Support Forums : DataStax

DataStax Support Forums » DataStax Enterprise

DSE 2.0, Solr and non-Solr related content, UnavailableException

(6 posts) (3 voices)

Started 1 year ago by
Latest reply from baskaran.md
This topic is resolved

Tags:

jas
Member

Hello:

I'm working on restructuring an application that is currently making use of Solr 3.5.0 and Cassandra 1.0.7 to instead use DSE 2.0. I'm at the very early stages, so I'm just starting a single node after setting up on my computer (Max OS X Lion) using dse-2.0-1-bin.tar.gz. After launching with 'dse cassandra -s', I was able to index the demo wikipedia content and play with demo a bit.

Not all of my application content is Solr based, though, so I also defined additional keyspaces and column families using cassandra-cli. I'm able to define these things, but then things go south after that. For example:

[default@unknown] create keyspace IsecMetrics;
13c37090-7c32-11e1-0000-9e64b5f2caf7
Waiting for schema agreement...
... schemas agree across the cluster
[default@unknown] use IsecMetrics;
Authenticated to keyspace: IsecMetrics
[default@IsecMetrics] CREATE COLUMN FAMILY TokenMetrics
... WITH comparator=TimeUUIDType
... AND key_validation_class=UTF8Type
... AND default_validation_class=UTF8Type;
1899fee0-7c32-11e1-0000-9e64b5f2caf7
Waiting for schema agreement...
... schemas agree across the cluster
[default@IsecMetrics] list TokenMetrics;
Using default limit of 100
null
UnavailableException()
at org.apache.cassandra.thrift.Cassandra$get_range_slices_result.read(Cassandra.java:12822)
at org.apache.cassandra.thrift.Cassandra$Client.recv_get_range_slices(Cassandra.java:762)
at org.apache.cassandra.thrift.Cassandra$Client.get_range_slices(Cassandra.java:734)
at org.apache.cassandra.cli.CliClient.executeList(CliClient.java:1382)
at org.apache.cassandra.cli.CliClient.executeCLIStatement(CliClient.java:267)
at org.apache.cassandra.cli.CliMain.processStatementInteractive(CliMain.java:220)
at org.apache.cassandra.cli.CliMain.main(CliMain.java:348)
[default@IsecMetrics]

I've seen some information regarding UnavailableException indicates there are not enough replicas available to satisfy the requested consistency level. I'm not sure what CL the CLI uses for the range_slice, but I have just the one node.

[default@IsecMetrics] describe IsecMetrics;
Keyspace: IsecMetrics:
Replication Strategy: org.apache.cassandra.locator.NetworkTopologyStrategy
Durable Writes: true
Options: [datacenter1:1]
Column Families:
ColumnFamily: TokenMetrics
Key Validation Class: org.apache.cassandra.db.marshal.UTF8Type
Default column value validator: org.apache.cassandra.db.marshal.UTF8Type
Columns sorted by: org.apache.cassandra.db.marshal.TimeUUIDType
Row cache size / save period in seconds / keys to save : 0.0/0/all
Row Cache Provider: org.apache.cassandra.cache.ConcurrentLinkedHashCacheProvider
Key cache size / save period in seconds: 200000.0/14400
GC grace seconds: 864000
Compaction min/max thresholds: 4/32
Read repair chance: 1.0
Replicate on write: true
Bloom Filter FP chance: default
Built indexes: []
Compaction Strategy: org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy
[default@IsecMetrics]
ColumnFamily: TokenMetrics
Key Validation Class: org.apache.cassandra.db.marshal.UTF8Type
Default column value validator: org.apache.cassandra.db.marshal.UTF8Type
Columns sorted by: org.apache.cassandra.db.marshal.TimeUUIDType
Row cache size / save period in seconds / keys to save : 0.0/0/all
Row Cache Provider: org.apache.cassandra.cache.ConcurrentLinkedHashCacheProvider
Key cache size / save period in seconds: 200000.0/14400
GC grace seconds: 864000
Compaction min/max thresholds: 4/32
Read repair chance: 1.0
Replicate on write: true
Bloom Filter FP chance: default
Built indexes: []
Compaction Strategy: org.apache.cassandra.db.compaction.SizeTieredCompactionStrategy
[default@IsecMetrics]

I'm not sure if NetworkTopologyStrategy is the right thing for this situation. The docs (http://www.datastax.com/docs/1.0/configuration/storage_configuration) indicate a properly configured snitch is required. However, looking at the wiki keyspace that was created for the demo:

