Bug 1742815
| Summary: | Online reencryption ignores optimal io size parameter | ||
|---|---|---|---|
| Product: | Red Hat Enterprise Linux 8 | Reporter: | Corey Marthaler <cmarthal> |
| Component: | cryptsetup | Assignee: | Ondrej Kozina <okozina> |
| Status: | CLOSED ERRATA | QA Contact: | guazhang <guazhang> |
| Severity: | medium | Docs Contact: | |
| Priority: | urgent | ||
| Version: | 8.1 | CC: | agk, guazhang, jbrassow, mbroz, okozina, prajnoha, rhandlin, wchadwic |
| Target Milestone: | rc | Flags: | pm-rhel:
mirror+
|
| Target Release: | 8.0 | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | cryptsetup-2.2.0-2.el8 | Doc Type: | No Doc Update |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2019-11-05 22:17:22 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Corey Marthaler
2019-08-16 21:02:07 UTC
I'll try to reproduce it locally but could you please add some more info here so that I'm sure it's the same problem? 1) cryptsetup luksDump output for the device after encryption finished 2) "lsblk -t" output with device tree underneath the luks_origin mapping (after encryption is finished) 3) Initialize reencryption operation only by adding --init-only parameter and then paste here cryptsetup luksDump command output again? I've reproduced it locally but only in limited scope on a test drive with 4k sector size with 512B emulation (logical size 512, physical sector size 4096). I can hit that when I ask explicitly for weird sized reencryption hotzone via --hotzone-size parameter (in my case --hotzone-size 2047s). In default mode I can't hit it yet. So I think it mostly depends on how you encrypted the device and what's beneath the dm-crypt device when you perform reencryption. I don't say this is not a bug, but I definitely need more info before I start tinkering on patch :) I haven't had time to boil this down yet, but this appears to be the test case causing these messages. I was just able to reproduce this now:
[root@hayes-03 ~]# vgcreate cache_sanity /dev/sd[bcdefghijklmnop]1
Volume group "cache_sanity" successfully created
[root@hayes-03 ~]# pvchange --addtag slow /dev/sdf1
Physical volume "/dev/sdf1" changed
1 physical volume changed / 0 physical volumes not changed
[root@hayes-03 ~]# pvchange --addtag fast /dev/sdl1
Physical volume "/dev/sdl1" changed
1 physical volume changed / 0 physical volumes not changed
[root@hayes-03 ~]# lvcreate --wipesignatures y -L 4G -n corigin cache_sanity @slow
Logical volume "corigin" created.
[root@hayes-03 ~]# lvcreate -L 4G -n resize cache_sanity @fast
WARNING: crypto_LUKS signature detected on /dev/cache_sanity/resize at offset 0. Wipe it? [y/n]: y
Wiping crypto_LUKS signature on /dev/cache_sanity/resize.
WARNING: crypto_LUKS signature detected on /dev/cache_sanity/resize at offset 16384. Wipe it? [y/n]: y
Wiping crypto_LUKS signature on /dev/cache_sanity/resize.
Logical volume "resize" created.
[root@hayes-03 ~]# lvcreate -L 12M -n resize_meta cache_sanity @fast
Logical volume "resize_meta" created.
[root@hayes-03 ~]# lvconvert --yes --type cache-pool --cachepolicy smq --cachemode writethrough -c 32 --poolmetadata cache_sanity/resize_meta cache_sanity/resize
WARNING: Converting cache_sanity/resize and cache_sanity/resize_meta to cache pool's data and metadata volumes with metadata wiping.
THIS WILL DESTROY CONTENT OF LOGICAL VOLUME (filesystem etc.)
Converted cache_sanity/resize and cache_sanity/resize_meta to cache pool.
[root@hayes-03 ~]# lvconvert --yes --type cache --cachemetadataformat 2 --cachepool cache_sanity/resize cache_sanity/corigin
Logical volume cache_sanity/corigin is now cached.
