VMware

Blog Posts

Blog Posts: 1,664

Recent Blog Posts

Statistics: Blogs: 785 Blog Posts: 1,664   1 ... 107 108 109 110 111 Previous Next Items per page
0

PICT0968.JPG

Now that's what I call a processor. When can I fit one of these onto my Proliant?

0 Comments Permalink
3

This script will list all snapshots for guests assigned to a specific host.

3 Comments Permalink
0

Here is a quick 2 line change to enable ssh for root.

cp /etc/ssh/sshd_config /etc/ssh/sshd_config.orig
sed 's/PermitRootLogin no/PermitRootLogin yes/' /etc/ssh/sshd_config > /etc/ssh/sshd_config

If you want the changes immediately before a reboot also execute:
service sshd restart

Again, very handy in kickstart.

0 Comments Permalink
0

One of the initial issues I had with Virtual Center is that it did not always show the correct firewall settings. The command esxcfg-firewall seemed to be the authority on open ports. By executing 'service mgmt-vmware restart' at the command prompt it syncs up VC with esxcfg-firewall. If you decide to script this and plan on having commands after this, you will want to add a 'sleep 20' command on the next line so that the VM management service has a chance to finish the sync.

0 Comments Permalink
1

I hacked the installation script for VmPerl so it would auto-install with the defaults. Have fun!

1 Comments Permalink
0

TSM v5.4.1.0 Installation
1. Execute: rpm -ivh TIVsm-API.i386.rpm
2. Execute: rpm -ivh TIVsm-BA.i386.rpm
3. Create/edit config files.

Configuration
The way I like to do it is to copy my config files from a static source and then edit them on the fly with a few replace commands. This facilitates kickstart deployment. I have included my sample config files below.

Commands to execute for tsm server and node replacement
sed -i.old -e 's/tcpserveraddr tsm1/tcpserveraddr tsm3/g' /opt/tivoli/tsm/client/ba/bin/dsm.sys
sed -i -e 's/nodename yourvm-bu/nodename vm01-bu/g' /opt/tivoli/tsm/client/ba/bin/dsm.sys

Included files:
dsm.sys - Base config file for TSM on ESX
dsm.opt - Base config file. Not typically used on ESX but included just to be safe.
inclexcl - Include/exclude file. This dynamically changes with my TSM backup script.

0 Comments Permalink
0

When running TSM for the first time on ESX, you are prompted for a username and password. Here is a script to auto answer the first run questions. It assumes that the dsm.sys file exists and is correct. It also assumes that the node has been created on the TSM server.

I would recommend this script if you wish to auto-configure TSM during kickstart.

Requirements:
TSM Configured
perl module IO::Tty
perl module Expect.pm

0 Comments Permalink
2

Here is a TSM backup Script that I automate on each of my servers. I actually run this off of a mount.

tsmex and tsmvm.pl should be in the same directory.

tsm is a cron job file that goes in /etc/cron.d. After copying execute the following statements:
chmod 0644 /etc/cron.d/*
service crond restart

2 Comments Permalink
0

Here is a simple way to change you NTP settings. I have found this perl script handy to run during kickstart. Open the script and replace ntp1.yourserver.com and ntp2.yourserver.com with your NTP servers.

0 Comments Permalink
0

Since applying the performance patch this morning (see previous post in this blog), we've had consistently good performance all day, and no one has reported performance issues. To tune performance further, we are akamaizing (i.e. moving to Akamai edge servers) some static files like .gifs. And while we continue to monitor the systems closely, it seems it's time to turn our attention to usability.

And with that in mind, John updated the VMware Communities Known Issues today, which we are using to prioritize and drive our development tasks. Be sure to read through it and let us know if you perceive things differently. We plan to provide more updates next week on when you can expect to start seeing these on the site -- we are working to get the top items as quickly as possible.

0 Comments Permalink
0

This morning, we implemented a patch to the Communities platform. The patch resolved an issue where a dispatcher of static objects (.gif files, etc.) was creating a bottleneck, which resulted in a throttling of throughput. We also implemented front-end monitoring of the system, which simulates community members' experience and will notify us if page load times become too long. This will allow us to catch more quickly issues like that experienced early this morning. The new monitoring is in addition to the application monitoring that we already had in place since we went live.

So, the good news, and yes, we all need good news right now: Performance has been fast and steady the last hour. Please post in the Performance Feedback thread if you are experiencing performance problems.

I will report next some more detail on our plan for usability improvements.

0 Comments Permalink
3

consolidation and scrap in RedIT-Blog

Posted by pfuhli Sep 26, 2007


After finishing the consolidation project in which we were migration most of our matured servers to VI3 we had a problem :0

Look at the attached pictures to get some impression what we had to drag out of our datacenter ...

http://communities.vmware.com/servlet/JiveServlet/downloadImage/38-1038-1049/DSCN3136.JPG

http://communities.vmware.com/servlet/JiveServlet/downloadImage/38-1038-1050/DSCN3137.JPG

http://communities.vmware.com/servlet/JiveServlet/downloadImage/38-1038-1051/DSCN3138.JPG

3 Comments Permalink
0

RedIT in RedIT-Blog

Posted by pfuhli Sep 26, 2007

Hi there,

my name ist Daniel and I am IT-Infrastructure Project Manager at the University Hospital in Leipzig (Germany).