[default@IsecMetrics] describe wiki;
Keyspace: wiki:
Replication Strategy: org.apache.cassandra.locator.NetworkTopologyStrategy
Durable Writes: true
Options: [Solr:1]
Column Families:
ColumnFamily: solr
Key Validation Class: org.apache.cassandra.db.marshal.UTF8Type
Default column value validator: org.apache.cassandra.db.marshal.UTF8Type
Columns sorted by: org.apache.cassandra.db.marshal.UTF8Type
Row cache size / save period in seconds / keys to save : 0.0/0/all
Row Cache Provider: org.apache.cassandra.cache.ConcurrentLinkedHashCacheProvider
Key cache size / save period in seconds: 200000.0/14400
GC grace seconds: 864000
Compaction min/max thresholds: 4/32
Read repair chance: 1.0
Replicate on write: true
Bloom Filter FP chance: default
Built indexes: [wiki.solr, config-only, config-only, wiki.solr, wiki.solr, wiki.solr, wiki.solr, wiki.solr]
Column Metadata:
Column Name: _docBoost
Validation Class: org.apache.cassandra.db.marshal.UTF8Type
Index Name: wiki_solr__docBoost_index
Index Type: CUSTOM
Index Options: {class_name=com.datastax.bdp.cassandra.index.solr.SolrSecondaryIndex}
Column Name: _solr_schema.xml
Validation Class: org.apache.cassandra.db.marshal.UTF8Type
Index Name: solr__solr_schemaxml_idx
Index Type: CUSTOM
Index Options: {class_name=com.datastax.bdp.util.ConfigOnlyIndex, schema.xml=
...

That gets rather long when the Solr config files are listed etc. This has the same replication strategy, so am I right in assuming that DSE 2.0 has a properly configured snitch?

I'm not sure what the Options refer to. The wiki keyspace has [Solr:1], and my keyspace has [datacenter1:1]. I'm not sure what the real meaning is there. Safe to say Solr indicates something search related, whereas mine has a default name? Does the :1 mean anything wrt to the replication factor?

I don't have any issues with the wiki keyspace:

[default@IsecMetrics] use wiki;
Authenticated to keyspace: wiki
[default@wiki] list solr limit 1;
-------------------
RowKey: 23739865
=> (column=_docBoost, value=1.0, timestamp=1333291396917000)
=> (column=body, value= {{Notability|date=July 2009}} {{Unreferenced|date=July 2009}} {{Infobox television | bgcolor = #c00 | fgcolor = #fff | show_name = Arangkada | image = | caption = | format = [[News Talk Show|News talk show]] | runtime = 1 hour | creator = [[ABS-CBN Regional Network Group]] | country = | language = | network = [[DXEC-TV|ABS-CBN TV-2 Cagayan de Oro]] | executive producer = Conchita May Tanate | starring = PJ dela Pe?a | first_aired = July 25, 2009 | last_aired = Present | num_episodes = n/a (airs daily) }} '''Arangkada''' is the regional morning news talk show of ABS-CBN Northern Mindanao that dishes what's happening in the [[Cagayan de Oro City]] and within the whole [[Mindanao]]. It discuss the news reports and problems in our country. ==About== Arangkada was started in 2009. It airs 7:00am on ABS-CBN Channel 2 and Channel 23 in [[Cagayan de Oro City]]. Arangkada also reaches the whole [[Northern Mindanao]], [[Caraga]] region and [[Zamboanga del Norte]]. It can also be seen via online at [http://www.abs-cbnnow.com http://www.abs-cbnNOW.com. ==See also== * [[DXEC-TV]] * [[ABS-CBN Regional Network Group]] ==External links== * [http://now.abs-cbn.com/shows-tvpreg.aspx?showid=487] [[Category:Philippine television series]] [[Category:ABS?CBN Corporation]] [[Category:DXEC]] [[Category:Cagayan de Oro]], timestamp=1333291396917000)
=> (column=date, value=25-AUG-2011 22:28:21.000, timestamp=1333291396917000)
=> (column=title, value=Arankada, timestamp=1333291396917000)

1 Row Returned.
Elapsed time: 3 msec(s).
[default@wiki]

Any ideas on how I can get past this? I don't see anything relevant in the Cassandra system.log.

Thanks,

Jeff

Posted 1 year ago #
tjake
Moderator

Since you are running DSE with the DseSimpleSnitch the Datacenter of the node you are on is "Solr"

In your new keyspace you set replication as {datacenter1:1}

It's complaining because it doesn't know where to find datacenter1. If you switch this to "Solr:1" then it will start working.

Posted 1 year ago #
jas
Member

Hi Jake:

Thanks for the pointer! I added to each of my keyspaces:

WITH placement_strategy = 'NetworkTopologyStrategy'
AND strategy_options=[{Solr:1}];

and things are working for me. Thank you.

One sort of related question: Given the keyspace.columnfamily notation when creating/updating Solr cores using the HTTP API, is there any crime in putting non-Solr related column families in that same keyspace? Or is it best to just use another keyspace? I ask because I have content that is related, but only part of it is indexed, and it'd handy for it all to live in one keyspace. This is just the initial move for me, as I will continue using SolrJ to talk to Solr and Hector for Cassandra and continue to treat them as separate entities for the short term.

Thanks again,

Jeff

Posted 1 year ago #
tjake
Moderator

Hi Jeff, There is no issue with putting non solr column families in the same keyspaces

Posted 1 year ago #
jas
Member

Great, thanks Jake.

Posted 1 year ago #
baskaran.md
Member

Thanks tjake.. That really saves a lot of time..

For me, the following works like a charm..

update KEYSPACE my_keyspace WITH placement_strategy = 'NetworkTopologyStrategy' AND strategy_options={ap-southeast:6};

Posted 7 months ago #

RSS feed for this topic

Reply

You must log in to post.