vSphere5

0

After more than two years vSphere5 is ready! This will be a great release with many new features.

I look forward to working with Storage DRS, the new Virtual Storage Appliance, VMFS5 etc.

There will also be some changes regarding the licensing. vRAM is the new word. Some customers will need to buy more licenses but I think most will manage with what they have. Pretty much so sum it up Enterprise Plus gives you 48GB RAM pr. license. Take this into account when sizing your servers. a Dual Socket system with 96GB RAM would make sense.

Truncate vpx_event table in vCenter database

7
One of my clients was having issues with his vCenter database. He has an environment of 5 hosts and approximately 50 virtual machines and the database was filling up rather quickly. The size was 64GB!! Normally for an installation of this size I would expect to see a database of 2GB data. The first thing you look into is transaction logging but the database was running in simple mode so that was not the problem.

 

We then ran  a SQL script that showed the biggest tables in the database. The two biggest were vpx_event and vpx_even_arg. They had more than 76 milllion rows each! I used some time on google and found the following document that described how to truncate the vpx_event_arg table. Unfortunately you cannot just truncate the vpx_event table because it has some constraints.

 

By looking at the entries in the vpx_event and vpx_event_arg tables we discovered that all of the entries came from the vMA (vSphere Management Assistant). The vMA was collecting logs from the ESXi servers and from the vCenter server. We issued the command “vilogger disable –server vcenter.acme.org” and everything went back to normal.

 

The following script supplied to another customer from VMware support did the trick. Remember to backup your vcenter database just in case.

 

1. Stop the vCenter Server Service.
2. Run the following SQL against the database (I recommend you take a backup first):

 

<---script start--->
alter table VPX_EVENT_ARG drop constraint FK_VPX_EVENT_ARG_REF_EVENT, FK_VPX_EVENT_ARG_REF_ENTITY alter table VPX_ENTITY_LAST_EVENT drop constraint FK_VPX_LAST_EVENT_EVENT

truncate table VPX_TASK

truncate table VPX_ENTITY_LAST_EVENT

truncate table VPX_EVENT

truncate table VPX_EVENT_ARG

alter table VPX_EVENT_ARG add

constraint FK_VPX_EVENT_ARG_REF_EVENT foreign key(EVENT_ID) references VPX_EVENT (EVENT_ID) on delete cascade, constraint FK_VPX_EVENT_ARG_REF_ENTITY foreign key (OBJ_TYPE) references VPX_OBJECT_TYPE (ID)

alter table VPX_ENTITY_LAST_EVENT add

constraint FK_VPX_LAST_EVENT_EVENT foreign key(LAST_EVENT_ID) references VPX_EVENT (EVENT_ID) on delete cascade
<---script end--->
3. restart the vCenter Server Service.

Password complexity ESXi

0

I have a lot of ESXi servers running in my lab and prefer to have a very simple password. This password could be “vmware”. A six letter word all in lower case. Not the most secure password, but in my lab environment it is not a worry.

The password complexity is defined in the file: “/etc/pam.d/system-auth” you need to look at line 12:  ”password   requisite    /lib/security/$ISA/pam_passwdqc.so retry=3 min=8,8,8,7,6″ This is actually what defines the password complexity. The way to interpret “8,8,8,7,6″ is the following: The first 8 is how long the password has to be if we only use a single character set (lower case, upper case, digtigts, other characters.). The second 8 is if we use two character classes The third is for password phrases and the last two is for 3 and 4 character classes password.

I want to use the password “vmware” a single character word with a length of 6. To accomplish this we change line 10 to one of the following:

  1. password   requisite    /lib/security/$ISA/pam_passwdqc.so retry=3 min=6,6,6,6,6
  2. password   requisite    /lib/security/$ISA/pam_passwdqc.so retry=3 min=8,8,8,7,6 enforce=none

The change will take affect immediately. No need to restart any services. Now go ahead and change your password with the “passwd” command.

The only thing you have to worry about is that the system-auth is not persistent through reboots. Your password is persistent but the file is not. To make sure the system-auth file is backed up you need to:

chmod +t /etc/pam.d/system-auth

maxthroughput_io_second

Performance: RDM vs. VMFS

4

Last night on twitter there was a discussion about RDM vs. VMFS performance. I have always told everybody that RDM did not have an performance advantage over a VMDK file on a VMFS volume. I have refered people to the following document by VMware: Performance Characteristics of VMFS and RDM but I have never done my own testing until now :-)

My setup is ESXi 4.1 build 260247 with 4gb fibre channel HBA connected to a SUN Storagetek 2540 SAN. I did the test on a 10GB LUN located on fibre channel disks. The test was done with Iometer on a virtual machine running Windows 2003 SP2. I performed all tests on the 10GB LUN as either: Physical RDM, Virtual RDM, Lazy Zeroed VMDK on a VMFS and Eager Zeroed VMDK on a VMFS.

