Bug 1684823 - [GSS] OpenShift Registry out of space - Unable to prune
Summary: [GSS] OpenShift Registry out of space - Unable to prune
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Red Hat Gluster Storage
Classification: Red Hat Storage
Component: fuse
Version: rhgs-3.4
Hardware: x86_64
OS: Linux
urgent
urgent
Target Milestone: ---
: ---
Assignee: Amar Tumballi
QA Contact: Rahul Hinduja
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2019-03-02 19:17 UTC by Cal Calhoun
Modified: 2019-05-21 04:55 UTC (History)
16 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2019-05-21 04:55:23 UTC
Embargoed:
sankarshan: needinfo-
sankarshan: needinfo-


Attachments (Terms of Use)

Comment 16 amansan 2019-03-04 14:26:10 UTC
Hi

The customer is waiting for an update, do you need any additional information ? 

Regards

Alicia

Comment 18 amansan 2019-03-06 14:18:43 UTC
Heal just finished:

```
sh-4.2# gluster v status vol_7cedfda39e63fbb59c2998233091cb10
Status of volume: vol_7cedfda39e63fbb59c2998233091cb10
Gluster process                             TCP Port  RDMA Port  Online  Pid
------------------------------------------------------------------------------
Brick 10.28.139.73:/var/lib/heketi/mounts/v
g_e48e0dbb3dd65dd250ce3c997effe892/brick_34
934b39c2999e4f55a0c89aecab5718/brick        49161     0          Y       18510
Brick 10.28.139.13:/var/lib/heketi/mounts/v
g_329c82fd567da85c0b62c5dda1c342f3/brick_5d
1fcdf29e210fbf066c910126d43524/brick        49159     0          Y       28716
Brick 10.28.139.166:/var/lib/heketi/mounts/
vg_f2e34e863e3a99b21011b4d8344ec848/brick_4
cd6d96d7b40cd17b59c4950b28cf3bc/brick       49159     0          Y       50806
Brick 10.28.139.73:/var/lib/heketi/mounts/v
g_c7be81950c8c69dbb33fbbea58d5d3a5/brick_78
9c14aa897f8d1a75d0954ff2dd0227/brick        49161     0          Y       18510
Brick 10.28.139.166:/var/lib/heketi/mounts/
vg_042c5a884998313285e3d00c6c59470c/brick_4
65fda0ee5c13e5d78e3279461e1fe99/brick       49159     0          Y       50806
Brick 10.28.139.13:/var/lib/heketi/mounts/v
g_2b36d9ace572cb665ea1cb28ebe45081/brick_f5
7189b014ed23f279d80db687b89ad5/brick        49159     0          Y       28716
Brick 10.28.139.13:/var/lib/heketi/mounts/v
g_2b36d9ace572cb665ea1cb28ebe45081/brick_d8
a46b7c8bd752f2031577ea3be69112/brick        49159     0          Y       28716
Brick 10.28.139.166:/var/lib/heketi/mounts/
vg_b2b1c3052cd87d070a3ae7cd9188b653/brick_7
677d50724618caa89055fdd921bb501/brick       49159     0          Y       50806
Brick 10.28.139.73:/var/lib/heketi/mounts/v
g_d54f514570ec9f63a8d1e7b578b03059/brick_00
cd7128c612491996944fafb80bc8f9/brick        49161     0          Y       18510
Brick 10.28.139.73:/var/lib/heketi/mounts/v
g_c7be81950c8c69dbb33fbbea58d5d3a5/brick_5b
00db39119354483417a28b5d55aaf1/brick        49161     0          Y       18510
Brick 10.28.139.166:/var/lib/heketi/mounts/
vg_042c5a884998313285e3d00c6c59470c/brick_7
076b52b2f6219162e8973b338859c4e/brick       49159     0          Y       50806
Brick 10.28.139.13:/var/lib/heketi/mounts/v
g_2b36d9ace572cb665ea1cb28ebe45081/brick_99
7a09c1ac081ddc7e805ff759b7f29a/brick        49159     0          Y       28716
Brick 10.28.139.166:/var/lib/heketi/mounts/
vg_042c5a884998313285e3d00c6c59470c/brick_9
bd4899a61d0998ef986e5d714a0df4f/brick       49159     0          Y       50806
Brick 10.28.139.73:/var/lib/heketi/mounts/v
g_d54f514570ec9f63a8d1e7b578b03059/brick_b6
dd1cbab3c28cb84a56ce1d65093935/brick        49161     0          Y       18510
Brick 10.28.139.13:/var/lib/heketi/mounts/v
g_6eb35f532caca1f72b88a54475950a7d/brick_92
d7d2aca1a7d8b5c02b94f61c807120/brick        49159     0          Y       28716
Brick 10.28.139.166:/var/lib/heketi/mounts/
