Nathan Cutler

DeepSea packaging stores (and possibly clobbers) files in /srv (not allowed in openSUSE)

4

First, this issue refers to the discussion at https://build.opensuse.org/request/show/503528 where the openSUSE maintainers explain why the DeepSea spec file, in its current form, cannot be accepted. One of the issues...

Cannot deploy RGW-over-SSL (only)

1

Using the tip of the "master" branch, I am trying to deploy RGW with a single HTTPS endpoint. The documentation says to do something like this, which **almost** works (I'm...

### Description of Issue/Question This failure just started happening on the tip of DeepSea master branch. It does not appear to happen on 0.9.14, which would appear to implicate one...

### Description of Issue/Question We have a transient CI failure. It looks like this: ``` 2019-02-28T14:18:42.269 INFO:teuthology.orchestra.run.target192168000033:Running: "sudo bash -c 'DEV_ENV=true timeout 60m deepsea --log-file=/var/log/salt/deepsea.log --log-level=debug salt-run state.orch ceph.stage.3 --simple-output'"...

### Description of Issue/Question When I deploy a cluster with the following alternative defaults in the pillar: ``` tuned_mgr_init: default-off tuned_mon_init: default-off tuned_osd_init: default-off ``` my expectation is that nothing...

Stage 3 sometimes hangs in the CI

43

### Description of Issue/Question In the master branch CI, we are seeing frequent, but transient, Stage 3 hangs/timeouts. They do not seem to be associated with any particular test. ```...

CLI reports wrong number of steps for Stage 3

1

4-node cluster like so: ``` roles: - - client.salt_master - mon.a - mgr.x - osd.0 - - mon.b - mgr.y - osd.1 - - mon.c - mgr.z - osd.2 -...

ceph.restart orchestration could check if all needed roles are deployed and show improved error message if any roles are missing

5

DeepSea master branch, tip is 12965311c0c6f4ad69e5854f498907df7f9d1cea On a single-node SLE15/SES6 cluster with 4 external drives, I run Stages 0-3, get HEALTH_OK. Then I do `salt-run state.orch ceph.smoketests` and this fails...

A user reports: during the execution of stage 0 the “/var/log/deepsea.log” shows several occurrences of errors when calling “/usr/bin/zypper”: ``` 2017-10-05 07:49:38,185 [INFO] deepsea.monitor: Start stage: ceph.stage.0 jid=20171005074921707802 2017-10-05 07:49:38,187...

Need to adjust rgw_frontends syntax on upgrade

5

From jewel/SES4 to luminous/SES5 there was a change in the rgw_frontends syntax when specifying multiple rgw ports. In jewel it was > rgw_frontends = "civetweb port=80, civetweb port=443s ssl_certificate=/etc/ceph/rgw.pem" whereas...