|
The 63Tb VMFS thing is a HUGE HUGE improvement that will make me far less miserable on my database servers, file servers, Exchange server, Sharepoint server, etc. The vCenter vApp actually can manage a decent sized network now, with 500 vSphere hosts or 5,000 virtual machines. I can really see using it on my network just cause it's so easy to deploy and won't cost as much since I won't need Windows / SQL licenses. Am I crazy or are they going to be making vSphere 5.5 available for download today? I thought he was announcing GA for it but maybe he was just announcing GA for hybrid clouds.
|
#
¿
Aug 26, 2013 19:28
|
|
|
|
| # ¿ Apr 28, 2024 09:13 |
|
|
https://www.vsphereresourcekit.com/practice-exams/ is by far the best practice bank I used when I was prepping for my VCP5-DCV. It's not a brain dump, they actually tell you why the right answers are right and the wrong answers are wrong, etc. It was recommended to me by the instructor of the optimize and scale class I took. Unfortunately, it looks semi-dead.Number19 posted:So is the server appliance actually worth considering now? Sure looks like it to me. I'd like to see some testing but I'm guessing the vast majority of us aren't running >500 host > 5,0000 VM environments. The only feature I'm really dying for is support for 4 or 8 threads in Fault Tolerance. I can't imagine a single scenario where an application needs that much reliability and only uses one thread. What does everyone think of NSX? I know Google was talking about software defined networking as a huge performance boost to their network, which I can't really imagine with this sort of overlay. I'm not 100% sold on using VMware's solutions here either, I'd rather they keep building APIs for third party vendors who's technologies we're already using in appliance form. I know F5 already makes a load balancer as a vApp and I'm hoping to see more high quality integration of vApps from firms like Imperva or Cisco Ironport, etc. Would anyone really use VMware's VPN solution over Cisco, Juniper, Sonicwall, Fortinet, Checkpoint, etc? I also think it probably needs deep integration with hardware like the Nexus line but across a lot of vendors. El_Matarife fucked around with this message at 22:08 on Aug 26, 2013 |
|
#
¿
Aug 26, 2013 22:02
|
|
|
And it's not like it's particularly easy to get time when you're in a relationship either. I could myself lucky if I get 4 hours a week for professional development on top of 50-60 hours of work. Ideally, it'd be more like 8 hours a week on top of 40 hours of work if I didn't want "Why don't you ever spend time with me" questions out of the girlfriend. I really want to get my VCDX too, but it's a ton of work. Once you've got your VCP, you need to take the VCAP-DCA for administration and DCD for design. Theoretically, taking the optimize and scale class should have you ready for the VCAP-DCA but I just don't know. It's a lab exam and if you don't know how to do everything, and do it fast, you won't pass. $500 is a lot of money to wager that you know, for example, PowerCLI if it isn't something you use day to day. The VCAP-DCD doesn't have a single test bank or brain dump available from the usual suspects, and requires drawing out a ton of Visio style diagrams under a test clock that is generally agreed to make it very challenging. I'd expect 3 months of studying for each of those certs. Now we get to the fun stuff. You need to write up the design and implementation of new network in ~300-400 pages of documentation. That includes business needs and requirement gathering, a run book, and all kinds of stuff. There's a blueprint and book out there, go take a look. That's easily a 500 hour project I'd expect to take a year or eighteen months. In fact, I hear at VMworld there's still people defending designs based on 4.1, though this or early next year is probably the last possible defense session for a 4.1 design. There's ~150 VCDXs in the world, and there's a good reason why. Even getting either of the VCAPs is a big deal, since there's maybe 10,000 of those. I think the VCDX is going to be a "write your own ticket, anywhere in the world" cert for anyone, but there's a drat good chance you're going to be a consultant or a VMware employee. In fact, I hear VMware employs like 60% of them, VARs / consulting employees another 20%, and maybe 20% are VP level guys at big shops. I really laugh when I see these head hunting shops posting "VCP, VCAP or VCDX" for sysadmin jobs. VCAP is a senior level qualification and VCDX is really something for either consulting or management / senior architect. El_Matarife fucked around with this message at 23:38 on Aug 29, 2013 |
|
#
¿
Aug 29, 2013 23:35
|
|
|
Dilbert As gently caress posted:Sounds like lots of fun! Within the next few years, it will make a CCIE look like something from a box of Crackerjacks. (At least the DCV one will, I don't think we know if VMware cloud or OpenStack will win that battle. And desktop is a total blank slate.) The best part is, once you have the VCDX, you can upgrade to the next version by just taking the DCD test again. As far as VMware versus OpenStack goes, I think OpenStack does have a lead in the some areas right now, especially on the Linux side with everyone combining OpenStack and Puppet. And they've definitely got a deployment advantage with Rackspace onboard among other big names and tons of startups. However, VMware's hypervisor (and VMware Tools) is so far above and beyond HyperV, KVM, and Xen that I can't see anything but it remaining the standard as long as people still need Windows VMs. Worst case, VMware just wholesale rebrands their own version of OpenStack the way that Oracle does with Xen. Still, the whole "private cloud" space is still in its infancy. I think we're really at where hypervisors were in 2006, where it just got good enough to start getting mass deployment.