I configured Iometer with settings from the unoffical storage performance thread on VMware communites and ran two tests: Maxthroughput and Real life. Here are my results:

Conclusion: VMFS and RDM have similar performance. Don’t choose RDM for performance.

FastTracks are back

0

I have some great news for those of you who are looking for VMware training. The Fast Track courses are back. So far three FastTrack courses have been announced:

  • VMware vSphere : Automation FastTrack (Install Configure Manage & Automation with PowerCLI)
  • VMware vSphere : Advanced FastTrack (Troubleshooting & Manage for Performance)
  • View DeskTop Fast Track (VIEW ICM & VIEW Design & ThinApp)

The Fast Track courses are 5 days each with extended hours. It is basically several courses made into one. If you want as much as possible in the least amount of days Fast Track is for you.

I am really psyched about this and looking forward to delivering Arrow ECS’s first Advanced FastTrack when it is available. It looks like it could be late April or early May.

vSphere 4.1 performance troubleshooting guide

0

VMware just published a new version of their performance troubleshooting guide. This is a great walk through to discover cpu, memory, storage og what ever problem you may be experiencing.

Check it out here: http://communities.vmware.com/docs/DOC-14905

If you need some hand on with a lot of these performance graphs and ESXTOP consider attending the VMware Manage for Performance Class.

vSphere 4.1 update 1 released

0

Great news. Another version of vSphere has just been released. It is time to update your vCenter server and ESX/ESXi hosts.

This one brings new OS support and some new features. Nothing major. Check it out here:

http://downloads.vmware.com/d/info/datacenter_downloads/vmware_vsphere_4/4_0

esxtop

CPU Ready 1000 ms. equals 5%

1

One of the key performance counters in a vSphere enviroment is: CPU ready (%rdy in ESXTOP)

CPU ready is the time a virtual CPU is ready to run but is not being scheduled on a physical CPU. This would under normal circumstances indicate that there is not enough physical CPU resources on an ESX/ESXi host. This is the first go-to counter when your users complain about bad performance.

The CPU ready counter is accessible from the vSphere Client and from ESXTOP. I have made two screenshots showing the a virtual machine and its ready time:

vCenter Performance Graphs  (Value 1035 milliseconeds)

ESXTOP (value 5.38%)

What we see is a virtual machine with a ready time of 1035 ms. or 5.38%. These numbers are actually telling us the same thing. When we are using the performance graphs the graph updates every 20 second (or 20,000 millisecond). With a ready time of 1035 ms. we can change it to a percentage: (1035 ms. x 100) / 20000 ms = 5,175%

To be able to interprept ready times it is essential to know the relationship between the percentage of ESXTOP and ms. of the Performance Graphs. You are seeing the same numbers. One is in milliseconds the other is a percentage.

1% = 200 ms.

5% = 1,000 ms.

10% = 2,000 ms.

100% = 20,000 ms.

In general you want to see virtual machines with a ready time lower than 1000 ms. or 5%.

Common ESXi upgrade questions

0

Today I upgraded another customer from ESX 4.0 to ESXi 4.1. He attended an Install, Configure Manage course that had me as an instructor just one month ago. I told the class that ESX was being phased out from the next major release. He took notice of that and when planning his upgrade he wanted to change from ESX to ESXi. I have always been very Pro-ESX. But from ESXi 4.1 I would never pick it unless I had some kind of third party application that relied on the service console.

When changing to ESXi here are some common questions I get:

Q: I can’t live without the Service Console. What do I do?

A: Well for most you are probably better off without it. ESXi requires a lot less patches and is more secury. No service console running using memory and cpu resources on your ESXi host.

Q: What about scripting and command line interface?

A: Yes! You have two options here. Use the “vSphere Command Line Interface” or “vSphere PowerCLI”. You can do a lot of scripting. I would recommend everyone to import the “vSphere Management Assistant” virtual machine. It comes bundled with the “vSphere Command Line Interface”

Q: I thought ESXi was free. I want the full VMware experience?

A: ESXi is free if you want to use it as a stand alone host without vCenter. But if you want to use it with all the cool features like vMotion, HA, vCenter, FT, DRS etc. you need to license it. ESXi is using the same licenses as ESX. So if you are running ESX today you already have the licenses for ESXi.

VCAP-DCD

0

I just applied for the VCAP-DCD exam. Hopefully it wont be long before I get approved. I want to take the test in January og February. When that is done I am going for the VCDX in 2011! :-)

Go to Top