|
I can't imagine people willingly buying new Oracle stuff when given the chance. I'd rather not use software than have to deal with Oracle. "Yes I would gladly pay a lot of money to a vendor and get abused everytime they think of some new licensing trickery."
|
# ? May 14, 2016 06:39 |
|
|
# ? May 8, 2024 07:27 |
|
The point is for many BI problems you don't have much of a choice.
|
# ? May 14, 2016 06:55 |
|
evil_bunnY posted:The point is for many BI problems you don't have much of a choice. I don't know. Aren't there a applications from IBM and the like? But then you are trading IBM for Oracle, which is worse?
|
# ? May 14, 2016 07:02 |
|
adorai posted:they bought sun so they could sue google. Heh, there's the newer Oracle vs. IBM lawsuit but I haven't had time to dig into that one. NippleFloss posted:Oracle is everywhere. Building new apps on Oracle is less common, but there's plenty of existing apps out there that run on it that will be around for a long time and will still be running on Oracle for all of it because change is difficult. Agreed, but it seems they've recognized they've got all their clients in a bind, their actively exploiting them and not in a good way. Mr Shiny Pants posted:I don't know. Aren't there a applications from IBM and the like? At the end of the day technology companies are just corporations but some are better than others.
|
# ? May 14, 2016 08:31 |
|
Tab8715 posted:At the end of the day technology companies are just corporations but some are better than others. And they all seem to be trying their best to actively piss off customers it seems. I don't understand why pissing off you customers looks like a good plan to these companies, don't they understand they will be dropped like a bad habit when it is feasible? And with the whole industry in flux like it is it seems like a smart move to keep your customers happy that they chose your technology stack.
|
# ? May 14, 2016 11:56 |
|
Mr Shiny Pants posted:And they all seem to be trying their best to actively piss off customers it seems.
|
# ? May 14, 2016 16:38 |
|
adorai posted:They are beholden to their shareholders. Those shareholders want short term returns over long term customer retention.
|
# ? May 14, 2016 16:49 |
|
Tab8715 posted:Heh, there's the newer Oracle vs. IBM lawsuit but I haven't had time to dig into that one. Pretty much every engineer in Silicon Valley is aware of this as well, so none of them want to go to work for Oracle. That means they will never develop good software again, they just have to keep exploiting.
|
# ? May 14, 2016 17:18 |
|
I'm having an unusual problem with a VM. It was previously mounting an ISO on the host's local storage. That's been changed to client device, but the VM insists that it's still using the local datastore for something. I've confirmed the virtual disks are both on the SAN, and I don't see any mention of the local datastore in the vmx file either. I've tried mounting and unmounting ISOs from a variety of locations, but I can't seem to break that connection. Any ideas what might be causing this? This is an exchange server, and is currently stuck on its current host due to this ghost datastore connection, which obviously is not desirable. e: pretty sure I've figured it out. There's an old snapshot on the VM, which will hang on to that datastore reference even after you switch to client or host drive. We're going to wait until tonight to delete it, but hopefully that will solve the problem. stubblyhead fucked around with this message at 23:25 on May 18, 2016 |
# ? May 18, 2016 20:22 |
|
Goodbye C# client, hello HTML5 client. http://blogs.vmware.com/vsphere/2016/05/goodbye-vsphere-client-for-windows-c-hello-html5.html
|
# ? May 18, 2016 20:39 |
|
|
# ? May 18, 2016 20:46 |
|
Dumb question, but here goes. Say I have two connections to a NIC and both of those are on the same vswitch. Do I need to put those in a LACP group on my switch? Any quick and dirty vmware networking guides out there?
|
# ? May 18, 2016 21:37 |
|
NippleFloss posted:Goodbye C# client, hello HTML5 client. Bison_YES_YES.mpg
|
# ? May 18, 2016 22:19 |
|
Mr-Spain posted:Dumb question, but here goes. Say I have two connections to a NIC and both of those are on the same vswitch. Do I need to put those in a LACP group on my switch? The depends on the load balance policy on the vswitch. If your switch is set to load balance based on originating port ID, which is the default, then you do not need any special switch configuration. The IP hash load balance mechanism implies that you have an etherchannel configured on the switch side. For standard vSwitches that will be a static etherchannel, for DVswitches it will be LACP.
|
# ? May 18, 2016 23:25 |
|
LACP will provide more aggregate bandwidth to a single VM and is a bit better/faster when it comes to detecting and recovering from a failed link but if you're doing a home lab the default is probably fine but LACP is good practice if your hardware supports it.
