split read_fs_info into linux & freebsd parts (#625)

FreeBSD doesn't have blkid, so we want to use geom to list devices and
their fstypes and labels.

This PR also adds `jail` to the list of is_container()
And we now also properly cache geom and blkid output!

A test is added to verify the new behaviour by correctly identifying
NoCloud on FreeBSD.

Co-authored-by: Scott Moser <smoser@brickies.net>

2 files changed, 137 insertions, 9 deletions

diff --git a/tests/unittests/test_ds_identify.py b/tests/unittests/test_ds_identify.py
index 9314b244..5f8a4a29 100644
--- a/tests/unittests/test_ds_identify.py
+++ b/tests/unittests/test_ds_identify.py
@@ -20,6 +20,8 @@ UNAME_MYSYS = ("Linux bart 4.4.0-62-generic #83-Ubuntu "
 UNAME_PPC64EL = ("Linux diamond 4.4.0-83-generic #106-Ubuntu SMP "
                  "Mon Jun 26 17:53:54 UTC 2017 "
                  "ppc64le ppc64le ppc64le GNU/Linux")
+UNAME_FREEBSD = ("FreeBSD fbsd12-1 12.1-RELEASE-p10 "
+                 "FreeBSD 12.1-RELEASE-p10 GENERIC  amd64")
 
 BLKID_EFI_ROOT = """
 DEVNAME=/dev/sda1
@@ -80,6 +82,7 @@ MOCK_VIRT_IS_VMWARE = {'name': 'detect_virt', 'RET': 'vmware', 'ret': 0}
 MOCK_VIRT_IS_VM_OTHER = {'name': 'detect_virt', 'RET': 'vm-other', 'ret': 0}
 MOCK_VIRT_IS_XEN = {'name': 'detect_virt', 'RET': 'xen', 'ret': 0}
 MOCK_UNAME_IS_PPC64 = {'name': 'uname', 'out': UNAME_PPC64EL, 'ret': 0}
+MOCK_UNAME_IS_FREEBSD = {'name': 'uname', 'out': UNAME_FREEBSD, 'ret': 0}
 
 shell_true = 0
 shell_false = 1
@@ -257,6 +260,10 @@ class TestDsIdentify(DsIdentifyBase):
         """EC2: bobrightbox.com in product_serial is not brightbox'"""
         self._test_ds_not_found('Ec2-brightbox-negative')
 
+    def test_freebsd_nocloud(self):
+        """NoCloud identified on FreeBSD via label by geom."""
+        self._test_ds_found('NoCloud-fbsd')
+
     def test_gce_by_product_name(self):
         """GCE identifies itself with product_name."""
         self._test_ds_found('GCE')
@@ -725,6 +732,26 @@ def blkid_out(disks=None):
     return '\n'.join(lines)
 
 
+def geom_out(disks=None):
+    """Convert a list of disk dictionaries into geom content.
+
+    geom called with -a (provider) and -s (script-friendly), will produce the
+    following output:
+
+      gpt/gptboot0  N/A  vtbd1p1
+         gpt/swap0  N/A  vtbd1p2
+    iso9660/cidata  N/A  vtbd2
+    """
+    if disks is None:
+        disks = []
+    lines = []
+    for disk in disks:
+        lines.append("%s/%s  N/A  %s" % (
+            disk["TYPE"], disk["LABEL"], disk["DEVNAME"]))
+        lines.append("")
+    return '\n'.join(lines)
+
+
 def _print_run_output(rc, out, err, cfg, files):
     """A helper to print return of TestDsIdentify.
 
@@ -807,6 +834,19 @@ VALID_CFG = {
             'dev/vdb': 'pretend iso content for cidata\n',
         }
     },
+    'NoCloud-fbsd': {
+        'ds': 'NoCloud',
+        'mocks': [
+            MOCK_VIRT_IS_KVM,
+            MOCK_UNAME_IS_FREEBSD,
+            {'name': 'geom', 'ret': 0,
+             'out': geom_out(
+                 [{'DEVNAME': 'vtbd', 'TYPE': 'iso9660', 'LABEL': 'cidata'}])},
+        ],
+        'files': {
+            '/dev/vtdb': 'pretend iso content for cidata\n',
+        }
+    },
     'NoCloudUpper': {
         'ds': 'NoCloud',
         'mocks': [
diff --git a/tools/ds-identify b/tools/ds-identify
index 4e5700fc..ec9e775e 100755
--- a/tools/ds-identify
+++ b/tools/ds-identify
@@ -92,7 +92,8 @@ _DI_LOGGED=""
 # set DI_MAIN='noop' in environment to source this file with no main called.
 DI_MAIN=${DI_MAIN:-main}
 
-DI_BLKID_OUTPUT=""
+DI_BLKID_EXPORT_OUT=""
+DI_GEOM_LABEL_STATUS_OUT=""
 DI_DEFAULT_POLICY="search,found=all,maybe=all,notfound=${DI_DISABLED}"
 DI_DEFAULT_POLICY_NO_DMI="search,found=all,maybe=all,notfound=${DI_ENABLED}"
 DI_DMI_CHASSIS_ASSET_TAG=""
@@ -231,8 +232,19 @@ ensure_sane_path() {
     done
 }
 
