|
BnT posted:Goons, I'm tasked with a vSphere 4.1 to 5.x upgrade (a dozen blades, 300 vms). Were you to be tasked with this would you (a) go to 5.0x or (b) go to 5.1x? I guess another question is now or later, as in go to a stable 5.0 version now or a 5.1 version as soon as they clean it up. I saw some 5.1 hate in the last couple pages, which is making me nervous. I really didn't have too much trouble with the vCenter 5.1 install. Hit a snag setting up the SSO database to our external SQL server, but after that it was running fine. I am a smaller environment than you (6 hosts) though. One big reason I wanted 5.1 so bad was since we are only running Standard, with 5.1 I could finally use Storage vMotion. Edit: I also didn't bother with update manager for my hosts. Our setup is pretty simple and had a handful of changes to make to each so I just rebuilt each one by hand. If it were a bigger environment and more complex, I would have gone a more automated route.
|
# ? Dec 14, 2012 18:52 |
|
|
# ? Jun 10, 2024 11:31 |
|
BnT posted:Goons, I'm tasked with a vSphere 4.1 to 5.x upgrade (a dozen blades, 300 vms). Were you to be tasked with this would you (a) go to 5.0x or (b) go to 5.1x? I guess another question is now or later, as in go to a stable 5.0 version now or a 5.1 version as soon as they clean it up. I saw some 5.1 hate in the last couple pages, which is making me nervous. Really depends, however I would go to 5.0u1, personally I haven't really had any problems with 5.1 however a large amount of people have. vSphere 5.0 overall has been stable, which is where I would go to, you can always upgrade to 5.1 however downgrading to 5.0 is not something you want to go through, and if you upgraded the virtual hardware oh boy. Are there any features you are looking for in 5.1 like SR-IOV, no reboots for tools, and a full list here http://www.vmware.com/files/pdf/products/vsphere/vmware-what-is-new-vsphere51.pdf . If you do see a good amount of features you want I would set it up in a lab and see what is needed to bring you up to 5.1. Dilbert As FUCK fucked around with this message at 19:21 on Dec 14, 2012 |
# ? Dec 14, 2012 19:16 |
|
Nitr0 posted:5.1. Make sure you setup SSO properly. Make sure your host profiles are up to date. Use update manager. Broken VDR, terrible web-interface that uses Flash (hey, Vmware, Adobes cancelling this!) and java processes, there is quite a list. Typical VMware rushed launch.
|
# ? Dec 14, 2012 19:34 |
|
vty posted:Broken VDR, terrible web-interface that uses Flash (hey, Vmware, Adobes cancelling this!) and java processes, there is quite a list. Typical VMware rushed launch. Are you not using Data Protection? http://www.vmware.com/products/datacenter-virtualization/vsphere/data-protection.html It is replacing VMware Data Recovery.
|
# ? Dec 14, 2012 19:39 |
|
don't use vdr, don't use the web interface
|
# ? Dec 14, 2012 19:43 |
|
I haven't had any problems with VDP since it actually became VDP...
|
# ? Dec 15, 2012 00:04 |
|
I like that the summary tab for vCenter servers, datacenters, and clusters show the total storage capacity and use for the object, but I've noticed the total capacity listed for my vCenter is twice the sum of all datacenter objects, and usage is just randomly different. Anyone know how it's calculated? Seems like a bug, but probably not worth my effort opening a ticket.
|
# ? Dec 17, 2012 22:28 |
|
Can you boot ESXi from a SAN. Seems there was a fair bit of handwringing about it back on 3.x and 4.x but 5 seems to be supported. Anyone doing it, is it finicky?
|
# ? Dec 18, 2012 18:52 |
|
KennyG posted:Can you boot ESXi from a SAN. Seems there was a fair bit of handwringing about it back on 3.x and 4.x but 5 seems to be supported. Anyone doing it, is it finicky?