|
|
#
¿
Aug 30, 2013 04:43
|
|
|
evol262 posted:This is, in a word, poo poo. Complete and utter bollocks. VMkernel is not "far and above" HyperV, KVM, or Xen, and specvirt (and a million other benchmarks) agree. It's a better all-arounder than any of the rest (Xen and HyperV can beat it at I/O, KVM at CPU and memory performance as well as density). It's that VAAI, dvSwitches, svMotion, and all the rest of the VMware sauce have no real comparisons in RHEV, HyperV, or XenServer. We (RHEV) can offer some of it, but not all. HyperV is the same. But if you wanted to compare KVM vs ESXi/VMkernel, you'd be surprised at how little performance difference there is, because essentially the entire Linux ecosystem has standardized on it (regular contributors include Google, Intel, IBM, and AMD). I'm sorry, I should have specified the ecosystem / features of VMware are far and away better for the reasons you just listed. I had no idea what the benchmarks were but now I'm curious. Would you say all hypervisors sit within a 10% band of each other on every metric? And how does VMware Transparent Page Sharing factor in? Last I heard, no one else had a similar feature. 1000101 posted:All in all it was a fun experience (I enjoyed the first and 3rd portions of the defense the most.) Was pretty jazzed to make it through completely and I've made sure to take every VCAP-DCD exam since passing (I started with 3.X back when the program was first released.) Wow, I met VCDX #69 about four weeks ago and he did his on ESX 5 I think. You must be VCDX number what, <20? Are there still no public VCDX designs that passed for reference, beyond the blueprint? Pantology posted:Strongly encouraged but not required, though even a real project may require some fictional elaboration to map to the blueprint. I believe that no one has ever passed with an entirely fictional design even though it's allowed, so "strongly discouraged" is putting it lightly, but it's really, really common for people to take a real design and add some fictional elements to make it touch every necessary area. El_Matarife fucked around with this message at 21:42 on Aug 30, 2013 |
|
#
¿
Aug 30, 2013 21:30
|
|
|
Let's talk about this from a business needs perspective for a minute. Assume for a minute you're not building massive scale web or similar applications on top of open source. Let's say you're at a Windows + IIS + Java + MS SQL 2008R2 shop with all kind of proprietary software in the mix. What's Open Stack going to get you in this scenario? Does OpenStack have the same catalog / policy based rapid deployment capabilities of the vCloud suite for "pets" not "cattle"? My sense of VMware's goal with vCloud is to make it easy to deploy new VMs from templates with disaster recovery, server hardening, network configuration, etc all attached and super easy. And now I understand OpenStack is all about rapidly spinning up or killing identical images in an API driven way that programmers can easily work with. Would you agree that cloud computing and virtualization are totally different models? If so, would you then agree that vCloud way more of a virtualization product and OpenStack is a cloud computing product?
