For all those not using the VirtualCenter Update Manager, Dominic uploaded an update on his esx-autopatch.pl script. You can now also patch 3.5 hosts with this script, which can come in handy for scripted installs!
VMworld, the aftermath
It’s been a weird couple of days. I arrived in Cannes with the flu. Especially Monday, but also Tuesday, I wasn’t in top condition to say the least. Same goes for my cellphone which for some reason decided to quit on me, apologies to the people that tried to reach me and got my voice mail instead. Besides these physical and technical problems VMworld was awesome! I’ve met a lot of cool people and attended a lot of cool sessions, especially the vendor sessions were great, EMC / Brocade / Netapp. Although I must say that it would be nice if VMware adds a few real in depth sessions which will leave your brain dazzled for days, for instance on VMotion/DRS/Networking/Storage.
What surprised me most were the products Stage Manager, Lifecycle Manager and Lab Manager. They all seem the same to me. There’s a slight difference here and there but the global idea is the same. Provision VM’s with the possibility to archive them, assign them to a team or people etc. I can imagine these products will merge into one product or at least Lab Manager and Stage Manager will. Lifecycle Manager should be turned into a plugin for VirtualCenter. I don’t have any hard evidence for this or even a remote clue… but this is what I would expect to happen over the next couple of months.
One of the most promising new products that I played with definitely is Site Recovery Manager. This will save a lot of time and money for the companies that now have to do a manual fail over when a disaster occurs. One of the things that Site Recovery Manager still lacks though is the “Fail Back” button, which definitely is a feature that should be included soon. If I can fail over automatically it would be very nice if I could fail back automatically, the test button is great but everyone should really test the fail over every once in a while and with out a fail back this will be very hard.
The two new announced “products” VMsafe and vServices both leave the competition light years behind again. Both products were in the line of expectations. VMsafe was already more or less announced when VMware introduced VMCI 6 months ago:
The Virtual Machine Communication Interface (VMCI) is an infrastructure that provides fast and efficient communication between a virtual machine and the host operating system and between two or more virtual machines on the same host. The VMCI SDK facilitates development of applications that use the VMCI infrastructure.
vServices is the next step for VirtualAppliances. This gives the system engineer a new view on his system, looking at the VM from a service perspective instead of a server perspective makes sense to me, especially when you consider most services span multiple VM’s/servers these days.
Also VDI/VDM is definitely a technology VMware heavily bets on. Especially the new “offline usage” feature and the “patch one patch many” feature is something that will attract all engineers. This will not only save a lot of time updating but also solves most Corporate Laptop problems we are currently facing. The Linked Clone option sounds awesome, but what will happen when a 1000 VDI Desktops running linked clones are accessing the SAN at the same time and expanding the linked clone file at the same time. (The 09:00 clock bubble) I can only imagine this would put an outrages amount of stress on the SAN, but only time will tell I guess.
So that’s it for now, I’m gonna spend some time with my family and will be blogging full speed again the upcoming days, and testing those 16.000 users on my Exchange server of course. Maybe one of you can hook me up with suitable test servers and SAN… ๐
VMworld here we come
In a couple of hours my colleague Remco and I will be flying to VMworld, Cannes. I will be leaving my laptop at home so no blogs over the next couple of days! I want to spend as much time as possible gathering information/knowledge and making new virtual contacts! ๐
vmkping results in “syscall version mismatch (compiled: 0x7616c4e3, kernel: 0x50f63116)”
When troubleshooting a VMotion problem on an unpatched ESX 3.0.2 update 1 host today I wanted to use vmkping. This constantly resulted in the following error message:
syscall version mismatch (compiled: 0x7616c4e3, kernel: 0x50f63116)
I did not find any useful info on google so I checked the VMware knowledge base, maybe some over at VMware can read my mind or it’s a pure coincidence they just updated the following KB article, it seems to be a bug which is fixed in ESX-1002424. It’s actually mentioned in the release notes… I should read those more often I guess.
New version of ESX 3.5 and patches for 3.0.x
VMware just released a new version of the ESX 3.5 ISO according to their download site. The release notes haven’t been update yet so unfortunately I can’t tell you what’s included, but my guess would be the patches released on the 17th of January.
Latest Released Version: 3.5.0 | 02/20/08 | 64607 | 565 MB
There are no new patches available for 3.5 at the moment, but they did patch 3.0.x and the following is fixed in these patches(8 patches for 3.0.2, over 140MB):
- Security Update to the Samba Package
- Security Updates to the Python Package
- Fix for aacraid SCSI Driver Security Issue
- Virtual Machines on the ESX Server host do not power ON automatically
- ESX Server host might stop responding when a CD-ROM drive is accessed
- iSCSI LUNs in the pass-through mode might cause the ESX Server host to stop responding
- Fixes for Remote-Console Black-out, OpenSolaris Boot Delay, Duplicate Packet Issue; Support for Virtual Machines to Share Generic SCSI Devices
- Fixes for VMware Tools Installer
By the way I signed in on the patch/updates email and haven’t received anything yet… Did any of you guys receive anything?