You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Radim Kolar <hs...@sendmail.cz> on 2011/09/13 16:22:45 UTC
adding node without bootstrap
If i add node without bootstrap and run repair on both nodes then no
data are moved to new node, just new writes are going to new node. This
is expected behaviour or it is some kind of user or program error?
Re: adding node without bootstrap
Posted by aaron morton <aa...@thelastpickle.com>.
Then there is nothing to repair.
Set a better token, cassandra-cli to increase the RF to 2 and then kick off repair.
A
-----------------
Aaron Morton
Freelance Cassandra Developer
@aaronmorton
http://www.thelastpickle.com
On 26/09/2011, at 10:12 AM, Radim Kolar wrote:
> Dne 25.9.2011 22:40, aaron morton napsal(a):
>> That message will be logged if there RF on the keyspace is 1 or if the other nodes are not up.
>> What's the RF ?
> rf is 1.
Re: adding node without bootstrap
Posted by Radim Kolar <hs...@sendmail.cz>.
Dne 25.9.2011 22:40, aaron morton napsal(a):
> That message will be logged if there RF on the keyspace is 1 or if the other nodes are not up.
> What's the RF ?
rf is 1.
Re: adding node without bootstrap
Posted by aaron morton <aa...@thelastpickle.com>.
That message will be logged if there RF on the keyspace is 1 or if the other nodes are not up.
What's the RF ?
You should also sort out the tokens before going to far.
Cheers
-----------------
Aaron Morton
Freelance Cassandra Developer
@aaronmorton
http://www.thelastpickle.com
On 25/09/2011, at 7:35 PM, Radim Kolar wrote:
>
>> If you join a node with auto_bootstrap=false you had better be working at quorum or higher to avoid stale/not found reads. You should then repair the node right away to get all the missing data back on the node. This is not suggested. It is best to leave auto_boostrap=true and let Cassandra handle this on the front end.
> This do not works. I joined ring with node without bootstrap and result is like this:
>
> 216.17.99.40 datacenter1 rack1 Up Normal 1.17 GB 99.64% 83030609119105147711596238577753588267
> 64.6.104.18 datacenter1 rack1 Up Normal 43.15 KB 0.36% 83648735508289295779178617154261005054
>
> Well, this was expected. But running repair on both nodes didnt do anything:
>
> INFO [GossipStage:1] 2011-09-25 08:18:34,287 Gossiper.java (line 715) Node /216.17.99.40 is now part of the cluster
> INFO [GossipStage:1] 2011-09-25 08:18:34,287 Gossiper.java (line 681) InetAddress /216.17.99.40 is now UP
> INFO [AntiEntropySessions:1] 2011-09-25 08:22:16,066 AntiEntropyService.java (line 648) No neighbors to repair with for test on (83030609119105147711596238577753588267,83648735508289295779178617154261005054]: manual-repair-04dd27f0-401b-4452-b0eb-853beeda197b completed.
>
> Data are not moved to new node. Maybe tokens are not too random. I deleted new node and retried:
>
> 64.6.104.18 datacenter1 rack1 Up Normal 45.52 KB 56.94% 9762979552315026283322466206354139578
> 216.17.99.40 datacenter1 rack1 Up Normal 1.17 GB 43.06% 83030609119105147711596238577753588267
>
> and still nothing, while running repair on both nodes.
>
> INFO [AntiEntropySessions:1] 2011-09-25 08:29:13,447 AntiEntropyService.java (line 648) No neighbors to repair with for test on (83030609119105147711596238577753588267,9762979552315026283322466206354139578]: manual-repair-87bfcc67-2b99-4285-8571-e5bd168ef5e0 completed.
>
> Can you try this too? i cant get scenario: make 1 node - add data, add second node without bootstrap then repair on both work.
Re: adding node without bootstrap
Posted by Radim Kolar <hs...@sendmail.cz>.
> If you join a node with auto_bootstrap=false you had better be working
> at quorum or higher to avoid stale/not found reads. You should then
> repair the node right away to get all the missing data back on the
> node. This is not suggested. It is best to leave auto_boostrap=true
> and let Cassandra handle this on the front end.
This do not works. I joined ring with node without bootstrap and result
is like this:
216.17.99.40 datacenter1 rack1 Up Normal 1.17 GB
99.64% 83030609119105147711596238577753588267
64.6.104.18 datacenter1 rack1 Up Normal 43.15 KB
0.36% 83648735508289295779178617154261005054
Well, this was expected. But running repair on both nodes didnt do anything:
INFO [GossipStage:1] 2011-09-25 08:18:34,287 Gossiper.java (line 715)
Node /216.17.99.40 is now part of the cluster
INFO [GossipStage:1] 2011-09-25 08:18:34,287 Gossiper.java (line 681)
InetAddress /216.17.99.40 is now UP
INFO [AntiEntropySessions:1] 2011-09-25 08:22:16,066
AntiEntropyService.java (line 648) No neighbors to repair with for test
on
(83030609119105147711596238577753588267,83648735508289295779178617154261005054]:
manual-repair-04dd27f0-401b-4452-b0eb-853beeda197b completed.
Data are not moved to new node. Maybe tokens are not too random. I
deleted new node and retried:
64.6.104.18 datacenter1 rack1 Up Normal 45.52 KB
56.94% 9762979552315026283322466206354139578
216.17.99.40 datacenter1 rack1 Up Normal 1.17 GB
43.06% 83030609119105147711596238577753588267
and still nothing, while running repair on both nodes.
INFO [AntiEntropySessions:1] 2011-09-25 08:29:13,447
AntiEntropyService.java (line 648) No neighbors to repair with for test
on
(83030609119105147711596238577753588267,9762979552315026283322466206354139578]:
manual-repair-87bfcc67-2b99-4285-8571-e5bd168ef5e0 completed.
Can you try this too? i cant get scenario: make 1 node - add data, add
second node without bootstrap then repair on both work.
Re: adding node without bootstrap
Posted by Edward Capriolo <ed...@gmail.com>.
2011/9/13 Radim Kolar <hs...@sendmail.cz>
> If i add node without bootstrap and run repair on both nodes then no data
> are moved to new node, just new writes are going to new node. This is
> expected behaviour or it is some kind of user or program error?
>
No. This is exactly what you are asking the node to do by setting false. The
goal of auto_bootstrap=true is to migrate data to the node automatically.
You can turn this method off if you have moved data their yourself out of
band (maybe rsync).
If you join a node with auto_bootstrap=false you had better be working at
quorum or higher to avoid stale/not found reads. You should then repair the
node right away to get all the missing data back on the node. This is not
suggested. It is best to leave auto_boostrap=true and let Cassandra handle
this on the front end.
Edward