[root@hayes-03 ~]# lvs --noheadings -o lv_name --select 'cachemetadataformat=2' | grep corigin
corigin
[root@hayes-03 ~]# echo Str0ngP455w0rd### | cryptsetup luksFormat /dev/cache_sanity/corigin
[root@hayes-03 ~]# echo Str0ngP455w0rd### | cryptsetup luksOpen /dev/cache_sanity/corigin luks_corigin
[root@hayes-03 ~]# mkfs.xfs -f /dev/mapper/luks_corigin
meta-data=/dev/mapper/luks_corigin isize=512 agcount=8, agsize=130560 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=1, sparse=1, rmapbt=0
= reflink=1
data = bsize=4096 blocks=1044480, imaxpct=25
= sunit=8 swidth=8 blks
naming =version 2 bsize=4096 ascii-ci=0, ftype=1
log =internal log bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
[root@hayes-03 ~]# mkdir -p /mnt/corigin
[root@hayes-03 ~]# mount /dev/mapper/luks_corigin /mnt/corigin
[root@hayes-03 ~]# /usr/tests/sts-rhel8.1/bin/checkit -w /mnt/corigin -f /tmp/checkit_origin_1 -n 1200
checkit starting with:
CREATE
Num files: 1200
Random Seed: 17661
Verify XIOR Stream: /tmp/checkit_origin_1
Working dir: /mnt/corigin
[root@hayes-03 ~]# /usr/tests/sts-rhel8.1/bin/checkit -w /mnt/corigin -f /tmp/checkit_origin_1 -v
checkit starting with:
VERIFY
Verify XIOR Stream: /tmp/checkit_origin_1
Working dir: /mnt/corigin
[root@hayes-03 ~]# sync
[root@hayes-03 ~]# lvs --noheadings -o lv_attr cache_sanity/corigin
Cwi-aoC---
[root@hayes-03 ~]# lvcreate -s /dev/cache_sanity/corigin -c 64 -n snap1 -L 4100.00m
Logical volume "snap1" created.
[root@hayes-03 ~]# lvs --noheadings -o lv_attr cache_sanity/snap1
swi-a-s---
[root@hayes-03 ~]# lvconvert --uncache /dev/cache_sanity/corigin
Logical volume "resize" successfully removed
Logical volume cache_sanity/corigin is not cached.
[root@hayes-03 ~]# lvcreate -L 5G -n resize cache_sanity @fast
WARNING: crypto_LUKS signature detected on /dev/cache_sanity/resize at offset 0. Wipe it? [y/n]: y
Wiping crypto_LUKS signature on /dev/cache_sanity/resize.
WARNING: crypto_LUKS signature detected on /dev/cache_sanity/resize at offset 16384. Wipe it? [y/n]: y
Wiping crypto_LUKS signature on /dev/cache_sanity/resize.
Logical volume "resize" created.
[root@hayes-03 ~]# lvcreate -L 12M -n resize_meta cache_sanity @fast
Logical volume "resize_meta" created.
[root@hayes-03 ~]# lvconvert --yes --type cache-pool --cachepolicy smq --cachemode writeback -c 32 --poolmetadata cache_sanity/resize_meta cache_sanity/resize
WARNING: Converting cache_sanity/resize and cache_sanity/resize_meta to cache pool's data and metadata volumes with metadata wiping.
THIS WILL DESTROY CONTENT OF LOGICAL VOLUME (filesystem etc.)
Converted cache_sanity/resize and cache_sanity/resize_meta to cache pool.
[root@hayes-03 ~]# lvconvert --yes --type cache --cachemetadataformat 1 --cachepool cache_sanity/resize cache_sanity/corigin
Logical volume cache_sanity/corigin is now cached.
[root@hayes-03 ~]# lvs --noheadings -o lv_name --select 'cachemetadataformat=1' | grep corigin
corigin
[root@hayes-03 ~]# echo Str0ngP455w0rd### | cryptsetup luksOpen /dev/cache_sanity/snap1 luks_snap1
[root@hayes-03 ~]# cryptsetup -v status luks_snap1
/dev/mapper/luks_snap1 is active.
type: LUKS2
cipher: aes-xts-plain64
keysize: 512 bits
key location: keyring
device: /dev/mapper/cache_sanity-snap1
sector size: 512
offset: 32768 sectors
size: 8355840 sectors
mode: read/write
Command successful.
