Support Questions

Find answers, ask questions, and share your expertise

Atlas Dashboard UI not working



I'm working with HDP 2.6 and i have a problem logging with Atlas dashboard UI.

It's inaccesible.

Any suggession. Thank you in advance



@nisrine elloumi

Can you share these 2 Atlas logs in /var/log/atlas

  • application.log
  • gc-worker.log.0.current

Did it work before is your cluster kerberized ? What OS?



this application.log

2018-10-17 17:39:38,095 INFO - [pool-2-thread-29 - 47593d06-ea3b-4f9d-ae54-1f52ed931dfc:] ~ Audit: UNKNOWN/.......... performed request GET http://..........:21000/api/atlas/admin/status (....) at time 2018-10-17T15:39Z (AUDIT:117)

this for gc-worker.log.0.current

{Heap before GC invocations=87 (full 0): par new generation total 552960K, used 501664K [0x0000000080000000, 0x00000000a5800000, 0x00000000a5800000) eden space 491520K, 100% used [0x0000000080000000, 0x000000009e000000, 0x000000009e000000) from space 61440K, 16% used [0x00000000a1c00000, 0x00000000a25e8390, 0x00000000a5800000) to space 61440K, 0% used [0x000000009e000000, 0x000000009e000000, 0x00000000a1c00000) concurrent mark-sweep generation total 1482752K, used 192764K [0x00000000a5800000, 0x0000000100000000, 0x0000000100000000) Metaspace used 100066K, capacity 101828K, committed 102016K, reserved 1138688K class space used 12079K, capacity 12406K, committed 12416K, reserved 1048576K 97088.938: [GC (Allocation Failure) 97088.938: [ParNew Desired survivor size 31457280 bytes, new threshold 6 (max 6) - age 1: 308504 bytes, 308504 total - age 2: 9640 bytes, 318144 total - age 3: 33576 bytes, 351720 total - age 4: 2168 bytes, 353888 total : 501664K->2871K(552960K), 0.0178481 secs] 694429K->195636K(2035712K), 0.0179751 secs] [Times: user=0.32 sys=0.19, real=0.02 secs]

yes it is a kerberized cluster.

thank you for your response.


@nisrine elloumi

I don't see anything significant in the logs you attached. Can you enable DEBUG.

From Ambari UI -->Atlas--Config-->Advanced atlas-log4j

Uncomment the below line and restart Atlas

 <!-- uncomment this block to generate performance traces 
<appender name="perf_appender"> 
<param name="File" value="{{log_dir}}/atlas_perf.log" /> 
<param name="datePattern" value="'.'yyyy-MM-dd" /> 
<param name="append" value="true" /> 
<layout class="org.apache.log4j.PatternLayout"> 
<param name="ConversionPattern" value="%d|%t|%m%n" /> 
<logger name="org.apache.atlas.perf" additivity="false"> 
<level value="debug" /> 
<appender-ref ref="perf_appender" /> 

This will generate logs with $atlas_perf.log upload those files

Connect to zookeeper and check if the hbase table are available

# /usr/hdp/current/zookeeper-client/bin/ 

Then on the zk CLI list the hbase tables

[zk: localhost:2181(CONNECTED) 0] ls /hbase-secure/table 
[ATLAS_ENTITY_AUDIT_EVENTS, hbase:meta, atlas_titan, hbase:namespace, hbase:acl]

Share your output