We are in the process of tuning our infrastructure for passing the 80
million file mark and uncovered some interesting facts.  I will admit up
front that we haven't been dedicated to running FSCK until recently and
the latest process has been running for about 10 days.  In addition to
that, we are experiencing a surge in system utilization.  Our current
policy is putting files on 2 devices. 

 

-          Over 7 million files are exhibiting a policy violation on the
current FSCK run (out of 71.8 million total)

-          Approximately 6.2 million of these files are perfectly fine,
but have leftover entries in file_to_replicate (also generating
policy_no_suggestions messages)

-          Approximately 450 thousand files are replicated on more than
2 devices (the policy limit)

-          Approximately 4.6 thousand files are replicated on 47 devices

 

Our infrastructure is composed of the following:

 

-          single master/slave database

-          3 trackers running off the database

-          7 mogstored nodes with 46 devices each and lighttpd set up as
the getport

 

Any thoughts or suggestions would be helpful.  Thanks in advance. 

 

Best,

Brian

 

Reply via email to