[root@hayes-03 ~]# mkdir -p /mnt/snap1
[root@hayes-03 ~]# mount -o nouuid /dev/mapper/luks_snap1 /mnt/snap1
[root@hayes-03 ~]# /usr/tests/sts-rhel8.1/bin/checkit -w /mnt/snap1 -f /tmp/checkit_origin_1 -v
checkit starting with:
VERIFY
Verify XIOR Stream: /tmp/checkit_origin_1
Working dir: /mnt/snap1
[root@hayes-03 ~]# lvs -a -o +devices
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert Devices
corigin cache_sanity owi-aoC--- 4.00g [resize] [corigin_corig] 0.07 10.87 0.00 corigin_corig(0)
[corigin_corig] cache_sanity owi-aoC--- 4.00g /dev/sdf1(0)
[lvol0_pmspare] cache_sanity ewi------- 12.00m /dev/sdb1(0)
[resize] cache_sanity Cwi---C--- 5.00g 0.07 10.87 0.00 resize_cdata(0)
[resize_cdata] cache_sanity Cwi-ao---- 5.00g /dev/sdl1(0)
[resize_cmeta] cache_sanity ewi-ao---- 12.00m /dev/sdl1(1280)
snap1 cache_sanity swi-aos--- 4.00g corigin 0.07 /dev/sdb1(3)
[root@hayes-03 ~]# /usr/tests/sts-rhel8.1/bin/checkit -w /mnt/corigin -f /tmp/checkit_origin_2 -n 1200
checkit starting with:
CREATE
Num files: 1200
Random Seed: 18105
Verify XIOR Stream: /tmp/checkit_origin_2
Working dir: /mnt/corigin
[root@hayes-03 ~]# lvs -a -o +devices
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert Devices
corigin cache_sanity owi-aoC--- 4.00g [resize] [corigin_corig] 0.07 10.87 0.00 corigin_corig(0)
[corigin_corig] cache_sanity owi-aoC--- 4.00g /dev/sdf1(0)
[lvol0_pmspare] cache_sanity ewi------- 12.00m /dev/sdb1(0)
[resize] cache_sanity Cwi---C--- 5.00g 0.07 10.87 0.00 resize_cdata(0)
[resize_cdata] cache_sanity Cwi-ao---- 5.00g /dev/sdl1(0)
[resize_cmeta] cache_sanity ewi-ao---- 12.00m /dev/sdl1(1280)
snap1 cache_sanity swi-aos--- 4.00g corigin 7.83 /dev/sdb1(3)
[root@hayes-03 ~]# echo Str0ngP455w0rd### | cryptsetup reencrypt --resilience journal --active-name luks_corigin
Finished, time 04:31.803, 4080 MiB written, speed 15.0 MiB/s
# While the above is running:
[...]
Aug 20 15:00:27 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7638736 len 31696) caused an alignment inconsistency
Aug 20 15:00:27 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7670432 len 685408) caused an alignment inconsistency
Aug 20 15:00:28 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7670432 len 31696) caused an alignment inconsistency
Aug 20 15:00:28 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7702128 len 653712) caused an alignment inconsistency
Aug 20 15:00:29 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7702128 len 31696) caused an alignment inconsistency
Aug 20 15:00:29 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7733824 len 622016) caused an alignment inconsistency
Aug 20 15:00:30 hayes-03 kernel: device-mapper: table: 253:10: adding target device (start sect 7765520 len 590320) caused an alignment inconsistency
[...]
