Ceph : Issueshttps://tracker.ceph.com/https://tracker.ceph.com/favicon.ico2020-09-16T14:31:07ZCeph
Redmine Orchestrator - Bug #47501 (Resolved): cephadm: Error bootstraping with '--container-init' optionhttps://tracker.ceph.com/issues/475012020-09-16T14:31:07ZRicardo Marquesrimarques@suse.com
<p>When I try to bootstrap a new cluster using the '--container-init' option, I get the following error:</p>
<pre>
['/usr/bin/podman', 'run', '--rm', '--net=host', '--ipc=host', '-e', 'CONTAINER_IMAGE=registry.opensuse.org/filesystems/ceph/master/u
pstream/images/ceph/ceph', '-e', 'NODE_NAME=node1', '-v', '/var/log/ceph/86501be0-f825-11ea-ba9f-525400bedfac:/var/log/ceph:z', '-v',
'/tmp/ceph-tmpp4xdmtqe:/etc/ceph/ceph.client.admin.keyring:z', '-v', '/tmp/ceph-tmph6bq7nup:/etc/ceph/ceph.conf:z', '--entrypoint',
'/usr/bin/ceph', 'registry.opensuse.org/filesystems/ceph/master/upstream/images/ceph/ceph', 'config', 'set', 'mgr', 'mgr/cephadm/cont
ainer_init', True, '--force']
Traceback (most recent call last):
File "/usr/sbin/cephadm", line 5859, in <module>
r = args.func()
File "/usr/sbin/cephadm", line 1248, in _default_image
return func()
File "/usr/sbin/cephadm", line 3032, in command_bootstrap
cli(['config', 'set', 'mgr', 'mgr/cephadm/container_init', args.container_init, '--force'])
File "/usr/sbin/cephadm", line 2831, in cli
).run(timeout=timeout)
File "/usr/sbin/cephadm", line 2479, in run
self.run_cmd(), desc=self.entrypoint, timeout=timeout)
File "/usr/sbin/cephadm", line 907, in call_throws
out, err, ret = call(command, **kwargs)
File "/usr/sbin/cephadm", line 799, in call
logger.debug("Running command: %s" % ' '.join(command))
TypeError: sequence item 22: expected str instance, bool found
</pre> Dashboard - Bug #47494 (Resolved): mgr/dashboard: Dashboard becomes unresponsive when SMART data ...https://tracker.ceph.com/issues/474942020-09-16T09:32:47ZRicardo Marquesrimarques@suse.com
<p>On my environment, 'smartmontools' is not configured, so I see a warning message on "Cluster > Device health" tab, but if I click on the "SMART" tab, the Dashboard UI stops responding and I'm no longer able to click on other tabs:</p>
<p><img src="https://tracker.ceph.com/attachments/download/5141/smart-error.gif" alt="" /></p> Orchestrator - Bug #46922 (Resolved): cephadm: IPv6 syntax inconsistencyhttps://tracker.ceph.com/issues/469222020-08-13T12:39:43ZRicardo Marquesrimarques@suse.com
<p>While trying to bootstrap a cluster with `--mon-ip` and `--apply-spec` options I found that someting IPv6 syntax is `[<IP>]`, and sometimes is`<IP>`.</p>
<p>The following examples show which combinations will work:</p>
<p><strong>1)</strong><br /><pre>
cephadm --verbose bootstrap --mon-ip fde4:8dba:82e1:0:5054:ff:fecd:3d4e --apply-spec /root/bootstrap-spec.yaml
</pre></p>
<p>Fails because we must use `[<IP>]` format on `--mon-ip`:<br /><pre>
INFO:cephadm:/usr/bin/monmaptool:stderr /usr/bin/monmaptool: invalid ip:port '[v2:fde4:8dba:82e1:0:5054:ff:fecd:3d4e:3300,v1:fde4:8dba:82e1:0:5054:ff:fecd:3d4e:6789]'
Traceback (most recent call last):
File "/usr/sbin/cephadm", line 5159, in <module>
r = args.func()
File "/usr/sbin/cephadm", line 1223, in _default_image
return func()
File "/usr/sbin/cephadm", line 2661, in command_bootstrap
tmp_monmap.name: '/tmp/monmap:z',
File "/usr/sbin/cephadm", line 2404, in run
self.run_cmd(), desc=self.entrypoint, timeout=timeout)
File "/usr/sbin/cephadm", line 884, in call_throws
raise RuntimeError('Failed command: %s' % ' '.join(command))
