accumulo-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bill Havanki (JIRA)" <>
Subject [jira] [Commented] (ACCUMULO-2657) Shard randomwalk fails for 1.6.0
Date Mon, 14 Apr 2014 15:42:15 GMT


Bill Havanki commented on ACCUMULO-2657:

The tablet server, for example, does in fact log a value for the property as it starts up,
which proves that it gets the property somehow. The table properties that I am able to dump
out also prove that a value shows up there somehow.

* The tablet server logs the value from the {{Accumulo.init()}} method, by looking at the
configuration retrieved from {{ServerConfiguration.getConfiguration()}}. That configuration
is a chained list of configurations, a {{ZooConfiguration}} defaulting to the site configuration,
which in turn defaults to a {{DefaultConfiguration}}.
* The {{TableConfiguration}} that is retrieved for a table is parented by a {{NamespaceConfiguration}},
which is in turn parented by the site configuration, which is parented by the default configuration.

So, in both scenarios, from the tablet server and from a table, the site configuration ultimately
decides what the property value is (when it's not set explicitly anywhere). The important
thing to know is that my accumulo-site.xml does not explicitly set the property. Therefore,
it delegates to the default configuration to figure out the value.

The default value is retrieved using {{DefaultConfiguration.get(Property)}}, which calls {{Property.getDefaultValue()}}.
It's that method that has the problematic logic. Because the property is annotated as interpolated,
and because it's an ABSOLUTEPATH type, the value is rendered as Java's user directory. It's
the fact that this can vary which, I believe, causes me trouble.

> Shard randomwalk fails for 1.6.0
> --------------------------------
>                 Key: ACCUMULO-2657
>                 URL:
>             Project: Accumulo
>          Issue Type: Test
>          Components: test
>         Environment: Seven-node CDH 4.5 cluster, 2 masters, 5 tservers, 3 ZK
>            Reporter: Bill Havanki
>              Labels: 16_qa_bug
> This has happened twice for me, while trying to run randomwalk for 1.6.0 release testing.
> {noformat}
> 10 09:40:04,358 [randomwalk.Framework] ERROR: Error during random walk
> java.lang.Exception: Error running node shard.ExportIndex
>         at org.apache.accumulo.test.randomwalk.Module.visit(
>         at
>         at org.apache.accumulo.test.randomwalk.Framework.main(
>         at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
>         at sun.reflect.NativeMethodAccessorImpl.invoke(
>         at sun.reflect.DelegatingMethodAccessorImpl.invoke(
>         at java.lang.reflect.Method.invoke(
>         at org.apache.accumulo.start.Main$
>         at
> Caused by: java.lang.Exception: Props not equals ST_index_test_server_com_3525_1397137262293
>         at org.apache.accumulo.test.randomwalk.shard.ExportIndex.visit(
>         at org.apache.accumulo.test.randomwalk.Module.visit(
>         ... 8 more
> {noformat}

This message was sent by Atlassian JIRA

View raw message