Just a quick update:
After being away from my systems for a week because I'm in the process of moving, I set up my workstations in the new house, along with two new toys. I had the opportunity to throw Windows 7 up on a VM, but haven't had much time to play with it. So far, I'm more impressed with Win7 than I was with Vista, in terms of usability, notification handling, and its more enterprise-friendly configuration options.
For the past week I've been working on getting my system back up and running- since upgrading to kernel 2.6.29 and a newer intel driver, X no longer starts correctly- basically everything I've experienced is documented in this gentoo bug report. I now have an intimate understanding on configuring Xorg-server, using masked packages, and installing X11-drivers on gentoo. I'm still struggling to get X to work properly, but so far am still having issues with it starting then freezing. As of right now, I have recompiled x11-drivers as well as followed the steps in this Gentoo how-to.
At work, I've finally been granted some Active Directory privileges, so I'm trying to take a crash course in AD group policy. I have learned basic AD administration, such as user and computer creation, managing organizational units, and user privileges. Right now, I'm working on creating group policy to implement auto logoff and something along the lines of Windows SteadyState. I'm also in the process of configuring Mac OSX to authenticate users using the University's LDAP tree. I've also had the opportunity to toy a bit with SmartDeploy, but have yet to actually start deploying workstations with it.
That's it for now, hopefully I'll be able to post a bit more frequently now that my move is almost complete.
Wednesday, July 8, 2009
Saturday, June 27, 2009
GCC Update Problem
So after running emerge --update --deep --newuse world, and then emerge --depclean, somehow I lost Firefox, my gentoo sources (used to compile/recompile the kernel), and I could no longer compile any emerged packages... Somehow the update shot GCC. I googled the problem, and found this forum post about my very problem, so I unset GCC_SPECS and reemerged gcc (which takes forever to compile), and was able to reinstall firefox and install gtk+, to play with some Gnome eye-candy later...
Alas, I still can't recompile my kernel to enable extra CIFS support required to use Windows WINS names. Everytime I try running make menuconfig, I get
and when I try to reemerge gentoo sources, I get
I went back and modularized CIFS support, with the following options:
Hopefully the added CIFS support (namesly CIFS extneded attributes and CIFS statistics) will get me working all the way.
Alas, I still can't recompile my kernel to enable extra CIFS support required to use Windows WINS names. Everytime I try running make menuconfig, I get
make: *** No rule to make target `menuconfig'. Stop.
and when I try to reemerge gentoo sources, I get
* Messages for package sys-kernel/gentoo-sources-2.6.29-r5:
* If you are upgrading from a previous kernel, you may be interested
* in the following document:
* - General upgrade guide: http://www.gentoo.org/doc/en/kernel-upgrade.xml
>>> Auto-cleaning packages...
>>> No outdated packages were found on your system.
* GNU info directory index is up-to-date.
I went back and modularized CIFS support, with the following options:
│ │ [*] CIFS statistics │ │
│ │ [*] Extended statistics │ │
│ │ [ ] Support legacy servers which use weaker LANMAN security (NEW) │ │
│ │ [ ] Kerberos/SPNEGO advanced session setup (NEW) │ │
│ │ [*] CIFS extended attributes │ │
│ │ [*] CIFS POSIX Extensions │ │
│ │ [*] Enable additional CIFS debugging routines │ │
│ │ [ ] CIFS Experimental Features (EXPERIMENTAL) (NEW)
Hopefully the added CIFS support (namesly CIFS extneded attributes and CIFS statistics) will get me working all the way.
Sunday, June 21, 2009
FreeBSD
Got bored waiting for the Gentoo box to emerge and compile all the new packages, so...