|
|
#
¿
Sep 4, 2013 18:52
|
|
|
I've got DR4000s with Backup Exec 2012 (Ugh) and it's a huge pain in the rear end. The DR4000s are OK, I guess, but the dedupe capability rests on a specialized driver and licensing Symantec PureDisk (that wasn't purchased by the previous team) so I've basically got a huge CIFS share. Everything I hear about Data Domain makes it sound truly fantastic, especially if paired with Avamar. My biggest complaint is the DR4000 web interface times out sometimes when making changes or the changes don't fully commit, which seems really buggy. I'm about a year behind in software upgrades so maybe they fixed the issue. Replication speeds didn't seem really great either. You could also consider vRanger now that Dell owns Quest who bought VizionCore. vRanger was like THE VMware backup product a few years ago but I have no idea how time has treated it.Misogynist posted:Pets don't have rapid deployment, at least not in the sense of what constitutes "rapid deployment" nowadays. That's precisely what makes them pets, and precisely why most organizations want cattle instead. This is why vCloud Director was such an apocryphal product: it adds a lot of complexity but doesn't fundamentally change the way that people get work done. Well, rapid in terms of hours instead of a day or two. If you look at your standard piece of commercial software, you need what, at least two web servers, two database servers, maybe two app servers? And then you may need them hardened for regulatory / compliance reasons, you need them replicating to your DR site, you want your shares / quotas / pools done, etc. If they're killing vCloud Director and putting all those capabilities in vCenter, that would be fantastic. Right now, unless you've got a bunch of advanced vCLI Powershell scripts, you've got to run the deploy VM from template wizard six times, then go in and run vCenter Configuration Manager against the new VMs to harden them, then add them to SRM, adjust all the pool and share and quota settings, etc. It's maybe an hour per VM and like 5 different control panels to touch, assuming you've got VAAI with a good SAN and your templates clone fast. Deploying an application might cost you a whole day in vCenter before you can even start installing software. Misogynist posted:If what you're looking for is something that incrementally enhances your productivity and takes some of the repetitive steps out of provisioning systems the exact same way you would do them with bare vSphere, vCloud Director is the right tool for the job. It's not the approach or the market that OpenStack or AWS are reaching for. OpenStack and AWS don't rely on images, per se, but they're designed for systems that function with a high degree of automation and minimal interaction. The catalog based sandbox deployment they use in the VMware Hands On Labs site is pretty much exactly what I'd like to do for dev /QA teams. And I'd love fancier templating with more attached policies and guest customization steps like "Install IIS" or "Enable this server role". If that's going to be in vCenter going forward, great, I'll save a ton of money not buying the full cloud suite.
|
|
#
¿
Sep 4, 2013 23:42
|
|
|
evol262 posted:This is why you're deploying sysprepped clones, right? With RunOnce deployment scripts? Or tied into SCCM? Honestly, I thought vCloud Director was going to become a full fledged configuration management system to replace SCCM and complement vCenter Configuration Manager. I really don't want two separate suites from Microsoft and VMware managing the same set of VMs and I think we're pretty set on vCenter Configuration Manager since we need something for hosts and guests. What other config managament products are worth looking into on Windows? Do you have any good source for off the shelf deployment scripts?
|
|
#
¿
Sep 5, 2013 00:46
|
|
|
Oh my god, why didn't anyone ever tell me about RVTools? http://www.robware.net/ This makes VCOPS look like hot garbage. I just pulled a list of all the NICs on my VMs to look for E1000s, I pulled a list of raw physical disks, and it's got a health check that found some really interesting issues.
|
|
#
¿
Sep 6, 2013 20:11
|
|
|
Yellow Bricks and vBrownBag really should both be in the OP too.
|
|
#
¿
Sep 6, 2013 20:38
|
|
|
Dilbert As gently caress posted:Thinking about doing some vmugs or vSAMugs on Wednesday night anyone interested? Are you thinking Google Hangouts or what? I don't suppose someone has a Webex / GoTo Meeting account from work we can borrow?
|
|
#
¿
Sep 8, 2013 19:54
|
|
|
http://www.datacenterknowledge.com/archives/2013/09/11/cisco-acquires-whiptail-for-415-million/ Sounds like Cisco is a believer in distributed storage on compute. If vSAN and CloudStack Swift / Cinder are the future, the big monolithic SAN guys should be worried. I always figured Cisco would buy NetApp or another mid-tier SAN vendor to complete their compute / network / storage stack. This also dovetails nicely with the read cache support in VMware 5.5, though I suspect this might be at the chassis / FEX level rather than individual blades. Or possibly they want to build a Dell VRTX style chassis with a "SAN" integrated as a blade. El_Matarife fucked around with this message at 18:43 on Sep 11, 2013 |
|
#
¿
Sep 11, 2013 18:26
|
|
|
The HA / FT stuff is a very common VMware job question so take note. HA basically just means "Can this VM be rebooted on the other hosts in the cluster if one host fails?", so you need shared storage, some leftover resources on the hosts, etc. I recommend setting an admission policy for the cluster, but use percent reserved instead of host failures tolerated because host failure calculations are really weird. It definitely helps to use memory reservations too. That way, you're sure you can have HA / DRS move VMs without causing memory contention, and with a reservation you no longer need a VSWP file which can save some space on the SAN.
