> On Jan. 21, 2013, 8:15 p.m., David Nalley wrote: > > setup/db/create-schema.sql, line 1613 > > <https://reviews.apache.org/r/9041/diff/1/?file=250528#file250528line1613> > > > > You are setting the max_hosts_per_cluster for all VMware versions to > > 32, yet I thought the proposal was only to affect 5.1 with the updated VMFS > > version.
Refer to my mail on http://mail-archives.apache.org/mod_mbox/incubator-cloudstack-dev/201301.mbox/%3c2529883e7b666f4e8f21f85aada43ca7010c8f172...@banpmailbox01.citrite.net%3E Also recently I saw another discussion on full clone support so this limitation of 8 in earlir versions shouldn't hold. > On Jan. 21, 2013, 8:15 p.m., David Nalley wrote: > > setup/db/db/schema-40to410.sql, line 81 > > <https://reviews.apache.org/r/9041/diff/1/?file=250529#file250529line81> > > > > You've altered the table and have now defined it as 32 in all versions > > here as well. Same as above - Koushik ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/9041/#review15550 ----------------------------------------------------------- On Jan. 21, 2013, 11:44 a.m., Koushik Das wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/9041/ > ----------------------------------------------------------- > > (Updated Jan. 21, 2013, 11:44 a.m.) > > > Review request for cloudstack. > > > Description > ------- > > For Vmware max. hosts in a cluster is specific to HV version. This > configuration is read from the hypervisor_capabilities table. > Read the dev list discussion here: > http://apache.markmail.org/thread/lva7loblbksiavm7 > > Scenarios: > - Adding a cluster to CS - From 1st host in the list of hosts discovered, > query the HV version. Based on the version read the max_hosts_per_cluster > limit from hypervisor_capabilities table and do the validation > - Adding a host to an existing cluster - Get the HV version from one of the > hosts already present in cluster from CS db., then similar to previous > scenario > > The implementation is specific to Vmware only, haven't made the validation > logic generic as there are no limits for other HVs currently. > For upgraded scenarios, HV versions may not be present for the existing host > table entries. In that case the 'default' version will get used to read the > limit. > > > This addresses bug CLOUDSTACK-672. > > > Diffs > ----- > > api/src/com/cloud/hypervisor/HypervisorCapabilities.java efb8a6a > core/src/com/cloud/hypervisor/HypervisorCapabilitiesVO.java 56e8e0a > > plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/VmwareServerDiscoverer.java > 684df54 > > plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/manager/VmwareManager.java > be3fe9f > > plugins/hypervisors/vmware/src/com/cloud/hypervisor/vmware/manager/VmwareManagerImpl.java > c450312 > server/src/com/cloud/configuration/Config.java 4ae144e > server/src/com/cloud/hypervisor/dao/HypervisorCapabilitiesDao.java 8f4a475 > server/src/com/cloud/hypervisor/dao/HypervisorCapabilitiesDaoImpl.java > 1a6fb72 > setup/db/create-schema.sql 174f53c > setup/db/db/schema-40to410.sql a9d168d > > Diff: https://reviews.apache.org/r/9041/diff/ > > > Testing > ------- > > Manually verifed the scenarios mentioned above. > > > Thanks, > > Koushik Das > >
