What follows are some notes taken on vmware server. Most are related to an install on ubuntu-server (NO GUI INSTALLED)…. the main point of this is to have the host system take as FEW resources away from the guests as possible.
This requires a few x libraries – but not full blown X gui.
Adobe has released updates to Acrobat Reader for Windows and Mac systems. I don’t see any mention of linux in the advisories, but I do see that the linux version available for download is now 7.0.8 as well (which is the same as the Mac/Windows versions.) *(Apparently they’re pushing more Yahoo! integration in this release as well.)
OK – just fresh off solving the reason I couldn’t get vmplayer to start on my 1Ghz AMD 64 1GB of memory machine…. I started looking at vmserver. (RC1). Now, vmplayer is a free download (so is vmware server). I’m really liking what I see from vmware server and am pretty sure I’m going to stick with it over the player. There are quite a few interesting features and I’ll try to give a skim over them. For starters, there is a management console that gives you the ability to connect to a vmware server on a different host (with credentials as a user on that system), or the localhost. On the localmachine it shows what virtual machines you have configured and gives you the chance to create another one, open an existing one not in the list, or tweak host settings.
I just came across this article about wine-doors which sounds VERY promising. Of course, let me set the stage. Wine is a windows compatibility api for linux. The goal of wine is to allow windows applications to run on top of a linux system without modification (of the original windows version.) There are codeweavers wine (a commercial version that drives most of the development – and is known for easier installs of software.), the main open source wine and cedaga (aimed towards games 3d directx implementation.)
One of the big linux news stories yesterday was the release of google earth for linux. Essentially the Google earth team has released “release 4” which is a beta version of the next release. It looks like there are greater “user contribution” capabilities with this release. I’ve tried the download for linux and can say that it installs well, the user interface looks fine (it’s not a wine-wrapper application – it’s a true linux port.) It’s not usable yet (for me…)
As you might recall some time back I talked about the release of vmplayer which is a free virtual machine “player” from vmware. Mostly, I wrote about the problems I had running it on my main desktop which was running Mandrake 10.1 at the time. The error was basically a signal 11 in the log file, which MOST everyone says is a hardware problem (or kernel…) Well, I tested the memory, everything seemed fine. Finally, I was having other issues and an itch for an upgrade, so I upgraded to an AMD64 3000 based system with 1GB of memory and did the upgrade to Mandriva 2006 as well (first with the original kernel, now with the updated kernel.) Vmplayer still refused to run.
I mentioned the other day revisiting the qemu site to discover new versions of both qemu and the non-redistributable kernel module kqemu (qemu is now at version 0.8.1 and kqemu is at 1.3.0pre7). There’s also a new startup switch -kernel-kqemu which enables FULL virtualization of both kernel and user code. I could feel a difference in speed, but I wanted to give a test with a stopwatch to see what the numbers were and this is what I found.
Ethereal has quite a reputation for itself, I’ve used it in analyzing traffic on the home network and experimenting with virtual images… anyway, it’s a packet sniffer and network protocol analyzer and it now has a new name and new home…. wireshark.org. Apparently the lead developer did not own the rights to the name ethereal and is transfering to a new job, the old company keeps the ethereal name. He said in the explanation that he is NOT going through a namechange like that again for the project and is in the process of trademarking and will ask for input shortly as to how the development team wants to hold the trademark.
It sounds like this vulnerability would take a great deal of user interaction, but cio-today is reporting on a browser vulnerability that affects pretty much every javascript enabled browser. According to Symantec …. “This issue is triggered by utilizing JavaScript ‘OnKeyDown’ events to capture and duplicate keystrokes from users,” and is a way that the attacker could scrape/log things that are typed in (bank information, passwords, etc.) Also, they say “In one scenario, a crafty programmer might be able to trick users into entering personal data into a seemingly secure field on an online payment form, giving the hacker access to anything typed within the field.”
It’s quite a dilemma when a software product is billed as more secure than another…. several days back when Mozilla Firefox released v. 1.5.0.4 which fixed a number of security issues, I saw someone comment “I thought firefox was supposed to be secure.” I think there’s a misunderstanding when it comes to software. I think the misunderstanding is that one piece of software can be secure and another not. Out of the box. Let’s take a stab at clarifying…. Security is not a product, it’s not a feature, it’s a way of doing things. Along those lines….