-read_fs_info() {
-    cached "${DI_BLKID_OUTPUT}" && return 0
+blkid_export() {
+    # call 'blkid -c /dev/null export', set DI_BLKID_EXPORT_OUT
+    cached "$DI_BLKID_EXPORT_OUT" && return 0
+    local out="" ret=0
+    out=$(blkid -c /dev/null -o export) && DI_BLKID_EXPORT_OUT="$out" || {
+        ret=$?
+        error "failed running [$ret]: blkid -c /dev/null -o export"
+        DI_BLKID_EXPORT_OUT="$UNAVAILABLE"
+    }
+    return $ret
+}
+
+read_fs_info_linux() {
     # do not rely on links in /dev/disk which might not be present yet.
     # Note that blkid < 2.22 (centos6, trusty) do not output DEVNAME.
     # that means that DI_ISO9660_DEVS will not be set.
@@ -244,20 +256,23 @@ read_fs_info() {
         return
     fi
     local oifs="$IFS" line="" delim=","
-    local ret=0 out="" labels="" dev="" label="" ftype="" isodevs="" uuids=""
-    out=$(blkid -c /dev/null -o export) || {
-        ret=$?
-        error "failed running [$ret]: blkid -c /dev/null -o export"
+    local ret=0 labels="" dev="" label="" ftype="" isodevs="" uuids=""
+
+    blkid_export
+    ret=$?
+    [ "$DI_BLKID_EXPORT_OUT" = "$UNAVAILABLE" ] && {
         DI_FS_LABELS="$UNAVAILABLE:error"
         DI_ISO9660_DEVS="$UNAVAILABLE:error"
+        DI_FS_UUIDS="$UNAVAILABLE:error"
         return $ret
     }
+
     # 'set --' will collapse multiple consecutive entries in IFS for
     # whitespace characters (\n, tab, " ") so we cannot rely on getting
     # empty lines in "$@" below.
 
     # shellcheck disable=2086
-    { IFS="$CR"; set -- $out; IFS="$oifs"; }
+    { IFS="$CR"; set -- $DI_BLKID_EXPORT_OUT; IFS="$oifs"; }
 
     for line in "$@"; do
         case "${line}" in
@@ -281,6 +296,74 @@ read_fs_info() {
     DI_ISO9660_DEVS="${isodevs#,}"
 }
 
+geom_label_status_as() {
+    # call 'geom label status -as', set DI_GEOM_LABEL_STATUS_OUT
+    cached "$DI_GEOM_LABEL_STATUS_OUT" && return 0
+    local out="" ret=0
+    out=$(geom label status -as) && DI_GEOM_LABEL_STATUS_OUT="$out" || {
+        ret=$?
+        error "failed running [$ret]: geom label status -as"
+        DI_GEOM_LABEL_STATUS_OUT="$UNAVAILABLE"
+    }
+    return $ret
+}
+
+
+read_fs_info_freebsd() {
+    local oifs="$IFS" line="" delim=","
+    local ret=0 labels="" dev="" label="" ftype="" isodevs=""
+
+    geom_label_status_as
+    ret=$?
+    [ "$DI_GEOM_LABEL_STATUS_OUT" = "$UNAVAILABLE" ] && {
+        DI_FS_LABELS="$UNAVAILABLE:error"
+        DI_ISO9660_DEVS="$UNAVAILABLE:error"
+        return $ret
+    }
+
+    # The expected output looks like this:
+    #   gpt/gptboot0 N/A vtbd1p1
+    #      gpt/swap0 N/A vtbd1p2
+    # iso9660/cidata N/A vtbd2
+
+    # shellcheck disable=2086
+    { IFS="$CR"; set -- $DI_GEOM_LABEL_STATUS_OUT; IFS="$oifs"; }
+
+    for line in "$@"; do
+        # shellcheck disable=2086
+        set -- $line
+        provider=$1
+        ftype="${provider%/*}"
+        label="${provider#*/}"
+        dev=$3
+
+        [ -n "$dev" -a "$ftype" = "iso9660" ] &&
+            isodevs="${isodevs},${dev}=$label"
+
+        labels="${labels}${label}${delim}"
+    done
+
+    DI_FS_LABELS="${labels%${delim}}"
+    DI_ISO9660_DEVS="${isodevs#,}"
+}
+
+read_fs_info() {
+    # After calling its subfunctions, read_fs_info() will set the following
+    # variables:
+    #
+    # - DI_FS_LABELS
+    # - DI_ISO9660_DEVS
+    # - DI_FS_UUIDS
+
+    if [ "$DI_UNAME_KERNEL_NAME" = "FreeBSD" ]; then
+        read_fs_info_freebsd
+        return $?
+    else
+        read_fs_info_linux
+        return $?
+    fi
+}
+
 cached() {
     [ -n "$1" ] && _RET="$1" && return || return 1
 }
@@ -319,6 +402,11 @@ detect_virt() {
                 *) virt="$out"
             esac
         }
+        out=$(sysctl -qn security.jail.jailed 2>/dev/null) && {
+            if [ "$out" = "1" ]; then
+                virt="jail"
+            fi
+        }
     fi
     _RET="$virt"
 }
@@ -331,7 +419,7 @@ read_virt() {
 
 is_container() {
     case "${DI_VIRT}" in
-        container-other|lxc|lxc-libvirt|systemd-nspawn|docker|rkt) return 0;;
+        container-other|lxc|lxc-libvirt|systemd-nspawn|docker|rkt|jail) return 0;;
         *) return 1;;
     esac
 }