Now I just need to familiarize myself with the different file structure, and get Gnome running faster. Install was comparatively painless, it did most of the hard part for me. First adjustment: no /etc/init.d, this is true blue Unix, so I'm looking at /etc/rc.d and /usr/local/etc/rc.d
Now I just need to familiarize myself with the different file structure, and get Gnome running faster. Install was comparatively painless, it did most of the hard part for me. First adjustment: no /etc/init.d, this is true blue Unix, so I'm looking at /etc/rc.d and /usr/local/etc/rc.d
Repositories
So I've got Samba server running flawlessly on my Gentoo box, but I'm still struggling to get the server to see my Vista machine's shares. I've approached this problem from two angles, the first trying to get nautilus to recognize smb:// shares and the second using smbclient to browser shares. I ensured that cifs support was enabled in the kernel (I think I'm going to go back and modularize it...) and after a little googling, I made sure that gnome-base/gnome-vfs was emerged. Even after that, and restarting X, I could not get nautilus to browser netwrok shares. I next moved on to try smbclient (smbclient -L
The Ubuntu VM is able to access both the shares on the Gentoo machine and the Vista machine using nautilus, although accessing the Vista share is definitely cumbersome and seems to work inconsistently and I'm not entirely sure why. When using smbclient from the console on the Ubuntu box, I am able to scan the shares on the Gentoo box, however I first had to enable lanman auth and client lanman auth in the Ubuntu VM's smb.conf to connect to the server. I'm not really sure why the server is requesting lanman authentication, although maybe some sort of authentication/encryption is required by samba to connect to shares that don't use user-level security. Theoretically, no authentication should be necessary using smbclient since guests are allowed. I'm still digging to find the answer...
In the process, I definitely had one big *duh* moment. After changing my USE flags in /etc/make.conf to add samba, I wanted to ensure all packages now would support samba (particularly gnome). I ran emerge --update --deep --newuse world, only to get stopped immediately with a "masked package error," as well as an error stating that I needed to update portage. So I updated portage and ran emerge --update --deep --newuse world again, only to hit a brick wall trying to update /sys-lib/timezone-data. Portage could not find timezone-data-2008c. This seemed a bit odd to me, as obviously, its now 2009, so why is portage trying to find a package marked 2008? So after a minute of twiddling my thumbs, I recalled having to always run 'apt-get update' on my Ubuntu laptop to update the repository tree... and I hadn't been doing the same on my gentoo machine with portage. AHA! Well a simple emerge --sync should do the trick, and I should be home free!
Well about this time, Charter decided to throw the Internet switch from "working" to "broken," for my router and I couldn't pull down the new portage tree, at least not within a reasonable amount of time. After a failed rsync, I tried again (I was still getting about 40% of my packets through, and I really wasn't in the mood to call Charter), and got responses from the rsync server... all of which were 404. So instead of giving in, and just calling Charter, I tried switching my SYNC parameter within etc/make.conf to a different server, first to Georgia Tech's (biting my tongue... I'm a UGA grad) with no success. After having a near aneurysm due to frustration, I figured out that I needed to emerge mirrorselect and run mirrorselect -r -i to find a new server. I managed to find a new rsync server, but finally gave in and decided it was about time to call Charter, as I was now experience around 80% packet loss *grrr.
So I dialed up the dreaded 1-888 number, I repeatedly hit '0' and was patched right through to a representative who was very helpful, and in fact got a technician out to me in a matter of two hours (unprecedented for Charter!), who came out and checked my levels, reran a cable, and swapped out my dying Motorola Surfboard modem, and got me back on my feet- he didn't even bitch about me using a router!
So now I was able to successfully sync portage, and again, ran emerge --update --deep --newuse world only to get another error that I needed to add a USE flag to do that...
So I just did it the quick, temporary way USE="###" emerge --update --deep --newuse world and it worked... until I got smacked with another error <sys-apps/man-pages-3 ("<sys-apps/man-pages-3" is blocking sys-apps/man-pages-posix-2003a). this forum post which seemed to the trick...
So that's where I am now, running a deep update with new use flags, hopefully not fruitlessly, currently on package 49 of 391. I guess at the very least, my server will be all up to date! More to come later....