|
# ? May 19, 2016 07:22 |
|
The HTML5 interface is pretty great. I've had some bugs but I expect that with flings. You have no idea how much I'm looking forward to that experience translating to the vSphere web client. I've actually gotten so used to the flash client that I prefer it to the C# one for just about everything. It's not super responsive but the memory of C# responsiveness has faded with time so I don't even complain about that anymore. I'm just praying that they finally support this on Safari so I can stop seeing every time I log in. It's actually worked 100% fine for the past two or so years but it still annoys me every time I see it
|
# ? May 19, 2016 14:22 |
|
Martytoof posted:I've actually gotten so used to the flash client that I prefer it to the C# one for just about everything. It's not super responsive but the memory of C# responsiveness has faded with time so I don't even complain about that anymore.
|
# ? May 19, 2016 17:08 |
|
This is gonna be fun when Chrome blocks Flash for most websites later this year.
|
# ? May 19, 2016 17:43 |
|
I've been having a spat of VMs rebooting and getting new NICs with no IPs while the old ones show as disconnected in device manager. There's a VMware kb from back in the 4.0/4.1 days that blames this on the hotplug feature and their "solution" is to disable it globally for the VM which sounds like a huge pain in the rear end. Is there something goofy going on with the latest tools release? I tested the 2008r2 rollup on 8 systems yesterday and five of them came up with new nics and no network access and I'm terrified over what is going to happen after next patch week. We had one do it in with May patches but it seemed like an isolated incident at that point.
|
# ? May 19, 2016 18:53 |
|
Bhodi posted:lmao, this is the most stockholm syndrome thing I've ever read. "it's not so bad once you forget how good the older version was" I only mentioned the speed of the C# client, not that it was in any other way better than the flash UI. I preferred the convenience and aesthetics of the flash UI from day one, but Flash itself is a bloated technology so it was never going to break any speed records. I'd trade in a loading bar every now and then to not have to RDP from my macbook into a Windows machine to run the C# client anyway. some kinda jackal fucked around with this message at 19:49 on May 19, 2016 |
# ? May 19, 2016 19:46 |
|
BangersInMyKnickers posted:I've been having a spat of VMs rebooting and getting new NICs with no IPs while the old ones show as disconnected in device manager. There's a VMware kb from back in the 4.0/4.1 days that blames this on the hotplug feature and their "solution" is to disable it globally for the VM which sounds like a huge pain in the rear end. Is there something goofy going on with the latest tools release? I tested the 2008r2 rollup on 8 systems yesterday and five of them came up with new nics and no network access and I'm terrified over what is going to happen after next patch week. We had one do it in with May patches but it seemed like an isolated incident at that point. My current theory is that the 2008R2 rollup (KB3125574) is bundled with hotfix KB2846340 https://support.microsoft.com/en-gb/kb/2846340 which is designed to cleanup instances where a new NIC is initialized during driver updates and this is doing something funky with either the template/sysprep process and the old vNIC from there or whatever happens when the NIC driver gets updated with VMware Tools patches and it ends up selecting the wrong disconnected interface from device manager to use as the primary which results in your static IP config getting lost and buried. The problem only seems to occur after installation of the rollup and only once. Be wary and test if you've imported the rollup to your catalog.
|
# ? May 19, 2016 21:04 |
|
I'm that guy who only opens the web client if I have to.
|
# ? May 19, 2016 21:51 |
|
As a frequent Linux user, it's nice not to have to keep a Windows VM on the side just for ESXi administration. Now I can even manage VMs from my Chromebook.
|
# ? May 19, 2016 23:23 |
|
Early on it was garbage. In 5.5 it was slightly less garbage. In 6.0 I think it's actually pretty decent, aside from the fact it's flash based.
|
# ? May 20, 2016 01:52 |
|
stubblyhead posted:Early on it was garbage. In 5.5 it was slightly less garbage. In 6.0 I think it's actually pretty decent, aside from the fact it's flash based. It's HTML5-based, isn't it?
|
# ? May 20, 2016 03:10 |
|
There's an HTML5 console that took until 2 months ago to actually work properly so you could click on the bottom 1/3 of the screen and a ton of the regular UI is flash which seems to cause memory leaks if you leave it up too long. It's bad.
|
# ? May 20, 2016 03:38 |
|
HPL posted:As a frequent Linux user, it's nice not to have to keep a Windows VM on the side just for ESXi administration. Now I can even manage VMs from my Chromebook. FYI -- libvirt is surprisingly capable for basic tasks. It falls down if it needs interaction, and can't manage vswitches/etc at all, but for creating/starting/stopping/migrating VMs, libvirt works. Not that you need it now, but if you still have an older VMware environment somewhere...
|
# ? May 20, 2016 04:33 |
|
evol262 posted:FYI -- libvirt is surprisingly capable for basic tasks. It falls down if it needs interaction, and can't manage vswitches/etc at all, but for creating/starting/stopping/migrating VMs, libvirt works. Not that you need it now, but if you still have an older VMware environment somewhere... Holy crap, I had no idea libvirt could manage ESXi VMs. Consider my mind blown. Still, the 6.0 web app is pretty good. There's some functionality missing from it that is in the Windows client, but I'm sure that'll be along shortly. Before the web fling came along, I used to ssh into my ESXi host and do my basic VM management from there.
|
# ? May 20, 2016 07:05 |
|
What is there in the fat client that still isn't in the web client? The last bastion I can think of was VUM, and it runs pretty nice in the web client now.
|
# ? May 20, 2016 07:26 |
|
stubblyhead posted:What is there in the fat client that still isn't in the web client? The last bastion I can think of was VUM, and it runs pretty nice in the web client now. Speed?
|
# ? May 21, 2016 15:51 |
|
NippleFloss posted:Goodbye C# client, hello HTML5 client. This is very cool; I should be able to drop my windows VM from my Macbook once I get this running (home use), correct?
|
# ? May 21, 2016 21:59 |
|
GobiasIndustries posted:This is very cool; I should be able to drop my windows VM from my Macbook once I get this running (home use), correct? It doesn't have 100% of the functionality of the Windows client yet, so you may want to hang on to it a little longer until you can confirm that the HTML5 client does everything you need it to.
|
# ? May 21, 2016 23:55 |
|
I just deployed the HTML5 web client appliance to my home lab and it is drat sexy. I am not able to resize the bottom panel though which makes it tough on my MacBook Air's smaller resolution, but overall very fast and responsive. Since I use a Mac for work, this is a very welcomed advancement because I am always RDP'd into a Windows VM to use the thick client along side the 6.0u2 web client.
|
# ? May 22, 2016 02:47 |
|
Is there any reason not to be mildly excited about Hyperconverged Storage Spaces Direct?
|
# ? May 23, 2016 20:10 |
|
It's Hyper-v? Nothing against Hyper-v, but I wish they'd finally support nested virt, and make Gen2 VMs running non-Microsoft operating systems less of a PITA. The RDMA or pair of NICs requirement is odd to me. I hope their best practices document suggests a more robust configuration. iWARP (which is probably what they're suggesting instead of ib or rcoe) is ok, but I dunno. IMO, a hyperconverged solution which requires specific hardware for the backend kind of misses the point (something which handles it all on relatively generic hardware), but it's interesting anyway, and I'd guess most clients who want to run this already have the infrastructure anyway.
|
# ? May 24, 2016 02:42 |
|
On the same subject and cross-posting, Is there a fundamental difference between Windows 10 Hyper-V and Windows Server Hyper-V aside from things like FT / HA? I bought an Intel Nuc for a home lab. Low and behold I can't add additional Ethernet drivers without hacking Windows Driver Validation. I just want to run VMs on an independent VLAN with the least amount of managerial overhead for a personal lab environment. I want this setup. Obviously, my home lab isn't directly exposed to the internet but with a P2S VPN in theory this will all come together. Gucci Loafers fucked around with this message at 05:52 on May 24, 2016 |
# ? May 24, 2016 05:37 |
|
Windows 10 Hyper-V doesn't do NIC teaming. But overall, Windows 10 is surprisingly powerful.
|
# ? May 24, 2016 05:52 |
|
evol262 posted:It's Hyper-v? Nothing against Hyper-v, but I wish they'd finally support nested virt, and make Gen2 VMs running non-Microsoft operating systems less of a PITA. Server 2016 does bring nested virt
|
# ? May 24, 2016 06:48 |
|
About time. I guess I haven't been keeping up. Then I looked: quote:Both hypervisors need to be the latest Windows Insider build (10565 or greater). Other hypervisors will not work. Just why? SVM is a little easier to implement nested virt on, but I can't fathom the "nested client must be Hyper-v" requirement unless they're not actually exposing vt-x to the guest and it's some bizarre syscall which invokes a hypercall on the parent.
|
# ? May 24, 2016 06:58 |
|
|
# ? May 8, 2024 07:27 |
|
Because it's more important to have the feature bullet point than have it be useful. IE, why most people only run MS hypervisors because the financials make sense for windows shops.
|
# ? May 24, 2016 10:32 |