Change in vdsm[master]: gluster: fix build error in F20 due to blivet version issue

tjeyasin at redhat.com tjeyasin at redhat.com
Wed Mar 25 20:11:53 UTC 2015


Timothy Asir has posted comments on this change.

Change subject: gluster: fix build error in F20 due to blivet version issue
......................................................................


Patch Set 10:

(2 comments)

https://gerrit.ovirt.org/#/c/38942/10/vdsm/gluster/storagedev.py
File vdsm/gluster/storagedev.py:

Line 250: 
Line 251:     stripeSize = raidParams.get('stripeSize', 0) / 1024.0
Line 252:     # bz#1198568: The latest version of blivet does not accepts float
Line 253:     # value for peSize in LVMVolumeGroupDevice function.
Line 254:     # It will be enhanced once the support is available in blivet.
> 1. I am not able to understand this bz.  Below you pass stripeSize in float
Below i am passing Size object (which may contain any float value) to _createVG if its support size object otherwise i am passing float value.
Because latest blivet version (in F21) support size obj (may contain any float) and the older version (in F20) does not support size obj and it will work with numerical value (in MiB) only.

The latest blivet version (in F21) should support float value also in addition to size obj to provide backward compatibility but it does not support any float value for peSize.

f20
~~
* support any numerical value

f21
~~
* support any numerical value in the size obj
* does not support float value
* does support int value

Note about the bug:
~~~~~~~~~~~~
The bug 1198568 created before we identify this issue. Actually the function LVMVolumeGroupDevice does not work if the size obj contain any float value before. But that part is fixed and its available in the latest version.
However working with a float value for the given peSize (not a size obj)) part is not yet fixed. Once that part is also fixed the bug status will be moved to next state. Hope you are clear now.
Line 255:     if stripeSize and supportSizeObj:
Line 256:         stripeSize = size.Size("%s MiB" % stripeSize)
Line 257:     vg = _createVG(vgName, pvDeviceList, stripeSize)
Line 258: 


Line 271:         metaDataSize = (metaDataSize - (metaDataSize % alignment)) / 1024.0
Line 272:         poolDataSize = (poolDataSize - (poolDataSize % alignment)) / 1024.0
Line 273:         # bz#1205553: The latest version of blivet does not accepts float
Line 274:         # value for size parm in LVMLogicalVolumeDevice function.
Line 275:         # This part will be enhanced once the support is available in blivet.
> Same as above
Same as above. There it is peSize param inLVMVolumeGroupDevice function but here it is size param in LVMLogicalVolumeDevice function.
Line 276:         if supportSizeObj:
Line 277:             metaDataSize = size.Size("%s MiB" % metaDataSize)
Line 278:             poolDataSize = size.Size("%s MiB" % poolDataSize)
Line 279:     pool = _createThinPool(poolName, vg, chunkSize, metaDataSize, poolDataSize)


-- 
To view, visit https://gerrit.ovirt.org/38942
To unsubscribe, visit https://gerrit.ovirt.org/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Ibf2af69637b7d247225433a7736c49f6c5fefe1d
Gerrit-PatchSet: 10
Gerrit-Project: vdsm
Gerrit-Branch: master
Gerrit-Owner: Timothy Asir <tjeyasin at redhat.com>
Gerrit-Reviewer: Bala.FA <barumuga at redhat.com>
Gerrit-Reviewer: Dan Kenigsberg <danken at redhat.com>
Gerrit-Reviewer: Eyal Edri <eedri at redhat.com>
Gerrit-Reviewer: Federico Simoncelli <fsimonce at redhat.com>
Gerrit-Reviewer: Max Kovgan <mvk at redhat.com>
Gerrit-Reviewer: Nir Soffer <nsoffer at redhat.com>
Gerrit-Reviewer: Ramesh N <rnachimu at redhat.com>
Gerrit-Reviewer: Sahina Bose <sabose at redhat.com>
Gerrit-Reviewer: Sandro Bonazzola <sbonazzo at redhat.com>
Gerrit-Reviewer: Timothy Asir <tjeyasin at redhat.com>
Gerrit-Reviewer: automation at ovirt.org
Gerrit-Reviewer: oVirt Jenkins CI Server
Gerrit-HasComments: Yes


More information about the vdsm-patches mailing list