|
# ? Dec 18, 2012 19:00 |
|
KennyG posted:Can you boot ESXi from a SAN. Seems there was a fair bit of handwringing about it back on 3.x and 4.x but 5 seems to be supported. Anyone doing it, is it finicky? Dilbert As FUCK fucked around with this message at 19:09 on Dec 18, 2012 |
# ? Dec 18, 2012 19:07 |
|
KennyG posted:Can you boot ESXi from a SAN. Seems there was a fair bit of handwringing about it back on 3.x and 4.x but 5 seems to be supported. Anyone doing it, is it finicky? I've had great luck with it on both 4.x and 5.x. Almost every Cisco UCS install I've seen has involved ESXi boot from SAN. Every Vblock boots from SAN. It's a well-traveled road--nothing to be afraid of. Autodeploy is neat, but somewhat finicky to setup. Pantology fucked around with this message at 20:19 on Dec 18, 2012 |
# ? Dec 18, 2012 20:03 |
|
Does esxi need its OS/boot storage once it's started up or does it load completely into RAM? We have a few servers that boot from sdcard, I've always wondered what would happen if the card failed.
|
# ? Dec 18, 2012 20:22 |
|
sanchez posted:Does esxi need its OS/boot storage once it's started up or does it load completely into RAM? We have a few servers that boot from sdcard, I've always wondered what would happen if the card failed. From my labs it seems to run just fine without it once fully booted, everything loads into ram. Pantology posted:Autodeploy is neat, but somewhat finicky to setup. Not to mention updates are: 1. Replace master image 2. Reboot host(s) Dilbert As FUCK fucked around with this message at 20:33 on Dec 18, 2012 |
# ? Dec 18, 2012 20:25 |
|
sanchez posted:Does esxi need its OS/boot storage once it's started up or does it load completely into RAM? We have a few servers that boot from sdcard, I've always wondered what would happen if the card failed. I think I posted about testing this earlier in the thread -- pulled the sd card out of a running 5.0 server for a week with no ill effects. KS fucked around with this message at 20:34 on Dec 18, 2012 |
# ? Dec 18, 2012 20:32 |
|
KS posted:I think I posted about testing this earlier in the thread -- pulled the sd card out of a running 5.0 server for a week with no ill effects. From what I've seen on the the command line, it copies everything to a memory disk to run from. Changes are saved on shut down when they get tarred, gzipped, and stuffed back onto the SD card. ESXi might get unhappy if you go to shutdown the server (maybe even enter maintenance mode) without an SD card in the socket.
|
# ? Dec 18, 2012 20:40 |
|
That makes sense and is good news. Problems on reboot are no big deal.
|
# ? Dec 18, 2012 20:41 |
|
I'm really beginning to hate this server.. A client we picked up a while back has a pretty unnecessary setup done by some part timer who skipped town. They have a single host running esxi 4.0 hosting one SBS 2003 VM Long story short, their backups to a LTO3 deck have been silently failing due to a bad drive, something I finally uncovered during some maintenance, but the server was under warranty. Dell sends us a part,I go out there, gracefully shut down the VM, then the ESX, replace the Tape deck. The PERC controller pops up on boot with an array degraded message.. gently caress ( raid 6 array 5x500Gb + an added on 1 TB standalone) ESXi boots up fine, I go to power the VM up, and gently caress me WINDOWS\SYSTEM32\CONFIG\SYSTEM corrupt. Tried the windows recovery console replaced that file then another one came up bad. There's only one failed disk in the array ( another ) WTF? Dell sent me a replacement disk, I pop it in, and hey, wouldn't ya know, the PERC controller doesn't like the new disk. ( apparently the actual FW on the disks has been updated, so I have to update the perc FW, but it won't loving do it when the array is degraded ) Anyhoo, I can get the hypervisor up connect to it with v-sphere or the datastore browser. I *really* need to at least get the VMDK's off this thing, mount them, pull out all the exchange data/files/etc. Problem is, the whole goddamned thing is thick provisioned, there's 2 TB of VMDK's to get off it. I can download the data with the browser or VMWare workstation. However, even over a directly connected crossover cable with the MTU set to 9k on both ends ( also tried default ) the data comes creeping in at about 15 MB per second, and we really dont have that kind of time. I tried my damndest to get it to mount a USB disk through the unsupported root shell to no avail ( also gently caress you VMWare for making everything in /dev have such ridiculous names ) I also tried using third party tools like FastSCP but it also ran like poo poo Any other ideas? I considered trying to up it to esxi 4.1 which supposedly solves the mounting of USB drives thing, but that could also go terribly wrong. This thing is killing me Farking Bastage fucked around with this message at 01:19 on Dec 19, 2012 |
# ? Dec 19, 2012 01:14 |
|
Corvettefisher posted:following best practices you should have some form of local storage on the host by use of flash drive or HDD, so if your autodeploy server is down you can still function. I just had the mental image of someone who thinks they're being all slick and sets up diskless hosts with some form of netbooting, then puts a service critical to netbooting (say DHCP, TFTP, etc.) on that group of hosts. The look on that person's face when they're trying to come back from a power failure and realize they've screwed themselves. ivemadeahugemistake.jpg
|
# ? Dec 19, 2012 03:00 |
|
wolrah posted:I just had the mental image of someone who thinks they're being all slick and sets up diskless hosts with some form of netbooting, then puts a service critical to netbooting (say DHCP, TFTP, etc.) on that group of hosts. We had a vendor that migrated the core OS partitions to the SAN that were needed to even allow connecting to the SAN. It worked okay until they rebooted.
