Closed rjfd closed 5 years ago
looks good(minus the linter issues), could we get this interweaved in the CI somehow? @smithfarm
@jschmid1 lint issues fixed.
Having this info in the CI logs is a great idea.
@jschmid1 @rjfd Here is the ganesha smoke test: https://github.com/SUSE/ceph/blob/ses6-downstream-commits/qa/tasks/scripts/ganesha_smoke_test.sh
Please provide details (what to run, when to run it), preferably in the form of a PR :-)
You could also provide the information in the form of a CI log which is edited to show which command should be run where.
Please provide details (what to run, when to run it), preferably in the form of a PR :-)
@smithfarm This report could be run after stage 4 is finished, and also in the middle of the execution of the ganesha_smoke_test.
I can open a PR adding this to the ganesha_smoke_test.sh (for running after stage 4, I don't know where to change), but I have one question, can we run a salt-run
command in that script?
@rjfd Running a salt-run
command (i.e. on the master) after Stage 4 is no problem. Running it in the middle of ganesha_smoke_test.sh
is somewhat more difficult, because that script is not guaranteed to run on the master node - see https://github.com/SUSE/ceph/blob/ses6-downstream-commits/qa/tasks/deepsea.py#L1867-L1888
There are at least two ways to approach this: either (1) split ganesha_smoke_test.sh
into two parts, or (2) run the command from the script via ssh
.
ad 1:
In this scenario, the existing ganesha_smoke_test.sh
would be split into two scripts - call them "before" and "after" - and the salt-run
command would be triggered on the master after "before" completes, but before "after" begins.
ad 2:
Alternatively, we could pass the hostname of the master node as an argument to the script, and have the script do something like ssh $MASTER_NODE sudo salt-run . . .
CI integration PR ready for review: https://github.com/SUSE/ceph/pull/331
What are the next steps here @smithfarm @rjfd ?
@jschmid1 @rjfd https://github.com/SUSE/ceph/pull/331 has been merged and cherry-picked to ses6-deepsea-pr
, so I'll trigger the nfs-ganesha test and see what happens.
@susebot run teuthology deepsea/tier2/nfs-ganesha
Commit 45ee069b41a102e1e0935cdaaf4f0e9172957582 is OK for suite deepsea/tier2/nfs-ganesha. Check tests results in the Jenkins job: http://ci.ses.suse.de:8080/job/pr-deepsea/288/
Ganesha report can be seen in all its glory.
if you'd like to see this feature in SES6 please create a backport for this @rjfd
This PR adds a function to the ganesha runner to output the information about the current nfs-ganesha gateways and their configuration.
Signed-off-by: Ricardo Dias rdias@suse.com