RuntimeError: Failed command: /usr/bin/podman run --rm --net=host --ipc=host -e CONTAINER_IMAGE=docker.io/ceph/ceph:v15 -e NODE_NAME=node1 -v /tmp/ceph-tmpiuac2yke:/tmp/monmap:z --entrypoint /usr/bin/monmaptool docker.io/ceph/ceph:v15 --create --clobber --fsid e308bb2a-dd5f-11ea-88c2-525400ca3d18 --addv node1 [v2:fde4:8dba:82e1:0:5054:ff:fecd:3d4e:3300,v1:fde4:8dba:82e1:0:5054:ff:fecd:3d4e:6789] /tmp/monmap
</pre></p>
<p><strong>2)</strong><br /><pre>
# cat /root/bootstrap-spec.yaml
service_type: mgr
service_name: mgr
placement:
hosts:
- 'node1'
---
service_type: mon
service_name: mon
placement:
hosts:
- 'node1:[fde4:8dba:82e1:0:5054:ff:fecd:3d4e]'
# cephadm --verbose bootstrap --mon-ip [fde4:8dba:82e1:0:5054:ff:fecd:3d4e] --apply-spec /root/bootstrap-spec.yaml
</pre></p>
<p>Fails because `service_spec` does not suppor `[<IP>]` format:<br /><pre>
DEBUG:cephadm:Running command: /usr/bin/podman run --rm --net=host --ipc=host -e CONTAINER_IMAGE=registry.suse.de/devel/storage/7.0/containers/ses/7/ceph/ceph -e NODE_NAME=node1 -v /var/log/ceph/3cebcb98-dd4f-11ea-9787-525400ca3d18:/var/log/ceph:z -v /tmp/ceph-tmp9bfj1zkg:/etc/ceph/ceph.client.admin.keyring:z -v /tmp/ceph-tmpg0xv6fke:/etc/ceph/ceph.conf:z -v /tmp/bootstrap-spec.yaml:/tmp/spec.yml:z --entrypoint /usr/bin/ceph registry.suse.de/devel/storage/7.0/containers/ses/7/ceph/ceph orch apply -i /tmp/spec.yml
DEBUG:cephadm:/usr/bin/ceph:stderr Error EINVAL: Traceback (most recent call last):
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/mgr_module.py", line 1167, in _handle_command
DEBUG:cephadm:/usr/bin/ceph:stderr return self.handle_command(inbuf, cmd)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/orchestrator/_interface.py", line 138, in handle_command
DEBUG:cephadm:/usr/bin/ceph:stderr return dispatch[cmd['prefix']].call(self, cmd, inbuf)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/mgr_module.py", line 311, in call
DEBUG:cephadm:/usr/bin/ceph:stderr return self.func(mgr, **kwargs)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/orchestrator/_interface.py", line 100, in <lambda>
DEBUG:cephadm:/usr/bin/ceph:stderr wrapper_copy = lambda *l_args, **l_kwargs: wrapper(*l_args, **l_kwargs)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/orchestrator/_interface.py", line 89, in wrapper
DEBUG:cephadm:/usr/bin/ceph:stderr return func(*args, **kwargs)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/orchestrator/module.py", line 1162, in _apply_misc
DEBUG:cephadm:/usr/bin/ceph:stderr spec = json_to_generic_spec(s)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/share/ceph/mgr/orchestrator/_interface.py", line 1217, in json_to_generic_spec
DEBUG:cephadm:/usr/bin/ceph:stderr return ServiceSpec.from_json(spec)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 42, in inner
DEBUG:cephadm:/usr/bin/ceph:profile rt=0.9018697738647461, stop=False, exit=None, reads=[12]
DEBUG:cephadm:/usr/bin/ceph:stderr return method(cls, *args, **kwargs)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 488, in from_json
DEBUG:cephadm:/usr/bin/ceph:stderr return _cls._from_json_impl(c) # type: ignore
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 495, in _from_json_impl
DEBUG:cephadm:/usr/bin/ceph:stderr v = PlacementSpec.from_json(v)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 42, in inner
DEBUG:cephadm:/usr/bin/ceph:stderr return method(cls, *args, **kwargs)
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 245, in from_json
DEBUG:cephadm:/usr/bin/ceph:stderr isinstance(host, str) else
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 134, in parse