|
# ? Dec 19, 2012 03:27 |
|
wolrah posted:I just had the mental image of someone who thinks they're being all slick and sets up diskless hosts with some form of netbooting, then puts a service critical to netbooting (say DHCP, TFTP, etc.) on that group of hosts.
|
# ? Dec 19, 2012 04:09 |
|
Misogynist posted:I once storage vMotioned a thin-provisioned VM onto an iSCSI datastore presented by OpenIndiana running in that VM How long did that take to catch fire? I could see it trying to comply, copying the virtual disk drive in, but then the thin provisioning means the file needs to expand, which then means it needs to expand, and so on until the whole thing eats its own face.
|
# ? Dec 19, 2012 04:16 |
|
Misogynist posted:I once storage vMotioned a thin-provisioned VM onto an iSCSI datastore presented by OpenIndiana running in that VM So you did something like the Farnsworth Parabox.
|
# ? Dec 19, 2012 17:06 |
|
Goon Matchmaker posted:So you did something like the Farnsworth Parabox. We joke that once we figure out a way to vMotion a VM onto a ESX install that is running as a VM, we will be able to get rid of that pesky hardware and just run everything as software.
|
# ? Dec 19, 2012 17:38 |
|
A controller, a backplane, and another disk later, the array is back. Booted up a backup exec recovery disk and captured what data I could. Finally getting around to rebuilding this fucker.
|
# ? Dec 19, 2012 19:59 |
|
Does anyone have any experience with Vmware Site Recovery Manager? My environment currently consists of two offices: Main Seattle office: 4 ESXi hosts, virtual center server, 12 VMs, iSCSI SAN Portland Office connected via MPLS: 1 ESXi host running 4 VMs on local storage. I'm hoping to implement some sort of two way SAN replication so that we have SAN's in both offices backing up each other. Currently we're paying a 3rd party company to move our BDR backups off-site and it's not exactly cheap or backing up at the VM layer.
|
# ? Dec 19, 2012 23:26 |
|
goobernoodles posted:Does anyone have any experience with Vmware Site Recovery Manager? We run SRM quite successfully. It works as advertised, and is pretty easy to configure.
|
# ? Dec 20, 2012 00:05 |
|
Everything you can do with SRM, you can do manually via scripting, etc. if you don't care about having to maintain scripts. It's an automation tool, but it's a pretty good automation tool.
|
# ? Dec 20, 2012 00:09 |
|
SRM's good if you have supported arrays.
|
# ? Dec 20, 2012 01:12 |
|
evil_bunnY posted:SRM's good if you have supported arrays. And if you don't, you're better off looking at 3rd party replication products. The scripting part of SRM is nice, but the software replication part is harder to set up than other products out there.
|
# ? Dec 20, 2012 03:01 |
|
The first of two, of many servers to come, came in for projects in 2013. Thus begins the start of the new financial quarter, and loads of work. Thankfully the M3's are so horrendously slow at boot as the M2's were. I really wish we had a Server Thread. Dilbert As FUCK fucked around with this message at 05:56 on Dec 20, 2012 |
# ? Dec 20, 2012 05:52 |
|
Corvettefisher posted:The first of two, of many servers to come, came in for projects in 2013. Are those one of them Cisco UCS servers (I can barely make out what looks like a Cisco logo on the left bracket of each server)? They look strikingly similar to IBM System x rack-mount servers, including the hexagonal vent holes which I'm pretty sure IBM has some stupid patent on. Oh and yeah a server thread would be awesome, I'd gladly contribute a ton of info on the IBM side of things.
