It is not expected that the users clear the distributed cache files. The
files are left as they are on the TT so that they can be used by other
jobs that refer to the same cache files on the same TT. They will be
deleted by the TT only when it is shutting down or when it tries to
purge some unused files when the total disk usage of all dist-cache
files hits a limit.
HTH,
+Vinod
Amogh Vasekar wrote:
My bad, its not exposed.
If nothing else's available, may be mimic its behaviour in the code?
Thanks,
Amogh
-----Original Message-----
From: #YONG YONG CHENG# [mailto:[email protected]]
Sent: Monday, September 07, 2009 10:27 AM
To: [email protected]
Subject: RE: DistributedCache purgeCache()
Thanks for your swift response.
But where can I find deletecache()?
Thanks.
-----Original Message-----
From: Amogh Vasekar [mailto:[email protected]]
Sent: Thu 9/3/2009 2:44 PM
To: [email protected]
Subject: RE: DistributedCache purgeCache()
AFAIK, releaseCache only works on cleaning reference to your file. Try using deletecache in synchronized manner.
Thanks,
Amogh
-----Original Message-----
From: #YONG YONG CHENG# [mailto:[email protected]]
Sent: Thursday, September 03, 2009 8:50 AM
To: [email protected]
Subject: DistributedCache purgeCache()
Good Day,
I have a question on the DistributedCache as follows.
I have used DistributedCache to move my executable(.exe) around the (onto the
local filesystems of) nodes in Hadoop and run the .exe (via addCacheArchive()
and getLocalCacheArchives()). But I discovered after my job, the .exe still
resides on the nodes. I have try using releaseCache() and/or purgeCache(), but
the .exe still on the nodes. Is there a way to remove the .exe from the nodes?
Thanks.
