thecodeteam / puppet-vnx Goto Github PK

We are making a change for Puppet 6.0.0 that affects this module (found by scanning all modules on the forge). See ticket https://tickets.puppetlabs.com/browse/PUP-7141 - where the change is described.

We found that In this location:

swindlife-emc_vnx-0.1.3/lib/puppet/type/vnx_storagesystemcache.rb
11:      value.gsub!(/\s+/, "")

There is a gsub! that mutates the given input. That will no longer work in Puppet 6.0.0 as there is a high probability that the input will be a frozen string.
Please see https://tickets.puppetlabs.com/browse/PUP-9098 for the corresponding ticket for some of the puppet maintained modules.

It would be great if puppet-vnx could be updated to make a dup of the string before mutating it (or be otherwise changed so that it does not mutate the original value).

I applied the attached manifest and created a storagepool and a couple of LUNs. This was entirely successful. I then re-applied the same manifest after a time. The re-apply failed attempting to execute navisphere commands a second time. In short, idempotency is broken for vnx_storagepool.

'''
puppet apply vnx.pp --verbose
Notice: Compiled catalog for lp01.vlab.local in environment production in 0.17 seconds
Info: Applying configuration version '1449629787'
Notice: /Stage[main]/Main/Vnx_storagepool[puppet-pool]/disks: disks changed [] to '0_0_9 0_0_10 0_0_11 0_0_12 0_0_13 1_1
_9 1_1_10 1_1_11 1_1_12 1_1_13'
Notice: /Stage[main]/Main/Vnx_storagepool[puppet-pool]/cancel_expand: defined 'cancel_expand' as 'false'
Error: /Stage[main]/Main/Vnx_storagepool[puppet-pool]: Could not evaluate: Command error, output:
The best practices check for the storagepool's expansion did not finish successfully. The storagepool will not be expand
ed
Rule Messages
Could not use the disk : 0_0_9. The disk is already part of a Storage Pool.
Could not use the disk : 0_0_10. The disk is already part of a Storage Pool.
Could not use the disk : 0_0_11. The disk is already part of a Storage Pool.
Could not use the disk : 0_0_12. The disk is already part of a Storage Pool.
Could not use the disk : 0_0_13. The disk is already part of a Storage Pool.
Could not use the disk : 1_1_9. The disk is already part of a Storage Pool.
Could not use the disk : 1_1_10. The disk is already part of a Storage Pool.
Could not use the disk : 1_1_11. The disk is already part of a Storage Pool.
Could not use the disk : 1_1_12. The disk is already part of a Storage Pool.
Could not use the disk : 1_1_13. The disk is already part of a Storage Pool.

Notice: Finished catalog run in 24.94 seconds
'''

we have tested in two versions, PE3.8 and Puppet4.2
when using storagepool, we hit this error.

Error: Parameter disks failed on Vnx_storagepool[wangzz_pool]: Validate method failed for class disks: undefined method `all?' for "1_0_7":String at /home/wangzz/modules/wzz_test/manifests/init.pp:20

Looked into lib/puppet/type/vnx_storagepool.rb
edit "validate do |value|" to “validate do |*value|” can make it work properly.