|
|
#
¿
Sep 16, 2013 20:19
|
|
|
Yeah, on almost every single small network I can think of, memory and disk size are your limiting factors. You need a LOT of users or a seriously intense application to peg out a modern CPU. Best practices, you want the minimum amount of vCPUs possible. Try to keep VM vCPU usage between 30-80%. If you go above 80%, add another vCPU. Generally, adding unused vCPUs adds overhead in memory and can result in some weird contention issues from CPU scheduling. Network contention can be an issue, but usually only if you're not using separate NICs for iSCSI / FCOE and vMotion. Generally, they should all be segregated unless you've got 10Gb. El_Matarife fucked around with this message at 20:25 on Sep 16, 2013 |
|
#
¿
Sep 16, 2013 20:22
|
|
|
skipdogg posted:So it turns out years and years ago when we first started virtualizing we bought 2 acceleration kits, so we have 2 licenses for vCenter. We have 2 main datacenters, one on East Coast, one on the West. vCenter currently lives on the west coast. Is it possible/recommended/feasible to setup some sort of failover vCenter in the east coast DC in case west coast drops for some reason? I'm going to dig through my books and google, but I figured what the hell, I'll ask here too. Honestly, I'd run a vCenter on the East Coast and one on the West Coast and put them in Linked Mode.
|
|
#
¿
Sep 19, 2013 02:35
|
|
|
http://blogs.vmware.com/kb/2013/09/vsphere-5-5-is-here-kbs-you-need-to-know-about.html Here's a bunch of KBs you should take a look at. VCOPS 5.7.2 available for vSphere 5.5 too: https://www.vmware.com/support/pubs/vcops-pubs.html https://www.vmware.com/support/vcops/doc/vcops-572-vapp-release-notes.html Still no IE10 support, still says only support for Chrome 24 and 25. Hell, still says only support for Configuration Manager 5.6 which is weird with 5.7 having been two months now. SRM 5.5 https://www.vmware.com/support/srm/srm-releasenotes-5-5-0.html El_Matarife fucked around with this message at 16:46 on Sep 23, 2013 |
|
#
¿
Sep 23, 2013 16:01
|
|
|
three posted:Correct, but very few people use FT due to its requirements and limitations (only 1 vCPU). And apparently they want the FT logging split out onto its own set of NICs too. Two for VM traffic, two for management / vMotion, two for storage, and now two for FT logging? I mean sure, if you've got two 10GbE NICs you're probably fine to combine everything but way to make FT even less attractive.
|
|
#
¿
Sep 26, 2013 20:31
|
|
|
parid posted:Think about what it has to do. It has to mirror ram. Do you really want that sharing with anything else? Even if you put it on 10gig you will want net I/O control/QOS to help prevent it from effecting other functions. Yeah, I get that it's pretty technically crazy to pull off. I just wish they'd use some crazy memory compression technique or something. Also, perhaps allow us to cluster over infiniband? Wouldn't give us any geographic diversity but infiniband would have enough bandwidth to lockstep a serious server with multiple threads. I'm really curious what you have that needs uptime so bad you need that kind of cluster but is light enough to survive on a single thread.
|
|
#
¿
Sep 27, 2013 16:45
|
|
|
EoRaptor posted:vCenter 5.1 to 5.5 upgrade wiped all my users and roles out. I only had 3, so not a big deal, but still annoying. I actually had a similar issue last week with my 5.1 upgrade, apparently you can't use local Active Directory accounts with single sign on and linked mode. And it also broke any VMware permissions based on local permissions, so all our users who were VMware admins from being local admins on the vCenter box broke.
|
|
#
¿
Sep 27, 2013 23:24
|
|
|
three posted:Here is a good article from Chris Wahl on CPU Hot Plug effects: http://wahlnetwork.com/2013/08/21/cpu-hot-plug-effects-on-vsphere-vcpu-count/ If hot add CPU disables vNUMA, I wouldn't use it on any large VMs. (And your large VMs are most likely to be your most sensitive VMs anyway, right) I can live with an extra memory hit from allowing hot memory add if I'd get to rightsize the vast majority of our VMs. I think hot CPU add is a lot trickier with per socket licenses potentially causing problems.
