ceph · dmick · Oct 30, 2013 · Oct 30, 2013 · Oct 31, 2013 · Sep 21, 2013
diff --git a/ceph-object-corpus b/ceph-object-corpus
diff --git a/configure.ac b/configure.ac
@@ -120,8 +120,41 @@ AX_C_PRETTY_FUNC
 # Checks for libraries.
 ACX_PTHREAD
 AC_CHECK_LIB([uuid], [uuid_parse], [true], AC_MSG_FAILURE([libuuid not found]))
+
+#
+# Check for res_nquery in libresolv. There are several variations. On OSX
+# res_nquery is a macro defined in resolv.h, so the typical AC_CHECK_LIB
+# doesn't work. On FreeBSD res_nquery can be found in libc. The required
+# library for linking (if any) is defined RESOLV_LIBS.
+#
+AC_CHECK_HEADER([resolv.h], [], [], [#include <netinet/in.h>])
+
+AC_DEFUN([CHECK_RESOLV_LIBS], [{
+  AC_MSG_CHECKING([if res_nquery will link (LIBS=$1)])
+  saved_LIBS="${LIBS}"
+  LIBS="$1"
+  AC_LINK_IFELSE([AC_LANG_PROGRAM([[
+        #include <netinet/in.h>
+        #include <resolv.h>
+      ]], [[res_nquery(0, 0, 0, 0, 0, 0);]])],
+      [AC_MSG_RESULT([yes])
+      [$2]],
+      AC_MSG_RESULT([no]))
+  LIBS="${saved_LIBS}"
+}])
+
+RESOLV_LIBS=""
+CHECK_RESOLV_LIBS([$RESOLV_LIBS], [resolv_libs="ok"])
+if test x"$resolv_libs" != "xok"; then
+  RESOLV_LIBS="-lresolv"
+  CHECK_RESOLV_LIBS([$RESOLV_LIBS], [resolv_libs="ok"])
+  if test x"$resolv_libs" != "xok"; then
+    AC_MSG_FAILURE([no resolv library found])
+  fi
+fi
+AC_SUBST([RESOLV_LIBS])
+
 AC_CHECK_LIB([m], [pow], [true], AC_MSG_FAILURE([libm not found]))
-AC_CHECK_LIB([resolv], [__res_nquery], [true], AC_MSG_FAILURE([libresolv not found]))
 if test x"$linux" = x"yes"; then
 	AC_CHECK_LIB([keyutils], [add_key], [true], AC_MSG_FAILURE([libkeyutils not found]))
 fi
@@ -539,6 +572,17 @@ AC_CHECK_FUNC([fallocate],
 	[AC_DEFINE([CEPH_HAVE_FALLOCATE], [], [fallocate(2) is supported])],
 	[])
 
+#
+# Test for time-related `struct stat` members.
+#
+
+AC_CHECK_MEMBER([struct stat.st_mtim.tv_nsec],
+  [AC_DEFINE(HAVE_STAT_ST_MTIM_TV_NSEC, 1,
+    [Define if you have struct stat.st_mtim.tv_nsec])])
+
+AC_CHECK_MEMBER([struct stat.st_mtimespec.tv_nsec],
+  [AC_DEFINE(HAVE_STAT_ST_MTIMESPEC_TV_NSEC, 1,
+    [Define if you have struct stat.st_mtimespec.tv_nsec])])
 
 AC_CHECK_HEADERS([arpa/nameser_compat.h])
 AC_CHECK_HEADERS([sys/prctl.h])

diff --git a/doc/install/build-ceph.rst b/doc/install/build-ceph.rst
@@ -50,7 +50,7 @@ installed on your host. ::
 
 On Debian/Wheezy, you may also need:: 
 
-	keyutils-dev libaio libboost-thread-dev
+	libkeyutils-dev libaio-dev libboost-thread-dev
 
 .. note:: Some distributions that support Google's memory profiler tool may use
    a different package name (e.g., ``libgoogle-perftools4``).

diff --git a/doc/install/get-packages.rst b/doc/install/get-packages.rst
@@ -71,9 +71,10 @@ Add Keys
 ========
 
 Add a key to your system's list of trusted keys to avoid a security warning. For
-major releases (e.g., ``cuttlefish``, ``emperor``) and development releases
-(``release-name-rc1``, ``release-name-rc2``), use the ``release.asc`` key. For
-development testing packages, use the ``autobuild.asc`` key (developers and QA).
+major releases (e.g., ``cuttlefish``, ``dumpling``, ``emperor``) and development
+releases (``release-name-rc1``, ``release-name-rc2``), use the ``release.asc``
+key. For development testing packages, use the ``autobuild.asc`` key (developers
+and QA).
 