We recently started a project to build up two new datacenters as the heart of a redundant infrastructure of our IT.


After two years testing and consolidation with ESX and VI3 we focus now standardizing our infrastructure using VMware. Today we use WMware VI3 for testing purposes as well as for production. We are running some of the most critical application servers on VI3 like e.g. our surgery operation documentation system MCC some of our Citrix presentation server nodes, intranet server, DHCP server, printing server and so on ...


If you are interested about our ongoing projekt RedIT come back for more information.


Please also do not hesitate to ask questions using the comments functionality of this blog.


Regards,


daniel

0 Comments Permalink
0

Performance: We implemented performance fixes today, which have decreased load times of pages and removed the intermittent delays many users were experiencing.

  • Configuration changes on the proxy
  • Increased cache size on the application server
  • Removal of Javascript from the homepage dropdowns and the right column

We are working to further reduce the amount of Javascript in the next few days, especially on the homepage. And we are continuing to monitor performance and work with Jive Software to resolve issues.

We are now turning our focus to the many usability fixes and improvements you all have been reporting.

Lost in Transition: To start off, we are working on the following Forums features missing in Clearspace:
  1. Threaded view
  2. See all posts from user -- now you can only see the new threads started by a user, not all posts and replies
  3. View communities with unread threads at a glance, so users can scan the list of communities and see which ones have new posts since they last visited
  4. Visible indicator of watches/favorites so users can track threads they were researching or working on answering.
  5. Set # of posts per page as a per-user configurable preference
  6. Search returns per-Message results, i.e. if a search term shows up in several messages of a thread, it should return one result instead of one result per message
  7. Turn off the editor and use a plain text editor
  8. User filter in search, i.e. ability to search by user
  9. Code markup missing, e.g. for [code], pre , and [url]{noformat}
  10. Search in this community (currently only search in all communities)
  11. Advanced search so users don't have to do a blank search to get to the advanced search functionality
  12. No PM notification if private email

These are roughly in priority order, although we may increase the priority of the editor fix, since that prevents some users from posting. We will update here and in the Missing Features thread.

Other Usability Fixes: We are also looking at other usability fixes, and will have a fully-prioritized list tomorrow. Among others, we are looking at the following:
  • Making Discussions the default tab (not All Content), which will make read tracking of threads more visible
  • Fixing OR search
  • Turn off default email notification

Low-Hanging Fruit: While we investigated the more important issues, we slipped in some easy fixes:
  • Fixing of Reply button so it doesn't fall off the page -- we enclosed those pages in a <table>
  • We will add tables to a few other pages with similar problems
  • Put "Up to Discussions" link at the bottom of threads as well -- they were only at the top
  • Put points in Top Members boxes
  • Ensure all listings have Previous and Next buttons -- listings on All Contents tab did not

I know many of you continue to be frustrated, but I ask you to bear with us while we fix these issues. With a little patience, I still firmly believe the benefits of the new features will become more and more apparent. One is already apparent to us behind the scenes -- this platform is much easier to customize and bug-fix than Forums. We are working to push out fixes as fast as we can.

0 Comments Permalink
0

End of Day 5... in VMware Communities Blog

Posted by ericni VMware Sep 26, 2007


Today was a better day... the team worked hard and every day we seem to make progress. This was the first day I've seen where we were able to focus on more then just stability and speed. We upgraded multiple times today and fixed what I hope to be the last big performance related issues. We are continuing to work on performance... but believe we are solving the big issues.

We have also started working on the list of useability changes that some of our active community members have supplied us. It was actually a pleasure for most of us to get some cycles to work on real usability enhancements, not just performance debugging.

It was also a pleasure to see over 240 users logged in during the day, this is also the average number of users we typically had seen before we started the upgrade last thursday. It was good to see that we were handling the load without seeing serious system regression. Most people reported better performance, even though we were reaching record people online.

Tomorrow we would like to hear from people who are still seeing lag times of more then 15 seconds. Possibly John T. will setup a document that people can edit with a line entry. (Time of Day, Symptom, City/Country). It is hard for us to know for sure if we have bugs worked out of the system. Time of day is important because we need to align this with any restarts. We are hoping to have fewer and fewer of these.

Thanks again for all the people who have worked on the site, given us feedback and allowed us to continue improving the environment.

EricNi25

Web Communities Team


0 Comments Permalink
1 ... 107 108 109 110 111 Previous Next

Actions

Notifications

Looking for a blog?

  • Can't find a specific blog? Try using the Blog page to browse and search blogs.

Communities