vg_b2b1c3052cd87d070a3ae7cd9188b653/brick_c
5a962f47423c1232d31e2c901d5de08/brick       49159     0          Y       50806
Brick 10.28.139.73:/var/lib/heketi/mounts/v
g_d54f514570ec9f63a8d1e7b578b03059/brick_2b
d672a66df2636fa91cc255cf4accf9/brick        49161     0          Y       18510
Brick 10.28.139.13:/var/lib/heketi/mounts/v
g_6eb35f532caca1f72b88a54475950a7d/brick_5f
2f7653e2732cdf542b8ac934e05352/brick        49159     0          Y       28716
Self-heal Daemon on localhost               N/A       N/A        Y       40089
Self-heal Daemon on storage02.appcanvas.net N/A       N/A        Y       97772
Self-heal Daemon on storage01.appcanvas.net N/A       N/A        Y       125437

Task Status of Volume vol_7cedfda39e63fbb59c2998233091cb10
------------------------------------------------------------------------------
There are no active volume tasks

sh-4.2# gluster v heal vol_7cedfda39e63fbb59c2998233091cb10 info
Brick 10.28.139.73:/var/lib/heketi/mounts/vg_e48e0dbb3dd65dd250ce3c997effe892/brick_34934b39c2999e4f55a0c89aecab5718/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.13:/var/lib/heketi/mounts/vg_329c82fd567da85c0b62c5dda1c342f3/brick_5d1fcdf29e210fbf066c910126d43524/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.166:/var/lib/heketi/mounts/vg_f2e34e863e3a99b21011b4d8344ec848/brick_4cd6d96d7b40cd17b59c4950b28cf3bc/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.73:/var/lib/heketi/mounts/vg_c7be81950c8c69dbb33fbbea58d5d3a5/brick_789c14aa897f8d1a75d0954ff2dd0227/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.166:/var/lib/heketi/mounts/vg_042c5a884998313285e3d00c6c59470c/brick_465fda0ee5c13e5d78e3279461e1fe99/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.13:/var/lib/heketi/mounts/vg_2b36d9ace572cb665ea1cb28ebe45081/brick_f57189b014ed23f279d80db687b89ad5/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.13:/var/lib/heketi/mounts/vg_2b36d9ace572cb665ea1cb28ebe45081/brick_d8a46b7c8bd752f2031577ea3be69112/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.166:/var/lib/heketi/mounts/vg_b2b1c3052cd87d070a3ae7cd9188b653/brick_7677d50724618caa89055fdd921bb501/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.73:/var/lib/heketi/mounts/vg_d54f514570ec9f63a8d1e7b578b03059/brick_00cd7128c612491996944fafb80bc8f9/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.73:/var/lib/heketi/mounts/vg_c7be81950c8c69dbb33fbbea58d5d3a5/brick_5b00db39119354483417a28b5d55aaf1/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.166:/var/lib/heketi/mounts/vg_042c5a884998313285e3d00c6c59470c/brick_7076b52b2f6219162e8973b338859c4e/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.13:/var/lib/heketi/mounts/vg_2b36d9ace572cb665ea1cb28ebe45081/brick_997a09c1ac081ddc7e805ff759b7f29a/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.166:/var/lib/heketi/mounts/vg_042c5a884998313285e3d00c6c59470c/brick_9bd4899a61d0998ef986e5d714a0df4f/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.73:/var/lib/heketi/mounts/vg_d54f514570ec9f63a8d1e7b578b03059/brick_b6dd1cbab3c28cb84a56ce1d65093935/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.13:/var/lib/heketi/mounts/vg_6eb35f532caca1f72b88a54475950a7d/brick_92d7d2aca1a7d8b5c02b94f61c807120/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.166:/var/lib/heketi/mounts/vg_b2b1c3052cd87d070a3ae7cd9188b653/brick_c5a962f47423c1232d31e2c901d5de08/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.73:/var/lib/heketi/mounts/vg_d54f514570ec9f63a8d1e7b578b03059/brick_2bd672a66df2636fa91cc255cf4accf9/brick
Status: Connected
Number of entries: 0

Brick 10.28.139.13:/var/lib/heketi/mounts/vg_6eb35f532caca1f72b88a54475950a7d/brick_5f2f7653e2732cdf542b8ac934e05352/brick
Status: Connected
Number of entries: 0