 
 APT

diff --git a/doc/install/upgrading-ceph.rst b/doc/install/upgrading-ceph.rst
@@ -250,6 +250,44 @@ Then add a new ``ceph.repo`` repository entry with the following contents.
    interface or the ``ceph-deploy`` tool.
 
 
+Dumpling to Emperor
+===================
+
+When upgrading from Dumpling (v0.64) you may perform a rolling
+upgrade.
+
+Replace any reference to older repositories with a reference to the
+Emperor repository. For example, with ``apt`` perform the following:: 
+
+	sudo rm /etc/apt/sources.list.d/ceph.list
+	echo deb http://ceph.com/debian-emperor/ $(lsb_release -sc) main | sudo tee /etc/apt/sources.list.d/ceph.list
+
+With CentOS/Red Hat distributions, remove the old repository. :: 
+
+	sudo rm /etc/yum.repos.d/ceph.repo
+
+Then add a new ``ceph.repo`` repository entry with the following contents and
+replace ``{distro}`` with your distribution (e.g., ``el6``, ``rhel6``, etc). 
+
+.. code-block:: ini
+
+	[ceph]
+	name=Ceph Packages and Backports $basearch
+	baseurl=http://ceph.com/rpm-emperor/{distro}/$basearch
+	enabled=1
+	gpgcheck=1
+	type=rpm-md
+	gpgkey=https://ceph.com/git/?p=ceph.git;a=blob_plain;f=keys/release.asc	
+
+
+.. note:: Ensure you use the correct URL for your distribution. Check the
+   http://ceph.com/rpm directory for your distribution. 
+
+.. note:: Since you can upgrade using ``ceph-deploy`` you will only need to add
+   the repository on Ceph Client nodes where you use the ``ceph`` command line 
+   interface or the ``ceph-deploy`` tool.
+
+
 Command Line Utility
 --------------------
 

diff --git a/doc/man/8/rbd.rst b/doc/man/8/rbd.rst
@@ -143,21 +143,21 @@ Commands
   specified. Size will be the same as the parent snapshot.
 
   The parent snapshot must be protected (see `rbd snap protect`).
-  This requires format 2.
+  This requires image format 2.
 
 :command:`flatten` [*image-name*]
   If image is a clone, copy all shared blocks from the parent snapshot and
   make the child independent of the parent, severing the link between
   parent snap and child.  The parent snapshot can be unprotected and
   deleted if it has no further dependent clones.
 
-  This requires format 2.
+  This requires image format 2.
 
 :command:`children` [*image-name*]
   List the clones of the image at the given snapshot. This checks
   every pool, and outputs the resulting poolname/imagename.
 
-  This requires format 2.
+  This requires image format 2.
 
 :command:`resize` [*image-name*] [--allow-shrink]
   Resizes rbd image. The size parameter also needs to be specified.
@@ -198,7 +198,7 @@ Commands
 
 :command:`cp` [*src-image*] [*dest-image*]
   Copies the content of a src-image into the newly created dest-image.
-  dest-image will have the same size, order, and format as src-image.
+  dest-image will have the same size, order, and image format as src-image.
 
 :command:`mv` [*src-image*] [*dest-image*]
   Renames an image.  Note: rename across pools is not supported.
@@ -226,14 +226,14 @@ Commands
   refer to this snapshot.  `rbd clone` will fail on a nonprotected
   snapshot.
 
-  This requires format 2.
+  This requires image format 2.
 
 :command:`snap` unprotect [*image-name*]
   Unprotect a snapshot from deletion (undo `snap protect`).  If cloned
   children remain, `snap unprotect` fails.  (Note that clones may exist
   in different pools than the parent snapshot.)
 
-  This requires format 2.
+  This requires image format 2.
 
 :command:`map` [*image-name*]
   Maps the specified image to a block device via the rbd kernel module.
@@ -351,7 +351,7 @@ To unmap an image::
 
 To create an image and a clone from it::
 
-       rbd import --format 2 image mypool/parent
+       rbd import --image-format 2 image mypool/parent
        rbd snap create --snap snapname mypool/parent
        rbd snap protect mypool/parent@snap
        rbd clone mypool/parent@snap otherpool/child
@@ -360,11 +360,11 @@ To create an image with a smaller stripe_unit (to better distribute small writes
 
        rbd -p mypool create myimage --size 102400 --stripe-unit 65536 --stripe-count 16
 
-To change an image from one format to another, export it and then
-import it as the desired format::
+To change an image from one image format to another, export it and then
+import it as the desired image format::
 
        rbd export mypool/myimage@snap /tmp/img
-       rbd import --format 2 /tmp/img mypool/myimage2
+       rbd import --image-format 2 /tmp/img mypool/myimage2
 
 To lock an image for exclusive use::
 

diff --git a/doc/rbd/rbd-snapshot.rst b/doc/rbd/rbd-snapshot.rst
@@ -177,7 +177,7 @@ a snapshot refers to the snapshot, so you **MUST** protect the snapshot before
 you clone it. The following diagram depicts the process.
 