[root@hayes-03 ~]# pvs
PV VG Fmt Attr PSize PFree
/dev/sdb1 cache_sanity lvm2 a-- 446.62g <442.61g
/dev/sdc1 cache_sanity lvm2 a-- 446.62g 446.62g
/dev/sdd1 cache_sanity lvm2 a-- 446.62g 446.62g
/dev/sde1 cache_sanity lvm2 a-- 446.62g 446.62g
/dev/sdf1 cache_sanity lvm2 a-- <1.82t 1.81t
/dev/sdg1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdh1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdi1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdj1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdk1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdl1 cache_sanity lvm2 a-- <1.82t 1.81t
/dev/sdm1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdn1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdo1 cache_sanity lvm2 a-- <1.82t <1.82t
/dev/sdp1 cache_sanity lvm2 a-- <1.82t <1.82t
[root@hayes-03 ~]# cryptsetup luksDump /dev/cache_sanity/corigin
LUKS header information
Version: 2
Epoch: 536
Metadata area: 16384 [bytes]
Keyslots area: 16744448 [bytes]
UUID: 94e9101a-d009-4d2f-990d-d4efb6516fd2
Label: (no label)
Subsystem: (no subsystem)
Flags: (no flags)
Data segments:
0: crypt
offset: 16777216 [bytes]
length: (whole device)
cipher: aes-xts-plain64
sector: 512 [bytes]
Keyslots:
1: luks2
Key: 512 bits
Priority: normal
Cipher: aes-xts-plain64
Cipher key: 512 bits
PBKDF: argon2i
Time cost: 4
Memory: 1048576
Threads: 4
Salt: 34 ce 58 46 b5 d7 ed f5 09 fb 41 62 02 9c bd 6d
01 17 98 d7 fa 13 7a 8e c1 cc b0 d0 6e a6 38 f2
AF stripes: 4000
AF hash: sha256
Area offset:290816 [bytes]
Area length:258048 [bytes]
Digest ID: 1
Tokens:
Digests:
1: pbkdf2
Hash: sha256
Iterations: 1000
Salt: d3 c8 9f 25 8e 14 67 a0 02 5b da a1 20 b7 76 9e
d0 9a cd 06 26 e5 80 20 ee 37 58 29 0f 81 c8 e9
Digest: a4 45 de 2a 28 a8 dc a2 05 1c 76 e9 93 c3 22 29
be 2e 22 b6 a9 60 91 27 a0 89 9d ca 6d b7 34 8f
[root@hayes-03 ~]# cryptsetup luksDump /dev/cache_sanity/snap1
LUKS header information
Version: 2
Epoch: 3
Metadata area: 16384 [bytes]
Keyslots area: 16744448 [bytes]
UUID: 94e9101a-d009-4d2f-990d-d4efb6516fd2
Label: (no label)
Subsystem: (no subsystem)
Flags: (no flags)
Data segments:
0: crypt
offset: 16777216 [bytes]
length: (whole device)
cipher: aes-xts-plain64
sector: 512 [bytes]
Keyslots:
0: luks2
Key: 512 bits
Priority: normal
Cipher: aes-xts-plain64
Cipher key: 512 bits
PBKDF: argon2i
Time cost: 4
Memory: 1048576
Threads: 4
Salt: ce d2 69 49 e0 64 49 7a f0 e9 7b 39 10 ab 84 ab
0e 0c cc e9 e1 1d 46 f6 64 8e 27 c4 35 6a dc 5d
AF stripes: 4000
AF hash: sha256
Area offset:32768 [bytes]
Area length:258048 [bytes]
Digest ID: 0
Tokens:
Digests:
0: pbkdf2
Hash: sha256
Iterations: 90269
Salt: 71 d4 ab 78 92 ef 64 36 59 22 b3 27 af 72 7d 4f
65 64 87 c0 7f ea 90 ce 82 38 cd 5e 8f b6 ed 1a
Digest: 1c 97 3e 57 af 75 c6 e0 d1 d4 d9 99 a5 18 ca 04
73 a9 50 a7 bd 51 24 5c a0 c7 bb e4 72 c0 dd 9a
[root@hayes-03 ~]# lsblk -t
NAME ALIGNMENT MIN-IO OPT-IO PHY-SEC LOG-SEC ROTA SCHED RQ-SIZE RA WSAME
sda 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsda1 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsda2 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsda3 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
sdb 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsdb1 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââcache_sanity-snap1-cow 0 4096 0 4096 512 1 128 4096 0B
ââcache_sanity-snap1 0 32768 32768 4096 512 1 128 4096 0B
ââluks_snap1 0 32768 32768 4096 512 1 128 4096 0B
sdc 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsdc1 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
sdd 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsdd1 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
sde 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
ââsde1 0 4096 0 4096 512 1 mq-deadline 256 4096 0B