|
# ? Dec 20, 2012 12:30 |
|
True story a Cisco reseller once quoted me 4 times the Dell MSRP for a bunch of x86 machines.adorai posted:Yes those are rack mount Cisco UC's servers. If you have a relationship with your Cisco account rep you can just threaten to buy HP instead and they have a lot of room on the price. evil_bunnY fucked around with this message at 13:57 on Dec 20, 2012 |
# ? Dec 20, 2012 12:47 |
|
cheese-cube posted:Are those one of them Cisco UCS servers (I can barely make out what looks like a Cisco logo on the left bracket of each server)? They look strikingly similar to IBM System x rack-mount servers, including the hexagonal vent holes which I'm pretty sure IBM has some stupid patent on. Yes those are rack mount Cisco UC's servers. If you have a relationship with your Cisco account rep you can just threaten to buy HP instead and they have a lot of room on the price.
|
# ? Dec 20, 2012 13:27 |
|
evil_bunnY posted:True story a Cisco reseller once quoted me 4 times the Dell MSRP for a bunch of x86 machines. That's a UCS for you, each of these were probably in the range of 24k each. cheese-cube posted:Are those one of them Cisco UCS servers (I can barely make out what looks like a Cisco logo on the left bracket of each server)? They look strikingly similar to IBM System x rack-mount servers, including the hexagonal vent holes which I'm pretty sure IBM has some stupid patent on. UCS yest, these are actually made by Cisco unlike the 7xxx series a long time ago which were in reality debranded HP's.
|
# ? Dec 20, 2012 14:09 |
|
I don't really get the appeal of the C-Series UCS servers, unless you're doing something really edge-casey and want to hook them into FIs for the unified fabric and profile management. I have a customer that's using them in that way for a Hadoop cluster, and they're pretty thrilled with it, but again, it's an edge case.
|
# ? Dec 20, 2012 15:15 |
|
Pantology posted:I don't really get the appeal of the C-Series UCS servers, unless you're doing something really edge-casey and want to hook them into FIs for the unified fabric and profile management. I have a customer that's using them in that way for a Hadoop cluster, and they're pretty thrilled with it, but again, it's an edge case. Depends how deep the ties with cisco run at your firm, that will leverage how much you push cisco servers. Granted I see it all as hardware that I divy up and provision as needed for the client; so I generally don't care who is the vendor, so long as the NIC's aren't broadcom.
|
# ? Dec 20, 2012 15:27 |
|
Has anybody had luck deploying Windows templates in vSphere 5.1? I built a new vCenter server for 5.1, reattached hosts, and added an existing 2008 R2 template to inventory. I built a new deployment specification, and I've confirmed that all the settings are the same as the specification I used in 5.0. The VM won't join the domain, and the local admin password is not set to what is specified in the specification. This means I can't even log in to look at the event viewer to see why it didn't join the domain. Has anyone seen this?
|
# ? Dec 20, 2012 16:58 |
|
adorai posted:Yes those are rack mount Cisco UC's servers. If you have a relationship with your Cisco account rep you can just threaten to buy HP instead and they have a lot of room on the price. Haha yeah vendor "special bid prices" are the best. I was once involved in a refresh project for a customer that had an existing HP+EMC infrastructure. We put together a quote for a dual-datacentre build with purely IBM hardware. The quote ended up being just under $2mil RRP however after sending it off to IBM they knocked it down to $600k just so they could steal business from HP+EMC!
|
# ? Dec 20, 2012 17:05 |
|
Erwin posted:Has anybody had luck deploying Windows templates in vSphere 5.1? I built a new vCenter server for 5.1, reattached hosts, and added an existing 2008 R2 template to inventory. I built a new deployment specification, and I've confirmed that all the settings are the same as the specification I used in 5.0. Are you using the unattend.xml script within the host? or customization from vcenter? If you live boot a PE environment/linux you might be able to navigate to where the logs are and copy them over to a VMDK you can remount somewhere else. No problems to report. Maybe rebuild the template? Dilbert As FUCK fucked around with this message at 17:41 on Dec 20, 2012 |
# ? Dec 20, 2012 17:07 |
|
|
# ? Jun 10, 2024 11:31 |
|
adorai posted:Yes those are rack mount Cisco UC's servers. If you have a relationship with your Cisco account rep you can just threaten to buy HP instead and they have a lot of room on the price. Corvettefisher posted:Depends how deep the ties with cisco run at your firm, that will leverage how much you push cisco servers. These two seem to sum up the entire Cisco Story pretty neatly for me.
|
# ? Dec 20, 2012 17:46 |