|
|
#
¿
Oct 1, 2013 22:24
|
|
|
Dilbert As gently caress posted:Is dell's 5.5 ISO out? usually it takes a while for them to release it. Last I looked at a Dell VMware ISO, they listed the source for all their driver packages as VMware.com which made me think I could probably take an off the shelf 5.1.1157734 ISO and use it after integrating PowerPath. Am I crazy?
|
|
#
¿
Oct 2, 2013 23:07
|
|
|
FISHMANPET posted:I believe the cert changes only when a new major release comes out. So VCP5 is good until VCP6 comes out, at which point should be able to take an upgrade course. I thought you weren't required to take an upgrade class. That price is totally completely ridiculous for what it's worth, the optimize and scale class is far cheaper and preps you for VCAP-DCA. Speaking of, I just took my first prep exam from the VCAP-DCD Official Cert guide and got a 392 with a passing score of 300. I assume now I can tackle the multiple choice portion, but how bad is the Visio style network diagramming going to hurt? I hear it usually takes two tries to pass.
|
|
#
¿
Oct 3, 2013 01:47
|
|
|
Optimize and scale class is heavily recommended for VCAP-DCA, and it qualifies you for the DCV, so if you don't have your VCP class and have enough experience to install / configure VMware, I'd take that class. For me, it was cheaper than Fast Track or ICM. It runs you through a bunch of command line stuff you probably don't use on a day to day basis like ESXTop or PowerCLI. VMware recommends two classes for VCAP DCD but I hear just reading http://www.amazon.com/VMware-vSphere-Design-Forbes-Guthrie/dp/1118407911 a few time might get you through. There's also an official cert guide now: http://www.amazon.com/VCAP5-DCD-Off...ywords=vcap+dcd thoguh you might want to order direct from Pearson since their eBook version has extra test prep. (The book does come with a 70% off code for the ebook and extra test prep.)
|
|
#
¿
Oct 4, 2013 21:53
|
|
|
If you've got >1 site, Linked Mode so you can manage more than one vCenter with one sign in is really nice and bordering on mandatory. The vCenter appliance can't do Linked Mode since requires Active Directory Application Mode (ADAM) and the VCSA is Linux. There's also some restrictions around VMware View so don't use it if you're planning on virtual desktops anytime soon.
|
|
#
¿
Oct 5, 2013 23:44
|
|
|
I believe there's currently no migration path. There's a lot of interest though, now that it's workable with larger environments.
|
|
#
¿
Oct 7, 2013 15:45
|
|
|
Dilbert As gently caress posted:I mean poo poo any advice is useful I want to make best of it! I WANT TO BE A VCDX! Honestly, go work for a partner or VAR. I thought I was going to get a VCDX out of my current job's brand new meganetwork I am going to be doing a lot of implementation on, but they won't let me be involved in the project planning or procurement process which is all happening at the VP level. They let me full upgrade our vBlock 300 though which was a very nice boost to my VCAP-DCA plans. If you're at a VAR or partner, you'll get thrown into network design tasks every day and eventually you take lead on one and have the perfect VCDX project. Helps to have a VCDX mentor there too. bull3964 posted:Really? SSO doesn't query DNS for the AD controllers and hard codes them in when the service was installed? That would explain it since we retired all our old 2003 domain controllers at the same time we did the domain level upgrade. VUM IPs are hardcoded into vCenter too, so if you retire a vCenter I think you have to go into ADSI edit to fix it. Edit: Two new interesting VMUG Webcasts: http://www.vmug.com/p/cm/ld/fid=38&source=5 One on performance tuning and one on upgrading to 5.5, both hour long lunch things in the next six weeks. (And an EMC backup event I'm sure they're going to pitch Data Domain / Avamar the whole time) El_Matarife fucked around with this message at 22:58 on Oct 10, 2013 |
|
#
¿
Oct 10, 2013 22:48
|
|
|
I'm shocked Apple won't just team up with Oracle and announce OSX for SUN x86 servers at ~$5,000 a socket. It'd take care of the business / enterprise space and Jobs was really, really close friends with Larry Ellison. So I apparently managed to install both my vCenter SSOs as primaries when doing my 5.1 upgrade, I'm going to have to completely uninstall and reinstall the DR site SSO. Total pain, I can't believe I missed that checkbox. The error message in the JoinTool log is spectacularly worthess too: Operation "Join instance XXXXX" failed: : Action: Join Instance Action: Prepare for Join Problem: Remote VC's XXXXXX SSO domain ID is does not match."