sdf 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdf1 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââcache_sanity-corigin_corig 0 512 0 512 512 1 128 4096 0B
ââcache_sanity-corigin-real 0 32768 32768 512 512 1 128 4096 0B
ââcache_sanity-corigin 0 32768 32768 512 512 1 128 4096 0B
â ââluks_corigin 0 32768 32768 512 512 1 128 4096 0B
ââcache_sanity-snap1 0 32768 32768 4096 512 1 128 4096 0B
ââluks_snap1 0 32768 32768 4096 512 1 128 4096 0B
sdg 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdg1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdh 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdh1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdi 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdi1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdj 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdj1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdk 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdk1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdl 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdl1 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââcache_sanity-resize_cdata 0 512 0 512 512 1 128 4096 0B
â ââcache_sanity-corigin-real 0 32768 32768 512 512 1 128 4096 0B
â ââcache_sanity-corigin 0 32768 32768 512 512 1 128 4096 0B
â â ââluks_corigin 0 32768 32768 512 512 1 128 4096 0B
â ââcache_sanity-snap1 0 32768 32768 4096 512 1 128 4096 0B
â ââluks_snap1 0 32768 32768 4096 512 1 128 4096 0B
ââcache_sanity-resize_cmeta 0 512 0 512 512 1 128 4096 0B
ââcache_sanity-corigin-real 0 32768 32768 512 512 1 128 4096 0B
ââcache_sanity-corigin 0 32768 32768 512 512 1 128 4096 0B
â ââluks_corigin 0 32768 32768 512 512 1 128 4096 0B
ââcache_sanity-snap1 0 32768 32768 4096 512 1 128 4096 0B
ââluks_snap1 0 32768 32768 4096 512 1 128 4096 0B
sdm 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdm1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdn 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdn1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdo 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdo1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sdp 0 512 0 512 512 1 mq-deadline 256 4096 0B
ââsdp1 0 512 0 512 512 1 mq-deadline 256 4096 0B
sr0 0 512 0 512 512 1 mq-deadline 64 128 0B
Online reencryption ignores optiomal-io size reported by device. To reproduce it easily create following scsi debug device: 1) modprobe scsi_debug dev_size_mb=256 opt_blks=64 opt_xferlen_exp=6 (results in /dev/sdx) 2) cryptsetup luksFormat /dev/sdx --offset 32768 3) cryptsetup open /dev/sdx sdx_crypt 4) cryptsetup reencrypt --resilience journal /dev/sdx (There's no risk to data AFAICT but it most probably cause performance degradation on more complex storage setup) Fixed upstream with https://gitlab.com/cryptsetup/cryptsetup/commit/6af4f7ca3d2ddcbe512fcf24b900e368bb255cfd The bug was introduced and fixed during 8.1 devel phase, no docs text needed. Fix verified in the latest rpms. cryptsetup-2.2.0-2.el8 BUILT: Fri Aug 30 07:54:20 CDT 2019 cryptsetup-libs-2.2.0-2.el8 BUILT: Fri Aug 30 07:54:20 CDT 2019 cryptsetup-reencrypt-2.2.0-2.el8 BUILT: Fri Aug 30 07:54:20 CDT 2019 Both the scenarios (comment #3 and comment #4) above that cause this no longer do. [root@hayes-01 ~]# modprobe scsi_debug dev_size_mb=256 opt_blks=64 opt_xferlen_exp=6 [root@hayes-01 ~]# cryptsetup luksFormat /dev/sdl --offset 32768 WARNING! ======== This will overwrite data on /dev/sdl irrevocably. Are you sure? (Type uppercase yes): YES Enter passphrase for /dev/sdl: Verify passphrase: [root@hayes-01 ~]# cryptsetup open /dev/sdl sdl_crypt Enter passphrase for /dev/sdl: [root@hayes-01 ~]# cryptsetup reencrypt --resilience journal /dev/sdl Enter passphrase for key slot 0: Auto-detected active dm device 