 .. note:: Ceph only supports cloning for ``format 2`` images (i.e., created with 
-  ``rbd create --format 2``), and is not yet supported by the kernel ``rbd`` module. 
+  ``rbd create --image-format 2``), and is not yet supported by the kernel ``rbd`` module. 
   So you MUST use QEMU/KVM or ``librbd`` directly to access clones in the current
   release.
 
@@ -321,4 +321,4 @@ For example::
 .. _QEMU: ../qemu-rbd/
 .. _OpenStack: ../rbd-openstack/
 .. _CloudStack: ../rbd-cloudstack/
-.. _libvirt: ../libvirt/
+.. _libvirt: ../libvirt/
diff --git a/doc/release-notes.rst b/doc/release-notes.rst
@@ -2,6 +2,50 @@
  Release Notes
 ===============
 
+v0.72.1 Emperor
+---------------
+
+Important Note
+~~~~~~~~~~~~~~
+
+When you are upgrading from Dumpling to Emperor, do not run any of the
+"ceph osd pool set" commands while your monitors are running separate versions.
+Doing so could result in inadvertently changing cluster configuration settings
+that exhaust compute resources in your OSDs.
+
+Changes
+~~~~~~~
+
+* osd: fix upgrade bug #6761
+* ceph_filestore_tool: introduced tool to repair errors caused by #6761
+
+This release addresses issue #6761.  Upgrading to Emperor can cause
+reads to begin returning ENFILE (too many open files).  v0.72.1 fixes
+that upgrade issue and adds a tool ceph_filestore_tool to repair osd
+stores affected by this bug.
+
+To repair a cluster affected by this bug:
+
+#. Upgrade all osd machines to v0.72.1
+#. Install the ceph-test package on each osd machine to get ceph_filestore_tool
+#. Stop all osd processes
+#. To see all lost objects, run the following on each osd with the osd stopped and
+   the osd data directory mounted::
+
+     ceph_filestore_tool --list-lost-objects=true --filestore-path=<path-to-osd-filestore> --journal-path=<path-to-osd-journal>
+
+#. To fix all lost objects, run the following on each osd with the
+   osd stopped and the osd data directory mounted::
+
+     ceph_filestore_tool --fix-lost-objects=true --list-lost-objects=true --filestore-path=<path-to-osd-filestore> --journal-path=<path-to-osd-journal>
+
+#. Once lost objects have been repaired on each osd, you can restart
+   the cluster.
+
+Note, the ceph_filestore_tool performs a scan of all objects on the
+osd and may take some time.
+
+
 v0.72 Emperor
 -------------
 
@@ -11,6 +55,14 @@ including multi-datacenter replication for the radosgw, improved
 usability, and lands a lot of incremental performance and internal
 refactoring work to support upcoming features in Firefly.
 
+Important Note
+~~~~~~~~~~~~~~
+
+When you are upgrading from Dumpling to Emperor, do not run any of the
+"ceph osd pool set" commands while your monitors are running separate versions.
+Doing so could result in inadvertently changing cluster configuration settings
+that exhaust compute resources in your OSDs.
+
 Highlights
 ~~~~~~~~~~
 
@@ -38,7 +90,10 @@ Upgrade sequencing
 ~~~~~~~~~~~~~~~~~~
 
 There are no specific upgrade restrictions on the order or sequence of
-upgrading from 0.67.x Dumpling.  
+upgrading from 0.67.x Dumpling. However, you cannot run any of the
+"ceph osd pool set" commands while your monitors are running separate versions.
+Doing so could result in inadvertently changing cluster configuration settings
+and exhausting compute resources in your OSDs.
 
 It is also possible to do a rolling upgrade from 0.61.x Cuttlefish,
 but there are ordering restrictions.  (This is the same set of
@@ -105,11 +160,6 @@ Upgrading from v0.67 Dumpling
   however continue to work just the same as long as the osd already exists
   in the crush map.
 
