[skip ci] Nimbus testbed for nfs volume nightly test#7853
[skip ci] Nimbus testbed for nfs volume nightly test#7853rajanashok merged 1 commit intovmware:masterfrom
Conversation
rajanashok
requested review from
matthewavery,
mhagen-vmware and
rogeliosanchez
Codecov Report
@@ Coverage Diff @@
## master #7853 +/- ##
==========================================
- Coverage 31.08% 26.52% -4.56%
==========================================
Files 295 36 -259
Lines 43723 5183 -38540
==========================================
- Hits 13590 1375 -12215
+ Misses 28951 3701 -25250
+ Partials 1182 107 -1075Continue to review full report at Codecov.
|
rajanashok
changed the title
| Close Connection | ||
|
|
||
| ${nfs} ${nfs_ip}= Deploy Nimbus NFS Datastore %{NIMBUS_USER} %{NIMBUS_PASSWORD} additional-args=--nimbus ${POD} | ||
| ${nfs} ${nfs_ro} ${esx1} ${nfs_ip} ${nfs_ro_ip} ${esx1_ip}= Deploy Nimbus NFS Datastore %{NIMBUS_USER} %{NIMBUS_PASSWORD} additional-args=--testbedSpecRubyFile /dbc/w3-dbc302/rashok/vic-nfs.rb --esxBuild ${ESX_VERSION} --esxPxeDir ${ESX_VERSION} --plugin testng |
There was a problem hiding this comment.
just make sure that path is accessible by the nightly user account
There was a problem hiding this comment.
Verified it works on svc.vicuser
| @@ -0,0 +1,58 @@ | |||
| #This file is needed because there are 3 functions in it which are used in this testbed. | |||
There was a problem hiding this comment.
I think this is wholy implied by the fact that you have required it, so not really sure the purpose of this comment
|
|
||
| testbed = { | ||
| 'version' => 3, | ||
| 'name' => "vsan-1esx-allflash-#{sharedStorageStyle}-#{esxStyle}", |
There was a problem hiding this comment.
it is not allflash, it apparently has one HDD that is 250GB in size?
| testbed = { | ||
| 'version' => 3, | ||
| 'name' => "vsan-1esx-allflash-#{sharedStorageStyle}-#{esxStyle}", | ||
| 'esx' => (0..0).map do |i| |
There was a problem hiding this comment.
this is a single iteration loop, really not needed
There was a problem hiding this comment.
Yeah the script had multiple ESXs to start with, will get rid of the loop.
| 'name' => 'dev-nfs', | ||
| 'type' => 'NFS', | ||
| 'nfsOpt' => 'ro', | ||
| 'numMem' => 64 * 1024, |
There was a problem hiding this comment.
I like the memory increase, might be overkill but is there any way to specify SSD for these? and we should specify a good sized disk capacity in addition to the mem update
There was a problem hiding this comment.
Sure, may be a 8 GB memory practical one? also, looking into adding a disk, as the local-0 is running out of space during my test runs.
There was a problem hiding this comment.
I think 16 is probably a happy medium
| 'ssds' => [ 200 * oneGB, 250 * oneGB ], | ||
| 'nics' => 2, | ||
| 'staf' => false, | ||
| 'desiredPassword' => 'ca$hc0w', |
There was a problem hiding this comment.
I am updating the password to e2eFunctionalTest after the ESX is created. https://github.com/rajanashok/vic/blob/15813ff2ec286414a36201f381079212ca9b55bc/tests/resources/Nimbus-Util.robot#L445
| 'numCPUs' => 6, | ||
| 'disks' => [ 250 * oneGB ], | ||
| 'disableNfsMounts' => true, | ||
| 'ssds' => [ 200 * oneGB, 250 * oneGB ], |
There was a problem hiding this comment.
two of these are really not needed I don't think.
| 'esxcli system settings kernel set -s skipPartitioningSsds -v "TRUE"', \ | ||
| 'services.sh restart', \ | ||
| 'esxcli vsan storage tag add -d mpx.vmhba1:C0:T1:L0 -t capacityFlash', \ | ||
| 'esxcli storage core adapter rescan --all', \ |
There was a problem hiding this comment.
we don't need vsan here
rajanashok
force-pushed
the
feature/nfs-testbed
branch
from
15813ff to
d9cf275
Compare
|
@mhagen-vmware Updated PR with the review comments. |
| 'numMem' => 16 * 1024, | ||
| }], | ||
| } | ||
| testbed |
| 'staf' => false, | ||
| 'desiredPassword' => 'e2eFunctionalTest', | ||
| }], | ||
| 'vsan' => true, |
tests/resources/Nimbus-Util.robot
Outdated
| \ ${status}= Run Keyword And Return Status Should Contain ${line} ${name}.esx.0' is up. IP: | ||
| \ ${ip}= Run Keyword If ${status} Fetch From Right ${line} ${SPACE} | ||
| \ Run Keyword If ${status} Set Suite Variable ${esx-ip} ${ip} | ||
| \ Exit For Loop If ${status} |
There was a problem hiding this comment.
these loops should be replaced with the 'Get IP' logic we are using in newer test cases like this:
https://github.com/vmware/vic/blob/master/tests/manual-test-cases/Group5-Functional-Tests/5-27-Selenium-Grid.robot#L33
There was a problem hiding this comment.
Yup thanks, these are pretty handy! Updated the PR.
tests/resources/Nimbus-Util.robot
Outdated
| @@ -405,25 +405,39 @@ Get Vsphere Version | |||
| Deploy Nimbus NFS Datastore | |||
There was a problem hiding this comment.
Deploy Simple NFS Testbed
rajanashok
force-pushed
the
feature/nfs-testbed
branch
from
d9cf275 to
5ec5df2
Compare
mhagen-vmware
left a comment
mhagen-vmware
left a comment
There was a problem hiding this comment.
lgtm, just a small consolidation left
| Open Connection %{NIMBUS_GW} | ||
| Wait Until Keyword Succeeds 10 min 30 sec Login %{NIMBUS_USER} %{NIMBUS_PASSWORD} | ||
| ${esx-ip}= Get IP ${name}.esx.0 | ||
| Close Connection |
There was a problem hiding this comment.
just combine these three into a single open connection/close connection
rajanashok
force-pushed
the
feature/nfs-testbed
branch
from
5ec5df2 to
bb96a8d
Compare
rajanashok
force-pushed
the
feature/nfs-testbed
branch
from
bb96a8d to
6d1d8ea
Compare
Updated the NFS Volume nightlies to build a full NFS test bed instead of relying on the unsupported pod placement calls. Fixes vmware#7642 (cherry picked from commit ff6baa9)
Updated the NFS Volume nightlies to build a full NFS test bed instead of relying on the unsupported pod placement calls. Fixes vmware#7642 (cherry picked from commit ff6baa9)
5 participants
Updated the NFS Volume nightlies to build a full NFS test bed instead of relying on the unsupported pod placement calls.
Fixes #7642