Skip to content

doc: New section in device management#35124

Merged
sebastian-philipp merged 1 commit intoceph:masterfrom
jmolmo:wip-doc-update-devices
Jun 2, 2020
Merged

doc: New section in device management#35124
sebastian-philipp merged 1 commit intoceph:masterfrom
jmolmo:wip-doc-update-devices

Conversation

@jmolmo
Copy link
Member

@jmolmo jmolmo commented May 19, 2020

A new section to describe the command used to blink lights in physical devices

Signed-off-by: Juan Miguel Olmo Martínez jolmomar@redhat.com

-->

Checklist

  • References tracker ticket
  • Updates documentation if necessary
  • Includes tests for new functionality or reproducer for bug

jenkins test docs
jenkins render docs

@sebastian-philipp
Copy link
Contributor

hm. this is a bit misleading, as this depends on (and is actually par of) the orchestrator.

I'm torn. On one hand, this is not part of ceph orch, but on the other hand, we really need to make people aware that this this depends on the orchestrator (i.e. either cephadm or rook)


ceph device ls

The second parameter sets the kind of light to blink. By default the
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
The second parameter sets the kind of light to blink. By default the
The second parameter of ``device light on|off`` command sets the kind of light to blink. By default the

as otherwise, reader (me, for instance) might want to search backwards for the command and the "second parameter". and the first match would be "ceph device ls".

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Changed. @tchaikov, what do you think now?

Comment on lines +31 to +32
You can blink lights on hardware enclosures to make the replacement of failed
disks easy and less error-prone. Use the following command::
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
You can blink lights on hardware enclosures to make the replacement of failed
disks easy and less error-prone. Use the following command::
You can blink the drive LEDs on hardware enclosures to identify the failed disk drive prior to replacement. Use the following command::

@zdover23 what do you think?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

changed

@ceph ceph deleted a comment from rzarzynski May 20, 2020
@jmolmo jmolmo force-pushed the wip-doc-update-devices branch from 81e4a78 to 6c63f29 Compare May 20, 2020 09:13
@jmolmo
Copy link
Member Author

jmolmo commented May 20, 2020

hm. this is a bit misleading, as this depends on (and is actually par of) the orchestrator.

I'm torn. On one hand, this is not part of ceph orch, but on the other hand, we really need to make people aware that this this depends on the orchestrator (i.e. either cephadm or rook)

@sebastian-philipp : We have two options:

a) Leave this functionality "inside" the "ceph device" commands set. This is my preferred , because I think that it is more easy for the user to have all the things around devices in the "same place". (despite we have this dependency)... after all, we will finish installing all the clusters using cephadm or rook...and therefore we will have the orch module enabled by default.

b) Move this command to the orchestrator command set:
ceph orch device light ....
In this case, i think that the user needs to know that this functionality is inside the orchestrator command set. I think seems less natural... but we avoid the dependency

I have added a note to outline this dependency, i think probably it is enough.

@jmolmo jmolmo requested a review from tchaikov May 20, 2020 09:28
@jmolmo
Copy link
Member Author

jmolmo commented May 25, 2020

@tchaikov are you ok with the changes?

Copy link
Contributor

@tchaikov tchaikov left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@jmolmo LGTM. just two nits. sorry for the latency. just saw your comment.


device light on|off <devid> [ident|fault] [--force]

The <devid> parameter is the device identification, and you can obtain this
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
The <devid> parameter is the device identification, and you can obtain this
The ``<devid>`` parameter is the device identification, and you can obtain this

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

thx! done.

@jmolmo jmolmo force-pushed the wip-doc-update-devices branch from 6c63f29 to 72e92f4 Compare May 27, 2020 08:57
@sebastian-philipp sebastian-philipp requested a review from asettle May 27, 2020 10:52
@tchaikov
Copy link
Contributor

/home/jenkins-build/build/workspace/ceph-pull-requests/src/test/smoke.sh:56: TEST_multimon:  timeout 8 rados -p foo bench 4 write -b 4096 --no-cleanup
hints = 1
Maintaining 16 concurrent writes of 4096 bytes to objects of size 4096 for up to 4 seconds or 0 objects
Object prefix: benchmark_data_slave-ubuntu11_11219
  sec Cur ops   started  finished  avg MB/s  cur MB/s last lat(s)  avg lat(s)
    0      16        16         0         0         0           -           0
    1      16        16         0         0         0           -           0
    2      16        16         0         0         0           -           0
    3      16        33        17 0.0220724 0.0221354    0.521384     2.28468
    4      16        33        17 0.0165655         0           -     2.28468
    5      16        33        17 0.0132578         0           -     2.28468
Total time run:         5.88443
Total writes made:      33
Write size:             4096
Object size:            4096
Bandwidth (MB/sec):     0.0219063
Stddev Bandwidth:       0.00989926
Max bandwidth (MB/sec): 0.0221354
Min bandwidth (MB/sec): 0
Average IOPS:           5
Stddev IOPS:            2.23607
Max IOPS:               5
Min IOPS:               0
Average Latency(s):     2.81336
Stddev Latency(s):      0.644136
Max latency(s):         3.40527
Min latency(s):         0.521384
/home/jenkins-build/build/workspace/ceph-pull-requests/src/test/smoke.sh:56: TEST_multimon:  return 1

@tchaikov
Copy link
Contributor

jenkins test make check

A new section to describe the command used to blink lights in physical devices

Signed-off-by: Juan Miguel Olmo Martínez <jolmomar@redhat.com>
@jmolmo jmolmo force-pushed the wip-doc-update-devices branch from 72e92f4 to e025727 Compare May 27, 2020 13:42
@tchaikov
Copy link
Contributor

[ RUN      ] TestDeepCopy.Snaps_LargerDstObjSize
creating image with stripe unit: 2097152, stripe count: 16
/home/jenkins-build/build/workspace/ceph-pull-requests/src/test/librbd/test_DeepCopy.cc:117: Failure
Expected equality of these values:
  read_size
    Which is: 4194304
  api::Io<>::read( *m_dst_ictx, offset, read_size, librbd::io::ReadResult{dst_result}, 0)
    Which is: 2097152
[  FAILED  ] TestDeepCopy.Snaps_LargerDstObjSize (417 ms)

@dillaman seems it's a new test failure. i filed a tracker ticket at https://tracker.ceph.com/issues/45731

@tchaikov
Copy link
Contributor

tchaikov commented Jun 2, 2020

jenkins test make check

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants