|
Corvettefisher posted:What verison of ESXi ESXi is running the latest 5.0U1 release. VT-D is enabled in the BIOS. He's running an Intel i7-3770 CPU. Storage is local SATA. He has an Intel dual port gigabit PCIExpress card. 32 GB of RAM in 4 sticks. Mobo has the Z77 chipset. The ISO used to install Windows Server 2008 R2 is stored on my machine and mounted after booting into BIOS (not stored in the datastore on the box) - it was downloaded from Microsoft's TechNet site and has been installed successfully a dozen or so times in my VirtualBox setup. Machines currently provisioned are four Linux VM's (one of them was setup for me as a SmoothWall Firewall for me to play around with). My roommate has the exact same machine at work running 40 or so VM's without a hitch, but I don't know if he's running any Server 2008 R2 VM's, and that's what's causing us to scratch our heads.
|
#
?
May 8, 2012 17:53
|
|
|
|
| # ? Apr 16, 2024 05:10 |
|
|
Daylen Drazzi posted:ESXi is running the latest 5.0U1 release. VT-D is enabled in the BIOS. He's running an Intel i7-3770 CPU. Storage is local SATA. He has an Intel dual port gigabit PCIExpress card. 32 GB of RAM in 4 sticks. Mobo has the Z77 chipset. The ISO used to install Windows Server 2008 R2 is stored on my machine and mounted after booting into BIOS (not stored in the datastore on the box) - it was downloaded from Microsoft's TechNet site and has been installed successfully a dozen or so times in my VirtualBox setup. Machines currently provisioned are four Linux VM's (one of them was setup for me as a SmoothWall Firewall for me to play around with). http://communities.vmware.com/community/vmtn/beta/workstationtp2012 And run ESXi in a vm, it works well enough. Check the mobo for Bios updates, and what not. Make sure the Mobo is set to AHCI mode and not IDE.
|
|
#
?
May 8, 2012 18:04
|
|
|
We'd like to move our DR location from our property to a 3rd-party location. Are there companies that do hosted ESXi? Or is it better (cheaper or easier) to just colocate our own hardware? We're only looking at 1 ESXi 5 host, plus a storage array if colocating.
|
|
#
?
May 8, 2012 19:51
|
|
|
Erwin posted:We'd like to move our DR location from our property to a 3rd-party location. Are there companies that do hosted ESXi? Or is it better (cheaper or easier) to just colocate our own hardware? We're only looking at 1 ESXi 5 host, plus a storage array if colocating. Shouldn't be any different than co-locating any other server. Isn't there something in the EULA that doesn't allow people to re-sell VM's on ESXi? We have a (Dell R710) server @ Rackspace that runs VMware ESX 4.0.0 and we can add/remove VM's as we please (actually, we have to call them to do it and they charge us more a month for each one) but that's the 'Private Cloud' product and each VM is managed by them.
|
|
#
?
May 8, 2012 19:58
|
|
|
Bob Morales posted:Shouldn't be any different than co-locating any other server. That sounds awful. I just want an ESXi host in someone else's building that is connected to my vCenter server, and that I do all the configuring on (I don't want it managed). I'm just trying to figure out if there are companies that rent out a physical box with ESXi installed, or if it's better to colocate my own hardware. I'd rather pull my existing hardware back to my production location as an extra host and then rent a host somewhere. edit: sorry, didn't mean to sound like a dick. I mean it sounds awful for my purposes to pay per VM. I want to have an ESXi host attached to my vCenter server to which I can replicate VMs with Veeam for failover, and run a few full-time offsite VMs. A Google search says people do host ESXi, so I guess I'm wondering if anyone has experience in hosting or colocating an ESXi host under vSphere licensing. Erwin fucked around with this message at 20:38 on May 8, 2012 |
|
#
?
May 8, 2012 20:04
|
|
|
Bob Morales posted:Isn't there something in the EULA that doesn't allow people to re-sell VM's on ESXi?
|
|
#
?
May 8, 2012 20:30
|
|
|
Erwin posted:That sounds awful. I just want an ESXi host in someone else's building that is connected to my vCenter server, and that I do all the configuring on (I don't want it managed). I'm just trying to figure out if there are companies that rent out a physical box with ESXi installed, or if it's better to colocate my own hardware. To be fair we're paying for the management and I think the people who signed the agreement with our host figured we'd be spinning up VM's (we don't have the capacity to really add more anyway) to add features or some poo poo down the road. Really, we should have just gotten 2 servers, one web and one db, or tried to design our stuff to actually use some sort of 'cloud' scaling. Right now the only advantage we really have using virtual machines is allocating ram here or there, and we're taking the slight performance hit of running on VM's (whatever that may be) We would be better served having say 3 web front-ends instead of one. But we'd have to either do it in physical hardware or jump up to a dual-hexcore setup with more RAM so we could spread resources out. Right now we have 4 vCPU's and no RAM left for another VM.
|
|
#
?
May 8, 2012 20:53
|
|
|
Welp Setting up some Hyper-V/Centos Xen clusters before I go, my place was shocked when I told them I couldn't offer support. These servers we have aren't on the HCL for 5 only upto esxi 4.1 4u. Going to be fun cramming in guides to XEN, along with how to use it...
|
|
#
?
May 8, 2012 21:03
|
|
|
Corvettefisher you should also throw in a Lync deployment how-to while you're at it, but I doubt they'd get the joke anyway.
|
|
#
?
May 8, 2012 22:20
|
|
|
Daylen Drazzi posted:Got an interesting issue with my roommate's box and hope someone can help resolve it. Generally speaking, ESX/ESXi does not hang; it should time out somewhere and eventually purple-screen. Complete hard-lockups are treated like hardware issues, usually, but I find it curious you can reproduce it. Probably now it'd be a possible erratum (hence recommending BIOS/firmware upgrades). What you can try to do is find out if you can NMI the system, and to do it while the host is hung. Firing an NMI on ESXi 5 (as long as that interrupt vector is enabled) will cause it to purple-screen, indicating it was not truly hung, but deadlocked somewhere in the hypervisor. Earlier products had to be configured to respond in this way. Putting myself in your buddy's shoes, it's not a nice outlook either way. I'm hopeful the BIOS/Firmware upgrades will do it, but if not, you might need to consider turning off virtualization options like vt-d. If he has other performance related options in the BIOS, they may need to be tweaked. CPU states (C/Sleep) should be disabled when troubleshooting anomalous behavior like that, as well as setting the performance profile to maximum. How much RAM is that VM booting with, anyway? And how many CPUs? At what stage exactly was the Windows install locking up the ESXi box? Edit: Kind of glossed over the fact that you had at least one working W2K8 R2 VM. Is there anything different between the two VMs' configurations? And just to be sure, is the system actually locked up? If your buddy is willing to reproduce it oen more time, try hitting alt+f12 on the console, then spin up another VM to hang it, and see if the activity on the screen stops, as well as Numlock, etc. See if that last activity on the screen shows anything useful (take a picture and post it if you want). It's near real-time logging from the host's syslog daemon. Might just be vmkernel components, I can't remember - but it's all that mattered here anyway. Kachunkachunk fucked around with this message at 23:10 on May 8, 2012 |
|
#
?
May 8, 2012 23:03
|
|
|
Oddhair posted:Corvettefisher you should also throw in a Lync deployment how-to while you're at it, but I doubt they'd get the joke anyway. I about ready to write a "How to be a good systems admin" guide for this job. I am too the point where it is basically click on deploy and servers are ready to go. E: hmm that wouldn't be a bad thread idea, "Systems Admin guide to automating/optimizing your environment" Dilbert As FUCK fucked around with this message at 23:55 on May 8, 2012 |
|
#
?
May 8, 2012 23:48
|
|
|
Corvettefisher posted:I about ready to write a "How to be a good systems admin" guide for this job. I am too the point where it is basically click on deploy and servers are ready to go. I'd read it.
|
|
#
?
May 9, 2012 00:04
|
|
P.S. STILL COMPLETELY DEVOID OF MERIT
|
Anyone know any Good Xen server management apps for windows? Terminal is not these peoples friends
|
|
#
?
May 9, 2012 00:29
|
|
|
Corvettefisher posted:E: hmm that wouldn't be a bad thread idea, "Systems Admin guide to automating/optimizing your environment" There's been a few Puppet/Chef configuration management threads, which I consider a major part of that process if you run *NIX, but they always die off. Not that I think it's a bad idea, maybe they just didn't have a strong OP or something.
|
|
#
?
May 9, 2012 02:13
|
|
|
Kachunkachunk posted:Box lockups are a toss-up between hardware and software issues, really. If there's one thing I learned, it was that ESX/ESXi was excellent at exposing hardware problems (architecture/design or physical issues), firmware issues, and errata. Thanks - I'll mention these things to him and see what he says. We talked for a couple minutes this evening about it and he has a suspicion something might be up with the RAM as it shows up in BIOS as DDR3 1333 and he specifically ordered DDR3 1600. As for the virtual machine, I just tried installing it and it finally took on the third try; I had a very uneasy feeling that it would crash again when installing the second virtual machine so it wasn't too big of a surprise, but we're reluctant to try it again since the ESXi box hosts his gateway and other services. The reality of the situation is that he does this type of stuff all day long as part of his job - I'm just a computer janitor and want to try and learn something about VM's so I can get a better job. I'm really great at breaking his toys; not so good when it comes to helping him fix them since he's working on a much higher level than I can assist with. Frankly I'm leaning towards taking my Core2Duo machine with 8GB of RAM and setting it up as my personal ESXi box and have it on it's own little VLAN. Then I can break it to my heart's content without worrying about taking down the entire house network. Until then I'm going back to working with my VM's I set up in VirtualBox on my primary machine.
|
|
#
?
May 9, 2012 03:11
|
|
|
The RAM speeds probably shouldn't cause problems there, but at least on that topic, I also had 16GB of the stuff not show up with the right speed on my recent build. Reconfiguring the BIOS to run it at its recommended speed and timings has shown no problems so far (more than a couple of weeks of full VM load already). Also in my post regarding turning off options - that's all just during troubleshooting. You'll basically only find certain features or components that seem to stress to the point of failure when used. In the order of what you can sacrifice, it's power saving, then virtualization extensions/capabilities. You *need* virtualization technology to be set, but vt-d is a [very] nice-to-have and not critical. I do however make it a necessity for my builds as I like to use directpath I/O. If you're really just looking to learn, you can still run a few ESXi hosts in VMware Workstation, if you like. To learn/repro some issues with Auto-Deploy and Host Profiles, I built a cluster of ESXi VMs on my laptop (16GB of RAM), and it was no problem at all. vCenter's the real hog, if anything. This plan of course requires you to have Workstation (or to have a trial/filez copy). Also sounds like you're both roomies. He doesn't actually work at VMware, does he? It's a strangely common living situation between a handful of employees from Support there, heh.
|
|
#
?
May 9, 2012 05:37
|
|
|
Corvettefisher posted:Anyone know any Good Xen server management apps for windows? Terminal is not these peoples friends
|
|
#
?
May 9, 2012 05:41
|
|
|
Oh how could I forget hyperVMDocjowles posted:There's been a few Puppet/Chef configuration management threads, which I consider a major part of that process if you run *NIX, but they always die off. Not that I think it's a bad idea, maybe they just didn't have a strong OP or something. I would make it a lengthy OP similar to that of this OP or the how to be an alchy thread with some detailed "how to's" for people not sure Dilbert As FUCK fucked around with this message at 13:51 on May 9, 2012 |
|
#
?
May 9, 2012 13:38
|
|
|
Misogynist posted:oVirt? Not that I can help with the original question, but oVirt is upstream RHEV. It's KVM only, basically, and it wants to either be bolted onto Fedora or to be installed on bare metal (as a Fedora spin). Not going to do what he wants. See here, though. I'm sure there's some dodgy PHP thing that works reasonably well on that page.
|
|
#
?
May 9, 2012 15:58
|
|
|
Kachunkachunk posted:Also sounds like you're both roomies. He doesn't actually work at VMware, does he? It's a strangely common living situation between a handful of employees from Support there, heh. Unfortunately for me he doesn't. Fortunately for me he has a kickass job that lets him work with a lot of different technologies, some of which we get to play around with at the house. As I said earlier, he operates on a whole different level than I do, and picks up concepts and theory of complicated subjects in less time than it takes me to actually figure out how to spell the words.
|
|
#
?
May 9, 2012 16:00
|
|
|
Are there any tools to check to see if a datastore is actively in use, or if any powered off VMs rely on that datastore? Using RVTools, the vDatastore tab is showing 0 VMs, but there seems to be orphaned snapshots that are stored there. I don't think I will have a problem deleting them, but just wondering if there is any other official way of verifying this.
|
|
#
?
May 9, 2012 16:45
|
|
|
Moey posted:Are there any tools to check to see if a datastore is actively in use, or if any powered off VMs rely on that datastore? I don't know of any tools, but there are console commands. Unfortunately, I do not do it often enough to know them off the top of my head.
|
|
#
?
May 9, 2012 16:51
|
|
|
Moey posted:Are there any tools to check to see if a datastore is actively in use, or if any powered off VMs rely on that datastore? Does browsing the datastore show and directories of known used VM's?
|
|
#
?
May 9, 2012 17:19
|
|
|
Corvettefisher posted:Does browsing the datastore show and directories of known used VM's? It has folders in there with the names of the VMs, then the orphaned snapshots inside. But when I look at the actual VM within vSphere, it doesn't show that it is using that datastore at all (and there are no snapshots for any of these VMs). There is also seemingly a full copy of a VM in one of the datastores, that has a 0kb VMX file, but the VM is perfectly fine running on a different datastore. What my old boss did to some of this stuff, I have no idea... Edit: gently caress it, time to blow this stuff out.
|
|
#
?
May 9, 2012 17:45
|
|
|
So we're going to be replacing some older infrastructure stuff, and we're trying to virtualize everything. I'm really only planning on virutalizing about 4 physical servers, none with any kind of heavy IO load at all. Think basic windows stuff, file/print share, DC, low usage IIS and WSUS. We're a HP shop so am I off base in thinking 2x DL360's with a LeftHand iSCSI SAN is a good starting point? We can tweak the actual specifications of the hardware such as cores and RAM. We do have some larger EMC SAN's so EMC isn't out of the question but I'm not familiar with their smaller offerings.
|
|
#
?
May 9, 2012 18:24
|
|
|
skipdogg posted:So we're going to be replacing some older infrastructure stuff, and we're trying to virtualize everything. I'm really only planning on virutalizing about 4 physical servers, none with any kind of heavy IO load at all. Think basic windows stuff, file/print share, DC, low usage IIS and WSUS. We're a HP shop so am I off base in thinking 2x DL360's with a LeftHand iSCSI SAN is a good starting point? We can tweak the actual specifications of the hardware such as cores and RAM. Are you looking at Lefthand physical or the VSA? VSA may be worth a look.
|
|
#
?
May 9, 2012 18:44
|
|
|
skipdogg posted:We do have some larger EMC SAN's so EMC isn't out of the question but I'm not familiar with their smaller offerings. We recently bought their very-low-end VNXe which seems like pretty good bang for the buck. I'm very happy with it but I admit to not having extensive experience (or needs) with storage. I know others in SH/SC kind of hate EMC. But if you want a speedy box with a lot of built in redundancy, a support contract and standard features like snapshots and replication, you could do worse IMO. Can't disclose our pricing obviously but you're looking at very low 5 figures depending on the exact drive loadout.
|
|
#
?
May 9, 2012 19:00
|
|
|
Moey posted:It has folders in there with the names of the VMs, then the orphaned snapshots inside. But when I look at the actual VM within vSphere, it doesn't show that it is using that datastore at all (and there are no snapshots for any of these VMs). If you edit the settings of each of those VMs, are the .vmdk files just the name of the VM, or is there a snapshot number attached? EG myvmname-0000007.vmdk. If that's the case, even if snapshot manager doesn't show any existing snapshots, it may be riding on that snapshot. Doing 'consolidate' in ESXi 5 fixes this, but I think the behavior is different in 4? Someone posted a powershell script a while back that shows all snapshots, and it did show the snapshots in situations like this, even if snapshot manager didn't. You may want to run that before blowing anything away. edit: or make an alarm for snapshot size over like 1k, and it should pick that up (but not guaranteed).
|
|
#
?
May 9, 2012 19:12
|
|
|
Why aren't you just storing the VMs on the existing storage?
|
|
#
?
May 9, 2012 19:18
|
|
|
It might be a long shot, but does anyone here use GhettoVCB as a temporary backup solution? My workplace runs three ESXi servers all running the latest release of ESXi 5.0 Update 1. We had GhettoVCB in place until we installed the latest patch, which seems to have broken it. The last time GhettoVCB ran it stalled when removing the snapshot after successfully backing up a host. So far I've been unable to find much of anything online relating to this problem, which leads me to believe that perhaps not-so-many people are using GhettoVCB any more. What would you guys use for a free, automat-able backup solution for ESXi?
|
|
#
?
May 9, 2012 19:54
|
|
|
I'm looking to pull the trigger on the whitebox configuration below. I'm planning on using my existing qnap 219p+ as a datastore using NFS. http://www.newegg.com/Product/Product.aspx?item=N82E16820139262 x 2 http://www.newegg.com/Product/Product.aspx?item=N82E16813182253 http://www.newegg.com/Product/Product.aspx?item=N82E16819115084 http://www.newegg.com/Product/Product.aspx?item=N82E16811144162 http://www.newegg.com/Product/Product.aspx?item=N82E16820220251 I'm pretty much hitting the exact price point I want with this build. Am I getting the best bang for my buck? I am somewhat worried about power consumption which is why I went with Intel vs AMD.
|
|
#
?
May 9, 2012 20:35
|
|
|
Erwin posted:If you edit the settings of each of those VMs, are the .vmdk files just the name of the VM, or is there a snapshot number attached? EG myvmname-0000007.vmdk. If that's the case, even if snapshot manager doesn't show any existing snapshots, it may be riding on that snapshot. Doing 'consolidate' in ESXi 5 fixes this, but I think the behavior is different in 4? The vmdk files that the running VM is using is named properly, so it is not relying on the snapshot at all. We are running ESXi 4.1, and I am not aware of any consolidate feature. What I was able to do was create a snapshot, and delete it right away. Even though those "orphaned" snapshots were not showing in the manager, removing the one I just created, removed the others. Aaearon posted:I'm looking to pull the trigger on the whitebox configuration below. I would look at getting the 1235 instead of the chip you have picked. Not a big price increase, does HT and also has integrated video so you do not have to toss in a video card anytime you need physical console access (or just for the base install). Moey fucked around with this message at 20:41 on May 9, 2012 |
|
#
?
May 9, 2012 20:36
|
|
|
The board has KVM over IP as well as a video card so I should be set there. I will spring for the 1230 as it seems to be the same as the 1235 minus built in video.
|
|
#
?
May 9, 2012 23:53
|
|
|
My vSphere 5 migrations are nearly complete. I just have 3 VMs left to migrate from ESX 3.5. Unfortunately those are the 3 most important VMs so I'm a lot more nervous than I was on the others. Both of our Active Directory servers (our primary, and our secondary/file server) and our Groupwise server are all that's left. I've had a few minor hiccups doing this migration, mostly relating to the upgraded VM Version. It seems that after migrating every VM thinks that it has a new network card, and the new network card booted with DHCP enabled giving the server a new IP address. The old network card was broken and wouldn't connect. In Linux this wasn't a problem, and it was trivial to remove the old settings and update it with the new. On the few windows boxes I've done I had to run a bunch of command line options before running the device manager, and then showing hidden devices, then deleting the old card before configuring the new one. The problem with doing that, was DNS resolution. After the server got it's DHCP address, and I changed it back to the proper static address, the DNS resolution kept reporting the old DHCP'd IP. No amount of /flushdns or /registerdns would get the new IP. Then, when I did get my local machine to see the IP properly, someone else's machine wouldn't. In one case I finally just set a new host record in our DNS server. Is there a way to avoid this whole DHCP/IP Changing issue? For the few servers we've moved so far it hasn't been a major problem, but it will be for our Active Directory and File Servers.
|
|
#
?
May 10, 2012 15:13
|
|
|
Frozen-Solid posted:My vSphere 5 migrations are nearly complete. I just have 3 VMs left to migrate from ESX 3.5. Unfortunately those are the 3 most important VMs so I'm a lot more nervous than I was on the others. Both of our Active Directory servers (our primary, and our secondary/file server) and our Groupwise server are all that's left. and yeah windows handling of old NICs is awful
|
|
#
?
May 10, 2012 15:23
|
|
|
Is there seriously no way to set the iLO password of an ESXi 4.1 host on HP hardware via software? In Linux there's hponcfg that can do this, and a .vib is provided by HP for hponcfg for ESXi 5.0, but I can't find one for 4.1. Don't really want to upgrade to ESXi 5 either.
|
|
#
?
May 10, 2012 17:01
|
|
|
It's not a blade by chance, is it? You can reset it via the chassis if so. If you're on a DL-series then I have no idea 
|
|
#
?
May 10, 2012 17:05
|
|
|
DL-series 
|
|
#
?
May 10, 2012 17:13
|
|
|
Good thing you're running vmware so you can just vmotion all the guests off to another vhost and put it into maintenance mode so you can reboot the machine and hit F8 during POST to reset it. Right? 
|
|
#
?
May 10, 2012 17:19
|
|
|
|
| # ? Apr 16, 2024 05:10 |
|
|
Well yeah but I'm getting tired of all these trips to the colo facility on the other side of town, and one host is running vms that aren't on shared storage for performance reasons (but downtime for these is begrudgingly tolerated). Guess which host has an iLO password tat nobody remembers?
|
|
#
?
May 10, 2012 17:52
|
|