XenServer "Warts" Exposed

6:37 PM
XenServer "Warts" Exposed -

There's always something a little surreal about blog posts on the Internet; especially as the older back around. One example is that of the blog Sysconfig which was recently dredged on twitter. In what he lists a number of things he had to fight, and details some of the steps necessary for them and this led to his final decision to migrate to solve Xen.

I readily admit that there are some pitfalls that could create problems for administrators, but as with many articles out there it lacks an important detail; the product version is not called to the front and center, leaving the reader to guess whether it is applicable.

As I sit here today, the current version is shipping XenServer 6.0.2 and I have two Master Classes XenServer with well over a thousand participants where we talk about the next Version XenServer Tampa. Going by the date in the blog, and some comments, I assume the author was running 5.6 but some of the problems related to 5.5.

For those of you query on release dates, 5.5 was released in June 09 and 5.6 in May 2010. If we look at some of the items listed, we can see the progress we have made in improving XenServer

replacement
NIC. Yup, this is a problem, but will be fixed in Tampa. The fundamental issue is that in our efforts to ensure that the conditions race in the list of Linux devices has no management interface that move and break bonds, we do support the case NIC replacement. This now been resolved; hopefully, once and for all. Oh, and the solution also works for cases where a XenServer backup is restored to a server where no equipment is common from the origin server.
  • network settings get watered member of the pool. I am pleased to say that this was a problem, but with 6.0.2 has been solved by using the reset function of the emergency network. I have personally used this a couple of times and I can say that it saved hours of reconfiguration previously needed. It is particularly useful if you use local storage for virtual machines from reinstalling XenServer is probably not a good option for those local VMs.
  • space reclamation. This used to be a big problem in the XenServer 5.x days, but since XenServer 6.0 we "coalescence of online maps", which is what makes the VM protection and recovery of an effective solution to the needs of database backup. While some storage issues are not fully addressed in Tampa, we lay the foundation to ensure that they are also a thing of the past in the very short term.
  • Code modified network bridge. We are guilty of doing it, but since XenServer 6.0 we now default to using the open virtual switch, so that our network stack is not only more powerful and flexible, but we base all our future capacity network around the OVS offers out of the box. A perfect example of this is the next support LACP Tampa obligations and containing up to 4 NICs. both are functions supported by the distributed Virtual switch (OVS) in XenServer and not implemented in the code old bridge.

At the end I salute blog entries like this to Sysconfig. It clearly states " XenServer is solid and very easy to install and use " and that " It runs and runs and runs " but also points out that, for the versions in question; " if something unexpected happens, you're seriously screwed ".

We have worked hard to improve the sustainability of XenServer and we have made significant progress in recent outings. We understand that it is not enough to build a world-class hypervisor that is running millions of mission critical workloads in hundreds of thousands of organizations worldwide; we must also ensure that the "10 Minutes to Xen" philosophy also extends to the ease of management and maintenance.

Previous
Next Post »
0 Komentar