'sdl_crypt' for data device /dev/sdl. Finished, time 00:04.947, 240 MiB written, speed 48.5 MiB/s [root@hayes-01 ~]# grep "caused an alignment inconsistency" /var/log/messages [root@hayes-01 ~]# SCENARIO - [cache_pool_resize_in_between_luks_encryption_operations] Create snapshots of encrypted luks cache origin with fs data, and then extend and reduce both the cache pool volume in between re-encryption stack operations *** Cache info for this scenario *** * origin (slow): /dev/sdm1 * pool (fast): /dev/sdo1 ************************************ Adding "slow" and "fast" tags to corresponding pvs Create origin (slow) volume lvcreate --wipesignatures y -L 4G -n corigin cache_sanity @slow WARNING: crypto_LUKS signature detected on /dev/cache_sanity/corigin at offset 0. Wipe it? [y/n]: [n] Aborted wiping of crypto_LUKS. 1 existing signature left on the device. Create cache data and cache metadata (fast) volumes lvcreate -L 4G -n resize cache_sanity @fast WARNING: crypto_LUKS signature detected on /dev/cache_sanity/resize at offset 16384. Wipe it? [y/n]: [n] Aborted wiping of crypto_LUKS. 1 existing signature left on the device. lvcreate -L 12M -n resize_meta cache_sanity @fast Create cache pool volume by combining the cache data and cache metadata (fast) volumes with policy: mq mode: writeback lvconvert --yes --type cache-pool --cachepolicy mq --cachemode writeback -c 32 --poolmetadata cache_sanity/resize_meta cache_sanity/resize WARNING: Converting cache_sanity/resize and cache_sanity/resize_meta to cache pool's data and metadata volumes with metadata wiping. THIS WILL DESTROY CONTENT OF LOGICAL VOLUME (filesystem etc.) Create cached volume by combining the cache pool (fast) and origin (slow) volumes lvconvert --yes --type cache --cachemetadataformat 1 --cachepool cache_sanity/resize cache_sanity/corigin Encrypting corigin volume cryptsetup luksFormat /dev/cache_sanity/corigin cryptsetup luksOpen /dev/cache_sanity/corigin luks_corigin Placing an xfs filesystem on origin volume Mounting origin volume Writing files to /mnt/corigin Checking files on /mnt/corigin syncing before snap creation... Making 1st snapshot of origin volume lvcreate -s /dev/cache_sanity/corigin -c 64 -n snap1 -L 4296704 Hack until live pool resize is supported: uncache and recreate pool w/ a larger size (lvconvert --uncache) cache_sanity/corigin Create cache data and cache metadata (fast) volumes lvcreate -L 5G -n resize cache_sanity @fast WARNING: crypto_LUKS signature detected on /dev/cache_sanity/resize at offset 0. Wipe it? [y/n]: [n] Aborted wiping of crypto_LUKS. 1 existing signature left on the device. lvcreate -L 12M -n resize_meta cache_sanity @fast Create cache pool volume by combining the cache data and cache metadata (fast) volumes with policy: cleaner mode: writethrough lvconvert --yes --type cache-pool --cachepolicy cleaner --cachemode writethrough -c 32 --poolmetadata cache_sanity/resize_meta cache_sanity/resize WARNING: Converting cache_sanity/resize and cache_sanity/resize_meta to cache pool's data and metadata volumes with metadata wiping. THIS WILL DESTROY CONTENT OF LOGICAL VOLUME (filesystem etc.) Create cached volume by combining the cache pool (fast) and origin (slow) volumes lvconvert --yes --type cache --cachemetadataformat 2 --cachepool cache_sanity/resize cache_sanity/corigin cryptsetup luksOpen /dev/cache_sanity/snap1 luks_snap1 Mounting 1st snap volume Checking files on /mnt/snap1 Writing files to /mnt/corigin (ONLINE) Re-encrypting corigin volume cryptsetup reencrypt --resilience journal --active-name luks_corigin [root@hayes-03 ~]# grep "caused an alignment inconsistency" /var/log/messages [root@hayes-03 ~]# *** Bug 1751003 has been marked as a duplicate of this bug. *** Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2019:3569 |