virtual_dave1's Posts

Hi all, Installed SRM 4.1 onto VCenter server (also 4.1). Install completes successfully and service is running. Using SQL auth as the SQL 2005 server is remote to the VCenter server. Database seems to be setup OK, documentation was a bit lacking but after a bit of trial and error got the SRM server installed fine, DB is populated with tables etc and no errors pop up during install. I download the plugin to my Vsphere client and install it fine. Restart VIC, log into vcenter as usual, click on Site Recovery and get a popup saying: "Connection to local Site Recovery Manager https://<vcenter-hostname>:8095/dr failed". Thats it no more info. VC\SRM server is 2008 R2. New DB created on SQL 2005 remote server, created new schema under database\Security, created new user (sql user not windows user), mapped user to DB using the new schema. Set user to be owner of Schema. Gave schema permissions to DB. Username matches schema name which also matches DB name. DSN is 32-bit (from c:\windows\syswow64). As it happens in an effort to troubleshoot I have given the user all server roles excluding sysadmin, but specifically it is a member of the bulkadmin server role. So SRM seems to have installed OK, no errors shown in event log or when service starts. But just can't connect the plugin to it. Client is Win7 x64, with UAC enabled, running as normal user - this is how i work day-to-day, and all other vmware stuff works fine like this (VC,VUM etc). Tried running the VIC as local and domain admins but error remains the same. any ideas or pointers? cheers

Hi all ESX 4.0u2 (soon to be esxi 4.1) Procurve 3500yl I have a vswitch for VMs with several port-groups for different VLANs. This vSwitch has 4 pNICs going to a single Procurve 3500yl. On the pswitch, the four ports are trunked so VLANs can be tagged into the vswitch, and on ESX the vswitch load balancing policy is set to IP-Hash (required with procurves). This is set at the vswitch level. In order to improve redundancy I have a second Procurve 3500yl. I would like to take two of the active NICs and place them into Standby. I would cable these into the second pswitch and reconfigure the trunking etc. These would then sit unused until called upon in a failover scenario. Bandwidth-wise, 2 active pnics is more than enough according to the performance metrics. When i go to create this configuration, I get a warning\error mesage - when moving two of the four active NICs down to standby (again at the vswitch level) I get a message saying: "The IP Hash based Load Balancing does not support Standby uplink physical adapters. Change all standby uplinks to active status". As far as I understand it, I need the IP-Hash LB policy to work with the Procurve trunks. Pswitch1 and Pswitch2 are connected via a 2 port (1Gb ports) LACP trunk and also support traffic\applications other than vmware. The inter-switch trunks are not busy at all during normal operations. The switches are not 'stacked' as in single management address. I am not sure if i can have 4 active nics, going to two different switches - they would obviously be two separate trunks at the pswitch end, but would that matter to ESX and the ip-hash based load balancing policy? Can anybody explain a bit about why vmware won't allow this configuraiton, and what I might be able to do to achieve the desired result? My goal is to be able to withstand an outage of pswitch1 (recent power issues at the datacenter have brought this on). outside of ESX, pSwitch1 has the primary IPVPN router connected to it, and pswitch2 has the secondary router, with HSRP running between the two routers across the inter-switch trunk. A complete failure of either pSwitch will still allow connectivity to devices not solely connected to either switch - hence my desire for this setup. Thanks.

Hey guys About to embark on the a consolidation\virtualisation project using VI3. Exciting times. We are in discussion with various 3rd parties who are each putting forward their tender if you like, encompassing both storage and vmware. In terms of storage, we seem to be down to two vendors: Sun Micrososystems and Dell\Equalogic. Sun are proposing one of their new 7000 series 'Amber Road' SAN\NAS servers, and we would be using iSCSI. Can i ask if anybody out there has experience of running vmware on the Sun 7000 SAN? Anybody got any tips or pointers, any gotchas to watch out for when speccing up this SAN? I personally favour the Equalogic approach in terms of simplicity and modularity\expandability. But can't ignore certain features of Sun like the ability to add an FC HBA in at a later date should we outgrow iSCSI. Hearing lots of things from Sun about ZFS and how it's the best thing since sliced bread blah blah. But looking at the features that we would be interested in (as it stands at the beginning of our virtualisation journey), Equalogic appears to have it all covered too. If not already implemented, then on the road map. But then we keep hearing 'yeah it's on the roadmap' from Sun too eg. de-dupe on-box in ZFS. Basically guys i'm wondering if anybody has any advice and more interestingly hands on experience with these new Sun 7000 series boxes. I'm a little dubious that we'll be paying over the odds for a bit of iron because it has a Sun sticker on the front of the box.... but would love to be corrected. For me as a techy, it is a win-win situation!! Sun or Equalogic - i can live with either! Just want to make sure we spend the bosses money on the right product for us. Thanks dave