|
|
#
¿
Oct 16, 2013 20:11
|
|
|
Does VMware have a GA mailing list or something to let us know when new downloads land? vCOPS 5.8 looks awesome and I really want to deploy it ASAP.
|
|
#
¿
Oct 24, 2013 21:09
|
|
|
Java 7 Update 45 breaks the Cisco UCS control panel. http://tools.cisco.com/Support/BugToolKit/search/getBugDetails.do?method=fetchBugDetails&bugId=CSCuj84421 Workaround according to Cisco? Use Java version 7 update 40 or below. There's no patch yet and I assume it's a few weeks out at least. Earlier releases are available from: http://www.oracle.com/technetwork/java/javase/downloads/java-archive-downloads-javase7-521261.html
|
|
#
¿
Oct 25, 2013 23:36
|
|
|
The HTML5 client is only on Mac right now, but apparently you can use the HTML5 console on every OS in a rather unsupported manner. http://www.virtuallyghetto.com/2013/09/how-to-generate-pre-authenticated-html5.html
|
|
#
¿
Oct 26, 2013 18:23
|
|
|
Posted this a few weeks ago but I think it's worth a mention if you're diving in to UCS right now: Java 7 Update 45 breaks the Cisco UCS console: http://tools.cisco.com/Support/BugToolKit/search/getBugDetails.do?method=fetchBugDetails&bugId=CSCuj84421 Workaround according to Cisco? Use Java version 7 update 40 or below. Earlier releases are available from: http://www.oracle.com/technetwork/java/javase/downloads/java-archive-downloads-javase7-521261.html Anyway, I think the blade profile capabilities of Cisco are pretty drat nice. Configure your BIOS and firmware settings, from a Java webpage, click apply and trigger a reboot and boom, whole new BIOS, NIC firmware, or BIOS settings, or anything. Lose a blade? Pop a hot spare in that slot and it will come up with the EXACT same settings / environment as the dead one. The OS or VMware will never even know the difference. You've got pools for MAC address, management IP, WWN, everything, so deploying additional blades is really easy. They've got an API for some seriously heavy scripting capabilities and the command line controls are pretty close to what you get on a Nexus or MDS switch so your network guys will like it. I will say the KVM and ISO mounting sucks compared to Dell DRAC 6 / 7 though. And the out of band RAID management capabilities blow for the Cisco C series rackmount servers. In fact, I couldn't get them to tell me if a RAID rebuilt properly or not, I'd have to reboot and check the BIOS. As always, I'm sure all the vendors have big warchests to go out and win deals from competitors. If you press them all hard with pricing from HP, Dell, and Cisco, you can get a pretty good deal. I'd be shocked if the total price band between all three was more than 20%. hackedaccount posted:Nah because organizations that need true fault tolerance would drop the cash on something like these (and they support vSphere too). Dilbert As gently caress posted:Don't worry guys vSMP FT is coming soon******* Our company ripped our Stratus FT boxes out, too many problems with them. We decided we could live with a 30 second VM reboot. Anyone want them? I can let them go for cheap. I've been trying to find a decent reseller to get rid of them. You know, I've been thinking about FT lately. Mellanox has a $700 twin 40 GbE / 56Gb Infiniband NIC for like $700. If they make crossover cables, you could pipe 80-112Gb between two boxes which would sure as hell fix any bandwidth problems to FT with multiple threads. They've got switches starting at $16K for ~30 ports, but no SFPs and I bet they're going to charge you for port licenses and all kinds of other hidden fees / prices since that's just too cheap to believe. I have a feeling Infiniband is eventually going to be a pretty popular interconnect on top of rack switches, especially with storage getting pushed to the "edges" of your network. Sooner or later, we'll all have like 16U of storage and 24U of compute, and 2U of networking, per rack. No more monolithic racks of all storage or all compute. El_Matarife fucked around with this message at 17:56 on Nov 10, 2013 |
|
#
¿
Nov 10, 2013 17:46
|
|
|
The web client in 5.1 is slightly slower with everything. Anything you click on there's a distinct wait time of anywhere from a half second up to two or three seconds. It's just enough friction to really slow you down and make it feel off. I wish VMware would make a new vSphere client from the View client, with all the console stuff PCOIP based.
