[
https://issues.apache.org/jira/browse/IMPALA-7220?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17692987#comment-17692987
]
Quanlong Huang commented on IMPALA-7220:
----------------------------------------
FWIW, this also fails in scale factor 1:
{code:java}
[scheduler]: Query primitive_shuffle_1mb_rows Failed: ImpalaBeeswaxException:
Query aborted:Could not create files in any configured scratch directories
(--scratch_dirs=/tmp/impala-scratch) on backend 'ip-172-31-1-9:27002'. 84.29 GB
of scratch is currently in use by this Impala Daemon (84.29 GB by this query).
See logs for previous errors that may have prevented creating or writing
scratch files. The following directories were at capacity:
Disk I/O error on ip-172-31-1-9:27002: fwrite() failed for
/tmp/impala-scratch/cf4991bf9159c324:c76425dd00000000_0e926416-4654-4ca0-baf7-15e7cf0154cd.
No space left on device. errno=28, range_length=2097152{code}
> Queries processing very large strings hit "memory limit exceeded" instead of
> spilling
> -------------------------------------------------------------------------------------
>
> Key: IMPALA-7220
> URL: https://issues.apache.org/jira/browse/IMPALA-7220
> Project: IMPALA
> Issue Type: Improvement
> Components: Backend
> Environment: m4.4xl, ubuntu 16.04
> Reporter: Jim Apple
> Priority: Major
> Labels: resource-management
>
> I saw {{primitive_shuffle_1mb_rows}} failing at scale factor 4 with 3
> impalads running:
> {noformat}
> [scheduler] [tpch4_parquet Thread 0]: Running Query:
> primitive_shuffle_1mb_rows
> [query_exec_functions] [tpch4_parquet Thread 0]: Connected to localhost:21001
> [query_exec_functions] [tpch4_parquet Thread 0]: Connected to localhost:21001
> [query_exec_functions] [tpch4_parquet Thread 0]: ImpalaBeeswaxException:
> Query aborted:Memory limit exceeded: Error occurred on backend
> ip-172-31-25-187:22001 by fragment cd4f016604f24316:c310bea60000000f
> Memory left in process limit: -410.47 MB
> {noformat}
> Command run was
> {noformat}
> ./bin/single_node_perf_run.py --iterations 2 --scale 4 --table_formats
> parquet/none --workload targeted-perf --num_impalads 3 --query_names '.*'
> --load --start_minicluster $HASH1 $HASH2
> {noformat}
> https://jenkins.impala.io/view/Experimental/job/perf-AB-test/182/consoleText
> cc: [~janulatha]
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
