[ 
https://issues.apache.org/jira/browse/KUDU-3068?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

huangtianhua resolved KUDU-3068.
--------------------------------
    Fix Version/s: 1.13.0
       Resolution: Fixed

> MemTracker CHECK failed on aarch64 when run cache-bench test
> ------------------------------------------------------------
>
>                 Key: KUDU-3068
>                 URL: https://issues.apache.org/jira/browse/KUDU-3068
>             Project: Kudu
>          Issue Type: Sub-task
>            Reporter: huangtianhua
>            Assignee: huangtianhua
>            Priority: Major
>             Fix For: 1.13.0
>
>
> I test kudu on aarch64 server based on https://gerrit.cloudera.org/#/c/14964/ 
> , the test cache-bench failed, error info as below:
> root@ubuntu:/home/workspace/kudu/build/debug/bin# ./cache-bench
> [==========] Running 4 tests from 1 test case.
> [----------] Global test environment set-up.
> [----------] 4 tests from Patterns/CacheBench
> [ RUN      ] Patterns/CacheBench.RunBench/0
> WARNING: Logging before InitGoogleLogging() is written to STDERR
> I0304 09:04:19.007313 19049 cache-bench.cc:172] Warming up...
> I0304 09:04:20.008733 19049 cache-bench.cc:175] Running benchmark...
> I0304 09:04:22.009366 19049 cache-bench.cc:183] ZIPFIAN ratio=1.00x 
> n_unique=262144: 3.28M lookups/sec
> I0304 09:04:22.009399 19049 cache-bench.cc:184] ZIPFIAN ratio=1.00x 
> n_unique=262144: 99.0% hit rate
> F0304 09:04:22.048135 19049 mem_tracker.cc:89] Check failed: consumption() == 
> 0 Memory tracker test-cache-sharded_lru_cache->root has unreleased 
> consumption 463171584
> *** Check failure stack trace: ***
> *** Aborted at 1583312662 (unix time) try "date -d @1583312662" if you are 
> using GNU date ***
> PC: @                0x0 (unknown)
> *** SIGABRT (@0x4a69) received by PID 19049 (TID 0xffffad19e010) from PID 
> 19049; stack trace: ***
>    @     0xffffae3f8688 ([vdso]+0x687)
>    @     0xffffad7384d8 raise
>    @     0xffffad738464 raise
> Aborted (core dumped)
> I run this test for many times and it's successful only one time, I have no 
> idea why the consumption is not zero? 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to