Reply
New Contributor
Posts: 1
Registered: ‎06-11-2018

Unable to start cloudera management service

After doing a `sudo yum update` and a restart to cloudera quickstart vm, the host monitor and service monitor are not restarting. here is the error log. can anyone please tell me what exactly is wrong and how to fix it.

Exception in thread "main" java.lang.RuntimeException: org.fusesource.leveldbjni.internal.NativeDB$DBException: Invalid argument: not an sstable (bad magic number)
	at org.fusesource.leveldbjni.internal.JniDBIterator.seek(JniDBIterator.java:68)
	at com.cloudera.cmon.tstore.leveldb.LDBUtils.getSchemaVersion(LDBUtils.java:305)
	at com.cloudera.cmon.tstore.leveldb.LDBUtils.openVersionedDB(LDBUtils.java:266)
	at com.cloudera.cmon.tstore.leveldb.LDBUtils.openVersionedDB(LDBUtils.java:212)
	at com.cloudera.cmon.tstore.leveldb.LDBTimeSeriesMetadataStore.openMetadataDB(LDBTimeSeriesMetadataStore.java:148)
	at com.cloudera.cmon.tstore.leveldb.LDBTimeSeriesMetadataStore.<init>(LDBTimeSeriesMetadataStore.java:138)
	at com.cloudera.cmon.firehose.Main.main(Main.java:474)
Caused by: org.fusesource.leveldbjni.internal.NativeDB$DBException: Invalid argument: not an sstable (bad magic number)
	at org.fusesource.leveldbjni.internal.NativeDB.checkStatus(NativeDB.java:194)
	at org.fusesource.leveldbjni.internal.NativeIterator.checkStatus(NativeIterator.java:121)
	at org.fusesource.leveldbjni.internal.NativeIterator.seek(NativeIterator.java:151)
	at org.fusesource.leveldbjni.internal.NativeIterator.seek(NativeIterator.java:145)
	at org.fusesource.leveldbjni.internal.NativeIterator.seek(NativeIterator.java:138)
	at org.fusesource.leveldbjni.internal.JniDBIterator.seek(JniDBIterator.java:63)
	... 6 more

 

Contributor
Posts: 35
Registered: ‎05-21-2018

Re: Unable to start cloudera management service

hello @shibu

 

Could you kindly share the detailed log file of the service.

Posts: 911
Topics: 1
Kudos: 208
Solutions: 111
Registered: ‎04-22-2014

Re: Unable to start cloudera management service

@shibu,

 

We can tell from the stack trace that either Host Monitor, Service Monitor, or Event Server couldn't start due to a problem in thieir leveldb files.

 

It is unlikely that you can recover the leveldb files and you may need to purge current leveldb files (on disk).

 

Before doing that, let us know:

 

- What role in the Cloudera Management Service is not starting? (try starting them one by one to see which fail)

- Let us know if it is OK to remove time series data recorded for that role.

- How much space do you have where /var is mounted?  (df -k)

- Which version of Cloudera Manager are you using?

 

-Ben

Announcements