Re: Riak performance problems when LevelDB database grows beyond 16GB

Jan.Evangelista at seznam.cz Jan.Evangelista at seznam.cz
Tue Oct 23 06:35:02 EDT 2012


> 1.  Did you realize that the "log_jan.txt" file from #1 below documented a 
hard disk failure?

I did not know about the corruption (I did not know that LevelDB logs are 
human readable application logs), thanks for telling me. I did look at syslog 
and I did not find any traces of a disk failure.

> You mentioned a failed drive once.  I am not sure if this is the same drive.

No, it is a different disk. I am not using the problematic node anymore.

> You also discuss a basho_bench failure.  Is this the same test run as the 
log_jan.txt file?  The hard drives had their first failure at:

Yes, it is. (It is an application-level test written as a basho_bench driver.)

> The next question from me is whether the drive / disk array problems are 
your only problem at this point.  
> The data in log_jan.txt looks ok until the failures start.  I am willing to 
work more, but I need to better understand your next level of problems.

I will look into it and report back.

Thanks!

Jan




More information about the riak-users mailing list