|
|
#
¿
Jan 10, 2014 21:06
|
|
|
DevNull posted:The webMKS (your console into the VM) is already HTML5, so at least part of it is done. What benefit would using PCoIP give? VMware console is terribly slow and generally awful on low bandwidth connections. It's got quite the delay on 6 Mbps DSL and some cellular hotspot or public / hotel WiFi use is unbearable. I understand not everyone has or can install VMware Tools at all times, but for VMs that have it (which should be nearly 100% of your environment or you're doin' it wrong), PCOIP would be a nice option. Anyway, apparently VCDX-001 & 002 are barnstorming around the country giving free VCAP-DCA, DCD and VCDX bootcamps. I found out about it through my local VMUG, but maybe you can get a heads up from your training or sales reps. El_Matarife fucked around with this message at 23:17 on Jan 13, 2014 |
|
#
¿
Jan 13, 2014 23:15
|
|
|
5.1 vCenter build 1123961 with ESXi build 1157734, and it's a little better than 5 GA and 4.1 were but still seems pokey. I've never sat down to measure the traffic, admittedly.
El_Matarife fucked around with this message at 23:31 on Jan 13, 2014 |
|
#
¿
Jan 13, 2014 23:26
|
|
|
ESXi 4.0/4.1 is moving out of general availability on May 21st, so I would HIGHLY recommend not building anything new on it. https://www.vmware.com/files/pdf/support/Product-Lifecycle-Matrix.pdf No more security patches or bug fixes can put you in a bad way, especially if you to pass audits or comply with standards like PCI or SOX. https://www.vmware.com/support/policies/lifecycle.html And ESX(note no i here) 4.0 is 100% dead in May, they'll no longer offer any support for it at all. https://www.vmware.com/support/esx/?docType=kc&externalId=2039567_draft Don't forget to check the highest supported EVC levels on all that mismatched hardware if you're building a cluster. And you can never mix and match AMD and Intel. El_Matarife fucked around with this message at 00:02 on Jan 28, 2014 |
|
#
¿
Jan 27, 2014 23:59
|
|
|
Mausi posted:(complete rebuild, yay! but I don't any hands on with the intervening versions) VMware announced a VCP5-DCV test based on 5.5 last week, so that's less of an issue. http://blogs.vmware.com/education/2014/01/new-vsphere-5-5-based-exam-for-vcp5-dcv-available.html I'm imagining 5.5 is going to have a pretty rapid uptake from 5.0 and 5.1 shops. Overall, I find it WAY easier to keep VMware up to date than to move versions of Windows Server for instance.
|
|
#
¿
Jan 28, 2014 00:57
|
|
|
Dilbert As gently caress posted:Nope it's super easy, but you need to go with 5.0 or later, your vCenter can be 5.5 and hosts rest at 4.x and later I want to point out even though VCSA 5.5 is WAY better, it still can't do Linked Mode (Necessary if you have >1 site or vCenter) and it won't do vSphere Update Manager which is the easiest and fastest way to upgrade your hosts. (Easy but more manually intensive to do with a CD going from box to box or something similar.) If you can afford a new license and have the time to build it out right, I still recommend a Windows base, but if you're lacking in money and time, VCSA is the way to go.
|
|
#
¿
Jan 28, 2014 22:30
|
|
|
You can still do it without Linked Mode, but you'll have two separate vCenters to login to when managing things, you will have to enter licenses separately on both sites and you can't do cold migrations without downloading and uploading a VMDK file. And I don't think you can get VMware Site Recovery Manager working at all, if you own that.
|
|
#
¿
Jan 28, 2014 23:01
|
|
|
Dilbert As gently caress posted:Welp congrats to the new VCDX's! They announced results? I thought the grading rubric usually takes a week.
|
|
#
¿
Feb 15, 2014 00:05
|
|
|
|
| # ¿ Apr 28, 2024 09:13 |
|
|
I'm running a study group tomorrow night for the VCAP5-DCA. We're going to start by covering the three objectives very few people use in real life: Autodeploy, Kickstart files, and Private VLANs. I'm using Jason Langer's VCAP5-DCA study guide http://www.virtuallanger.com/vcap-dca-5/ and Chris Wahl's DCA study guide http://wahlnetwork.com/2012/07/02/the-vcap5-dca-study-sheet/ plus a hands on lab that covers some autodeploy. Anyone have any other good suggestions? I was going to recommend readings in Scott Lowe's Mastering vSphere 5 book too. (And if you want to attend, PM me.)
|
|
#
¿
Mar 11, 2014 03:23
|
|