summaryrefslogtreecommitdiffstats
path: root/tests/bitrot
diff options
context:
space:
mode:
authorRaghavendra Bhat <raghavendra@redhat.com>2019-03-11 12:16:50 -0400
committerKotresh HR <khiremat@redhat.com>2019-04-25 05:19:41 +0000
commit647e2d3a9fbe36f2fbf062feda53b7cb6aa4830b (patch)
tree345e4894e312d9290c066417467202cc32a74f22 /tests/bitrot
parent8d51cb05d9745643f2dfaa6ef35938db9d2d47f6 (diff)
features/bit-rot: Unconditionally sign the files during oneshot crawl
Currently bit-rot feature has an issue with disabling and reenabling it on the same volume. Consider enabling bit-rot detection which goes on to crawl and sign all the files present in the volume. Then some files are modified and the bit-rot daemon goes on to sign the modified files with the correct signature. Now, disable bit-rot feature. While, signing and scrubbing are not happening, previous checksums of the files continue to exist as extended attributes. Now, if some files with checksum xattrs get modified, they are not signed with new signature as the feature is off. At this point, if the feature is enabled again, the bit rot daemon will go and sign those files which does not have any bit-rot specific xattrs (i.e. those files which were created after bit-rot was disabled). Whereas the files with bit-rot xattrs wont get signed with proper new checksum. At this point if scrubber runs, it finds the on disk checksum and the actual checksum of the file to be different (because the file got modified) and marks the file as corrupted. FIX: The fix is to unconditionally sign the files when the bit-rot daemon comes up (instead of skipping the files with bit-rot xattrs). Change-Id: Iadfb47dd39f7e2e77f22d549a4a07a385284f4f5 fixes: bz#1700078 Signed-off-by: Raghavendra Bhat <raghavendra@redhat.com>
Diffstat (limited to 'tests/bitrot')
-rw-r--r--tests/bitrot/bug-1700078.t87
1 files changed, 87 insertions, 0 deletions
diff --git a/tests/bitrot/bug-1700078.t b/tests/bitrot/bug-1700078.t
new file mode 100644
index 0000000..f273742
--- /dev/null
+++ b/tests/bitrot/bug-1700078.t
@@ -0,0 +1,87 @@
+#!/bin/bash
+
+. $(dirname $0)/../include.rc
+. $(dirname $0)/../volume.rc
+
+cleanup;
+
+## Start glusterd
+TEST glusterd;
+TEST pidof glusterd;
+
+## Lets create and start the volume
+TEST $CLI volume create $V0 $H0:$B0/${V0}1
+TEST $CLI volume start $V0
+
+## Enable bitrot for volume $V0
+TEST $CLI volume bitrot $V0 enable
+
+## Turn off quick-read so that it wont cache the contents
+# of the file in lookup. For corrupted files, it might
+# end up in reads being served from the cache instead of
+# an error.
+TEST $CLI volume set $V0 performance.quick-read off
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" get_bitd_count
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT 'Active' scrub_status $V0 'State of scrub'
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT '/var/log/glusterfs/bitd.log' scrub_status $V0 'Bitrot error log location'
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT '/var/log/glusterfs/scrub.log' scrub_status $V0 'Scrubber error log location'
+
+## Set expiry-timeout to 1 sec
+TEST $CLI volume set $V0 features.expiry-time 1
+
+##Mount $V0
+TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+
+## Turn off quick-read xlator so that, the contents are not served from the
+# quick-read cache.
+TEST $CLI volume set $V0 performance.quick-read off
+
+#Create sample file
+TEST `echo "1234" > $M0/FILE1`
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT 'trusted.bit-rot.signature' check_for_xattr 'trusted.bit-rot.signature' "/$B0/${V0}1/FILE1"
+
+##disable bitrot
+TEST $CLI volume bitrot $V0 disable
+
+## modify the file
+TEST `echo "write" >> $M0/FILE1`
+
+# unmount and remount when the file has to be accessed.
+# This is to ensure that, when the remount happens,
+# and the file is read, its contents are served from the
+# brick instead of cache.
+EXPECT_WITHIN $UMOUNT_TIMEOUT "Y" force_umount $M0
+
+##enable bitrot
+TEST $CLI volume bitrot $V0 enable
+
+# expiry time is set to 1 second. Hence sleep for 2 seconds for the
+# oneshot crawler to finish its crawling and sign the file properly.
+sleep 2
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT "1" get_bitd_count
+
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT 'Active' scrub_status $V0 'State of scrub'
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT '/var/log/glusterfs/bitd.log' scrub_status $V0 'Bitrot error log location'
+EXPECT_WITHIN $PROCESS_UP_TIMEOUT '/var/log/glusterfs/scrub.log' scrub_status $V0 'Scrubber error log location'
+
+## Ondemand scrub
+TEST $CLI volume bitrot $V0 scrub ondemand
+
+# the scrub ondemand CLI command, just ensures that
+# the scrubber has received the ondemand scrub directive
+# and started. sleep for 2 seconds for scrubber to finish
+# crawling and marking file(s) as bad (if if finds that
+# corruption has happened) which are filesystem operations.
+sleep 2
+
+TEST ! getfattr -n 'trusted.bit-rot.bad-file' $B0/${V0}1/FILE1
+
+##Mount $V0
+TEST glusterfs --volfile-id=$V0 --volfile-server=$H0 $M0 --attribute-timeout=0 --entry-timeout=0
+
+TEST cat $M0/FILE1
+
+cleanup;