246:
---
We don't use Bloom Filters at the mo
[
https://issues.apache.org/jira/browse/HBASE-2023?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806620#action_12806620
]
Jean-Daniel Cryans commented on HBASE-2023:
---
Maybe a low hanging fruit would be t
[
https://issues.apache.org/jira/browse/HBASE-1246?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806627#action_12806627
]
stack commented on HBASE-1246:
--
We'll be reviving bloomfilters in the near future. Lets keep
[
https://issues.apache.org/jira/browse/HBASE-2023?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12806629#action_12806629
]
stack commented on HBASE-2023:
--
I wonder if it'd be possible to do a mock regionserver impleme
37 matches
Mail list logo
