FAQ
Say I have a server with 12 disks each (3TB each) presented as a raw volume
and each disk corresponds to a mount point and each mount point is part of
data directories. Does a disk failure of one of the 12 disks, take the node
out of operation until the disk is replaced or the reads/writes continue
with remaining disks?

To unsubscribe from this group and stop receiving emails from it, send an email to scm-users+unsubscribe@cloudera.org.

Search Discussions

  • Kumar y at Oct 4, 2013 at 1:44 am
    Any disk failure might bring down datanode if DN cannot continue to write
    to the configured data volume. you can this behavior with this property
    "dfs.datanode.failed.volumes.tolerated". if you give value 2 for this
    property then DN will tolerate 2 vol/disk failures and continue to operate
    and it will shutdown itself when there are more than 2 failures.

    On Thu, Oct 3, 2013 at 6:21 PM, wrote:

    Say I have a server with 12 disks each (3TB each) presented as a raw
    volume and each disk corresponds to a mount point and each mount point is
    part of data directories. Does a disk failure of one of the 12 disks, take
    the node out of operation until the disk is replaced or the reads/writes
    continue with remaining disks?

    To unsubscribe from this group and stop receiving emails from it, send an
    email to scm-users+unsubscribe@cloudera.org.
    To unsubscribe from this group and stop receiving emails from it, send an email to scm-users+unsubscribe@cloudera.org.

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupscm-users @
categorieshadoop
postedOct 4, '13 at 1:21a
activeOct 4, '13 at 1:44a
posts2
users2
websitecloudera.com
irc#hadoop

2 users in discussion

Kumar y: 1 post Skr989: 1 post

People

Translate

site design / logo © 2022 Grokbase