DEBUG:cephadm:/usr/bin/ceph:stderr raise e
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib/python3.6/site-packages/ceph/deployment/service_spec.py", line 131, in parse
DEBUG:cephadm:/usr/bin/ceph:stderr ip_address(six.text_type(network))
DEBUG:cephadm:/usr/bin/ceph:stderr File "/usr/lib64/python3.6/ipaddress.py", line 54, in ip_address
DEBUG:cephadm:/usr/bin/ceph:stderr address)
DEBUG:cephadm:/usr/bin/ceph:stderr ValueError: '[fde4:8dba:82e1:0:5054:ff:fecd:3d4e]' does not appear to be an IPv4 or IPv6 address
</pre></p>
<p><strong>3)</strong><br />The combination that works is:<br /><pre>
# cat /root/bootstrap-spec.yaml
service_type: mgr
service_name: mgr
placement:
hosts:
- 'node1'
---
service_type: mon
service_name: mon
placement:
hosts:
- 'node1:fde4:8dba:82e1:0:5054:ff:fecd:3d4e'
# cephadm --verbose bootstrap --mon-ip [fde4:8dba:82e1:0:5054:ff:fecd:3d4e] --apply-spec /root/bootstrap-spec.yaml
</pre></p>
<hr />
<p>Should we be consistent on the supported/required IPv6 syntax? If yes, which should be the correct one? Or should we support both?</p> Dashboard - Bug #46818 (Resolved): mgr/dashboard: Unable to edit iSCSI logged-in clienthttps://tracker.ceph.com/issues/468182020-08-03T15:35:11ZRicardo Marquesrimarques@suse.com
<p>Using "gwcli" tool, it's possible to do the following actions on a logged-in client:</p>
<pre><code>- add/remove disks</code></pre>
<pre><code>- set/remove/change auth</code></pre>
<p>For consistency, the same operations should be supported by Ceph Dashboard.</p> Orchestrator - Bug #46777 (Resolved): cephadm: Error bootstraping a cluster with '--registry-json...https://tracker.ceph.com/issues/467772020-07-30T12:33:26ZRicardo Marquesrimarques@suse.com
<p>When I try to bootstrap a cluster with the new '--registry-json' option, I get the following error:</p>
<pre>
INFO:cephadm:Non-zero exit code 22 from /usr/bin/podman run --rm --net=host --ipc=host -e CONTAINER_IMAGE=registry.opensuse.org/filesystems/ceph/octopus/upstream/images/ceph/ceph -e NODE_NAME=node1 -v /var/log/ceph/88cd09d8-d254-11ea-b540-5254009ce6f4:/var/log/ceph:z -v /tmp/ceph-tmp8q3k7c2q:/etc/ceph/ceph.client.admin.keyring:z -v /tmp/ceph-tmpxa4ehfb0:/etc/ceph/ceph.conf:z --entrypoint /usr/bin/ceph registry.opensuse.org/filesystems/ceph/octopus/upstream/images/ceph/ceph config set mgr mgr/cephadm/registry_url 192.168.1.102:5000
INFO:cephadm:/usr/bin/ceph:stderr Error EINVAL: unrecognized config option 'mgr/cephadm/registry_url'
</pre>
<p>We may need to use the '--force' option, example: <a class="external" href="https://github.com/ceph/ceph/blob/83f068d04322fdfbfea7dc0481474bbec687a4c6/src/vstart.sh#L917">https://github.com/ceph/ceph/blob/83f068d04322fdfbfea7dc0481474bbec687a4c6/src/vstart.sh#L917</a></p> Dashboard - Backport #46436 (Resolved): octopus: mgr/dashboard: Unable to edit iSCSI target which...https://tracker.ceph.com/issues/464362020-07-09T14:57:39ZRicardo Marquesrimarques@suse.com
<p><a class="external" href="https://github.com/ceph/ceph/pull/35997">https://github.com/ceph/ceph/pull/35997</a></p> Dashboard - Backport #46435 (Resolved): nautilus: mgr/dashboard: Unable to edit iSCSI target whic...https://tracker.ceph.com/issues/464352020-07-09T14:57:32ZRicardo Marquesrimarques@suse.com
<p><a class="external" href="https://github.com/ceph/ceph/pull/35998">https://github.com/ceph/ceph/pull/35998</a></p> Dashboard - Bug #46383 (Resolved): mgr/dashboard: Unable to edit iSCSI target which has active se...https://tracker.ceph.com/issues/463832020-07-07T10:43:37ZRicardo Marquesrimarques@suse.com