-* 'ceph osd crush set <id> <weight> <loc..>' no longer adds the osd to the
-  specified location, as that's a job for 'ceph osd crush add'.  It will
-  however continue to work just the same as long as the osd already exists
-  in the crush map.
-
 * The OSD now enforces that class write methods cannot both mutate an
   object and return data.  The rbd.assign_bid method, the lone
   offender, has been removed.  This breaks compatibility with

diff --git a/doc/start/os-recommendations.rst b/doc/start/os-recommendations.rst
@@ -37,7 +37,7 @@ specific distributions aside from the kernel and system initialization
 package (i.e., sysvinit, upstart, systemd).
 
 
-Emperor (0.71)
+Emperor (0.72)
 --------------
 
 +----------+----------+--------------------+--------------+---------+------------+

diff --git a/doc/start/quick-start-preflight.rst b/doc/start/quick-start-preflight.rst
@@ -100,9 +100,6 @@ For Debian and Ubuntu distributions, perform the following steps:
 #. Add the release key::
 
 	wget -q -O- 'https://ceph.com/git/?p=ceph.git;a=blob_plain;f=keys/release.asc' | sudo apt-key add -
-	echo deb http://ceph.com/debian-dumpling/ $(lsb_release -sc) main | sudo tee /etc/apt/sources.list.d/ceph.list
-	sudo apt-get update
-	sudo apt-get install ceph-deploy
 
 #. Add the Ceph packages to your repository. Replace ``{ceph-stable-release}``
    with a stable Ceph release (e.g., ``cuttlefish``, ``dumpling``, etc.). 

diff --git a/man/ceph-authtool.8 b/man/ceph-authtool.8
@@ -1,4 +1,4 @@
-.TH "CEPH-AUTHTOOL" "8" "April 29, 2013" "dev" "Ceph"
+.TH "CEPH-AUTHTOOL" "8" "November 18, 2013" "dev" "Ceph"
 .SH NAME
 ceph-authtool \- ceph keyring manipulation tool
 .
@@ -34,7 +34,7 @@ level margin: \\n[rst2man-indent\\n[rst2man-indent-level]]
 .nf
 \fBceph\-authtool\fP \fIkeyringfile\fP [ \-l | \-\-list ] [ \-C | \-\-create\-keyring
 ] [ \-p | \-\-print ] [ \-n | \-\-name \fIentityname\fP ] [ \-\-gen\-key ] [ \-a |
-\-\-add\-key \fIbase64_key\fP ] [ \-\-caps \fIcapfils\fP ]
+\-\-add\-key \fIbase64_key\fP ] [ \-\-caps \fIcapfile\fP ]
 .fi
 .sp
 .SH DESCRIPTION
@@ -69,33 +69,16 @@ will create a new keyring, overwriting any existing keyringfile
 .UNINDENT
 .INDENT 0.0
 .TP
-.B \-g, \-\-gen\-key
+.B \-\-gen\-key
 will generate a new secret key for the specified entityname
 .UNINDENT
 .INDENT 0.0
 .TP
-.B \-a, \-\-add\-key
+.B \-\-add\-key
 will add an encoded key to the keyring
 .UNINDENT
 .INDENT 0.0
 .TP
-.B \-u, \-\-set\-uid
-sets the auid (authenticated user id) for the specified entityname
-.UNINDENT
-.INDENT 0.0
-.TP
-.B \-\-gen\-print\-key
-will generate and print a new secret key without adding it to the keyringfile
-
-NOTE: will work without a given keyringfile
-.UNINDENT
-.INDENT 0.0
-.TP
-.B \-\-import\-keyring
-will import the content of a given keyring into the keyringfile
-.UNINDENT
-.INDENT 0.0
-.TP
 .B \-\-cap subsystem capability
 will set the capability for given subsystem
 .UNINDENT

diff --git a/man/ceph-conf.8 b/man/ceph-conf.8
@@ -1,4 +1,4 @@
-.TH "CEPH-CONF" "8" "April 29, 2013" "dev" "Ceph"
+.TH "CEPH-CONF" "8" "November 18, 2013" "dev" "Ceph"
 .SH NAME
 ceph-conf \- ceph conf file tool
 .
@@ -110,7 +110,6 @@ information.
 .SH SEE ALSO
 .sp
 \fBceph\fP(8),
-\fBmkcephfs\fP(8)
 .SH COPYRIGHT
 2010-2013, Inktank Storage, Inc. and contributors. Licensed under Creative Commons BY-SA
 .\" Generated by docutils manpage writer.