```

It shows correct here now, but only after killing 03 and forcing a heal:
```
[root@storage01 docker_registry]# ls -l docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
total 2
-rw-r--r--. 1 1000030000 2000 1992 Feb  7 12:02 data
[root@storage01 docker_registry]#
```

But not sure if there are any other files that could have been missed, curious if we can force a re-heal/rebalance or have a different way for OCS to double check everything within the volume.

Even though the split-brain for the data dir has gone, still issues as its now an empty file:

```
WMac5159685:~ c09010a$ docker pull docker-registry-default.apps.appcanvas.net/da-saas-images/software-app-saas@sha256:61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
Error response from daemon: manifest for docker-registry-default.apps.appcanvas.net/da-saas-images/software-app-saas@sha256:61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2 not found
WMac5159685:~ c09010a$
```

File pretty much empty via FUSE:
```
[root@storage01 docker_registry]# cd docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
[root@storage01 61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2]# du -shc *
2.0K	data
2.0K	total
[root@storage01 61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2]#
```

storage01:
```
sh-4.2# for file in `gluster v info vol_7cedfda39e63fbb59c2998233091cb10 | grep 10.28.139.13 | awk -F ':' '{print $3}'`; do du -shc $file/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2; done
4.0K	/var/lib/heketi/mounts/vg_329c82fd567da85c0b62c5dda1c342f3/brick_5d1fcdf29e210fbf066c910126d43524/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
4.0K	total
0	/var/lib/heketi/mounts/vg_2b36d9ace572cb665ea1cb28ebe45081/brick_f57189b014ed23f279d80db687b89ad5/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_2b36d9ace572cb665ea1cb28ebe45081/brick_d8a46b7c8bd752f2031577ea3be69112/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_2b36d9ace572cb665ea1cb28ebe45081/brick_997a09c1ac081ddc7e805ff759b7f29a/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_6eb35f532caca1f72b88a54475950a7d/brick_92d7d2aca1a7d8b5c02b94f61c807120/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_6eb35f532caca1f72b88a54475950a7d/brick_5f2f7653e2732cdf542b8ac934e05352/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
sh-4.2#
```

