Hi Jerome,

You should use Guava 27.0-jre with Accumulo 1.10.0 and Hadoop 3.2.1.

1.9.3 does not support Hadoop 3.2.1, precisely for reasons such as the
Guava issue you encountered.
I fixed the issue you encountered in 1.10.0 in
https://github.com/apache/accumulo/pull/1688
See the release notes for more information about 1.10.0:
https://accumulo.apache.org/release/accumulo-1.10.0/

On Fri, Oct 16, 2020 at 10:35 AM Jansen, Jerome R. (ITADD) (CON)
<[email protected]> wrote:
>
> Now that I am registered with the new address, hopefully this message posts.
>
> Jerome Jansen
>
> Prometheus Cloud
> Federal Bureau of Investigation
> ITADD\EDAS\DAFU
>   703-633-5733
>
> ________________________________
> From: Jansen, Jerome R. (ITADD) (CON)
> Sent: Friday, October 16, 2020 7:12 AM
> To: [email protected] <[email protected]>
> Cc: Jansen, Jerome [USA] <[email protected]>
> Subject: Accumulo gc is in constant waiting state due to error in the 
> accumulo gc-log:
>
> Accumulo gc is in constant waiting state due to error  in the accumulo gc-log:
>
>
> I am running accumulo 1.9.3 on hdfs 3.2.1 using guava-27.0-jre.jar (prepacked 
> in the hadoop-3.2.1 tar ball)
>
> At first I downgraded guava to version 16 but then the accumulo-gc failed as 
> it was unable to get the hdfs HA information.
>
> Next, after mor googling, I downgraded the guava jar to guava-20.jar - the gc 
> does not fail but still received the same message in the error log.
>
> Exception in thread "gc" java.lang.IllegalAccessError: tried to access method 
> com.google.common.base.Stopwatch.<init>()V from class 
> org.apache.accumulo.gc.replication.CloseWriteAheadLogReferences
>         at 
> org.apache.accumulo.gc.replication.CloseWriteAheadLogReferences.run(CloseWriteAheadLogReferences.java:89)
>         at 
> org.apache.accumulo.gc.SimpleGarbageCollector.run(SimpleGarbageCollector.java:602)
>         at 
> org.apache.accumulo.gc.SimpleGarbageCollector.main(SimpleGarbageCollector.java:166)
>         at org.apache.accumulo.gc.GCExecutable.execute(GCExecutable.java:34)
>         at org.apache.accumulo.start.Main$1.run(Main.java:130)
>         at java.lang.Thread.run(Thread.java:748)
>
> Does anyone know a work around ? Which guava jar version to use for hadoop 
> 3.2.1 and accumulo 1.9.3 ?
>
> Respectfully,
>
> Jerome Jansen
>
> Prometheus Cloud
> Federal Bureau of Investigation
> ITADD\EDAS\DAFU
>   703-633-5733

Reply via email to