<p>iSCSI target "Edit" button is disabled when that target has an active session, but it's possible to edit it using "gwcli".</p>
<p>Note that Ceph dashboard should support same features that are supported by "gwcli".</p> Orchestrator - Bug #46233 (Resolved): cephadm: Add "--format" option to "ceph orch status"https://tracker.ceph.com/issues/462332020-06-26T15:08:06ZRicardo Marquesrimarques@suse.com
<p>ATM it's not possible to specify the output format for "ceph orch status":</p>
<pre>
ceph orch status --format=json
</pre> Dashboard - Bug #45810 (New): mgr/dashboard: Expand/collapse OSD row changes row selectionhttps://tracker.ceph.com/issues/458102020-06-02T09:36:04ZRicardo Marquesrimarques@suse.com
<p>Expand/collapse an OSD row changes row selection:</p>
<p><img src="https://tracker.ceph.com/attachments/download/4898/2020-06-02_10-32.gif" alt="" /></p> Dashboard - Bug #45159 (New): mgr/dashboard: Errors on OSD page during OSD creationhttps://tracker.ceph.com/issues/451592020-04-21T13:28:46ZRicardo Marquesrimarques@suse.com
<p>While creating OSDs using ceph orchestrator:</p>
<pre>
node1:~ # cat cluster.yaml
service_type: osd
placement:
host_pattern: 'node*'
data_devices:
all: true
node1:~ # ceph orch apply -i cluster.yaml
</pre>
<p>I see errors on Dashboard when selecting an OSD that is being created:</p>
<p><img src="https://tracker.ceph.com/attachments/download/4843/2020-04-21_09-37-01.png" alt="" /></p> Dashboard - Feature #43260 (Pending Backport): mgr/dashboard: Support creating a placement rule t...https://tracker.ceph.com/issues/432602019-12-11T16:04:12ZRicardo Marquesrimarques@suse.com
<p>It's already possible to select an existing "Crush ruleset" in pool creation form, but we should also be able to create, rename and delete "Crush ruleset"s.</p> Dashboard - Bug #42937 (In Progress): mgr/dashboard: iSCSI menu entry should indicate how many ga...https://tracker.ceph.com/issues/429372019-11-21T17:34:22ZRicardo Marquesrimarques@suse.com
<p>If one or more gateways are down, iSCSI menu entry should be "red" and show the number of gateways that are "down".</p>
<p>Something similar to what we have in RBD mirroring:</p>
<p><img src="https://tracker.ceph.com/attachments/download/4600/mirroring_error.png" alt="" /></p> Dashboard - Bug #39667 (New): mgr/dashboard: Optimize RBD list request on iSCSI formhttps://tracker.ceph.com/issues/396672019-05-10T09:22:08ZRicardo Marquesrimarques@suse.com
<p>iSCSI form is taking too long to open due to a request to the RBD images list that is then used to populate the "Images" dropdown:</p>
<p><img src="https://tracker.ceph.com/attachments/download/4169/rbd-request-on-iscsi-form.png" alt="" /></p>
<p>This request is returning a lot of information that is not needed in this form (e.g. disk usage):</p>
<p><img src="https://tracker.ceph.com/attachments/download/4170/iscsi-rbd-response.png" alt="" /></p>
<p>We should only request the required RBD image fields in order to optimize the initial load of this form.</p> Dashboard - Cleanup #25137 (New): mgr/dashboard: Security permission names should be consistent w...https://tracker.ceph.com/issues/251372018-07-27T12:59:06ZRicardo Marquesrimarques@suse.com
<p><strong>(Waiting for confirmation)</strong></p>
<p><ins>Whenever possible</ins>, security permission names should be consistent with the corresponding UI elements.</p>
<p>The following permissions should be renamed:</p>
<ul>
<li>rbd-image -> block-image</li>
<li>rbd-mirroring -> block-mirroring</li>
<li>rgw -> object-gateway</li>
</ul>