storage02:
```
sh-4.2# for file in `gluster v info vol_7cedfda39e63fbb59c2998233091cb10 | grep 10.28.139.73 | awk -F ':' '{print $3}'`; do du -shc $file/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2; done
4.0K	/var/lib/heketi/mounts/vg_e48e0dbb3dd65dd250ce3c997effe892/brick_34934b39c2999e4f55a0c89aecab5718/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
4.0K	total
0	/var/lib/heketi/mounts/vg_c7be81950c8c69dbb33fbbea58d5d3a5/brick_789c14aa897f8d1a75d0954ff2dd0227/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_d54f514570ec9f63a8d1e7b578b03059/brick_00cd7128c612491996944fafb80bc8f9/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_c7be81950c8c69dbb33fbbea58d5d3a5/brick_5b00db39119354483417a28b5d55aaf1/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_d54f514570ec9f63a8d1e7b578b03059/brick_b6dd1cbab3c28cb84a56ce1d65093935/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_d54f514570ec9f63a8d1e7b578b03059/brick_2bd672a66df2636fa91cc255cf4accf9/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
sh-4.2#
```

storage03:
```
sh-4.2# for file in `gluster v info vol_7cedfda39e63fbb59c2998233091cb10 | grep 10.28.139.166 | awk -F ':' '{print $3}'`; do du -shc $file/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2; done
4.0K	/var/lib/heketi/mounts/vg_f2e34e863e3a99b21011b4d8344ec848/brick_4cd6d96d7b40cd17b59c4950b28cf3bc/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
4.0K	total
0	/var/lib/heketi/mounts/vg_042c5a884998313285e3d00c6c59470c/brick_465fda0ee5c13e5d78e3279461e1fe99/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_b2b1c3052cd87d070a3ae7cd9188b653/brick_7677d50724618caa89055fdd921bb501/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_042c5a884998313285e3d00c6c59470c/brick_7076b52b2f6219162e8973b338859c4e/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_042c5a884998313285e3d00c6c59470c/brick_9bd4899a61d0998ef986e5d714a0df4f/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
0	/var/lib/heketi/mounts/vg_b2b1c3052cd87d070a3ae7cd9188b653/brick_c5a962f47423c1232d31e2c901d5de08/brick/docker/registry/v2/blobs/sha256/61/61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
0	total
sh-4.2#
```

Yesterday however, I could pull the image:
```
$ time docker pull docker-registry-default.apps.appcanvas.net/da-saas-images/software-app-saas:saas-rc-190128-201902071645@sha256:61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
sha256:61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2: Pulling from da-saas-images/software-app-saas
c325120ebc8d: Pull complete
c9d123037991: Pull complete
521a1d24b010: Pull complete
0e82bb9f4fba: Pull complete
aee3da753a34: Pull complete
f8dee6aae8b9: Pull complete
c147fa80b11a: Pull complete
a60ea87d6b32: Pull complete
Digest: sha256:61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2
Status: Downloaded newer image for docker-registry-default.apps.appcanvas.net/da-saas-images/software-app-saas@sha256:61cfb307586f943e554f5d9345c34f0865c0e2dac0f6e0d6f4b8549e67ee47a2

real	5m2.891s
user	0m0.136s
sys	0m0.095s
$

I suppose this happened again after adding storage03, what could we need to have a better idea ?

Comment 26 Dan Yocum 2019-03-07 16:52:32 UTC
Csaba,

What's the state of the patch to resolve the df issue discussed by Sonal in https://bugzilla.redhat.com/show_bug.cgi?id=1684823#c17 ?

I'm just the TAM, not an engineer maintaining the code.

Dan

Comment 27 Amar Tumballi 2019-03-12 11:50:33 UTC
> What's the state of the patch to resolve the df issue discussed by Sonal in https://bugzilla.redhat.com/show_bug.cgi?id=1684823#c17 ?


Dan, this issue is resolved in version RHGS 3.4.1 (Batch Update 1).

Is there anything more we should be looking at in this bug?

Comment 28 Amar Tumballi 2019-05-21 04:55:23 UTC
Closing with WORKSFORME (as per comment#17 etc). Please reopen if that is not the case.


Note You need to log in before you can comment on or make changes to this bug.