The Ubuntu VM is able to access both the shares on the Gentoo machine and the Vista machine using nautilus, although accessing the Vista share is definitely cumbersome and seems to work inconsistently and I'm not entirely sure why. When using smbclient from the console on the Ubuntu box, I am able to scan the shares on the Gentoo box, however I first had to enable lanman auth and client lanman auth in the Ubuntu VM's smb.conf to connect to the server. I'm not really sure why the server is requesting lanman authentication, although maybe some sort of authentication/encryption is required by samba to connect to shares that don't use user-level security. Theoretically, no authentication should be necessary using smbclient since guests are allowed. I'm still digging to find the answer...
In the process, I definitely had one big *duh* moment. After changing my USE flags in /etc/make.conf to add samba, I wanted to ensure all packages now would support samba (particularly gnome). I ran emerge --update --deep --newuse world, only to get stopped immediately with a "masked package error," as well as an error stating that I needed to update portage. So I updated portage and ran emerge --update --deep --newuse world again, only to hit a brick wall trying to update /sys-lib/timezone-data. Portage could not find timezone-data-2008c. This seemed a bit odd to me, as obviously, its now 2009, so why is portage trying to find a package marked 2008? So after a minute of twiddling my thumbs, I recalled having to always run 'apt-get update' on my Ubuntu laptop to update the repository tree... and I hadn't been doing the same on my gentoo machine with portage. AHA! Well a simple emerge --sync should do the trick, and I should be home free!
Well about this time, Charter decided to throw the Internet switch from "working" to "broken," for my router and I couldn't pull down the new portage tree, at least not within a reasonable amount of time. After a failed rsync, I tried again (I was still getting about 40% of my packets through, and I really wasn't in the mood to call Charter), and got responses from the rsync server... all of which were 404. So instead of giving in, and just calling Charter, I tried switching my SYNC parameter within etc/make.conf to a different server, first to Georgia Tech's (biting my tongue... I'm a UGA grad) with no success. After having a near aneurysm due to frustration, I figured out that I needed to emerge mirrorselect and run mirrorselect -r -i to find a new server. I managed to find a new rsync server, but finally gave in and decided it was about time to call Charter, as I was now experience around 80% packet loss *grrr.
So I dialed up the dreaded 1-888 number, I repeatedly hit '0' and was patched right through to a representative who was very helpful, and in fact got a technician out to me in a matter of two hours (unprecedented for Charter!), who came out and checked my levels, reran a cable, and swapped out my dying Motorola Surfboard modem, and got me back on my feet- he didn't even bitch about me using a router!
So now I was able to successfully sync portage, and again, ran emerge --update --deep --newuse world only to get another error that I needed to add a USE flag to do that...
So I just did it the quick, temporary way USE="###" emerge --update --deep --newuse world and it worked... until I got smacked with another error <sys-apps/man-pages-3 ("<sys-apps/man-pages-3" is blocking sys-apps/man-pages-posix-2003a). this forum post which seemed to the trick...
So that's where I am now, running a deep update with new use flags, hopefully not fruitlessly, currently on package 49 of 391. I guess at the very least, my server will be all up to date! More to come later....
Friday, June 19, 2009
Slowloris DoS
This looks like it could be some fun... and some trouble. As of yet, Apache has no patch. Too bad it doesn't work for IIS, however given my personal experience, IIS doesn't need a DoS attack to be brought down, just let it run for an hour, and it'll eventually crash
45 Seconds
The time it takes for my Gentoo box to cold boot.
I still need to time my Vista machine, but I imagine it won't be anywhere close!
I still need to time my Vista machine, but I imagine it won't be anywhere close!
Wednesday, June 17, 2009
POP vs IMAP
E-mail infrastructure has always been one of those things that I've never really taken the time to think about. You set up your client to use POP, or IMAP, or Exchange, and don't think twice about it. I've always been more familiar with how POP (post office protocol) works, as it is one of the oldest methods of e-mail retrieval, but IMAP has always been a bit more obscure to me, simply because I never really needed to use it because I always just used POP.
Well after reviewing this article, I've learned that IMAP may in fact be the superior protocol, at least for my purposes. IMAP, unlike traditional POP, supports online processing, using a more interactive client-server model. The obvious advantage to IMAP is that it allows users to access their email from different machines at different times (similar to just using a web mail client), wheras the POP paradigm is best suited for individuals that use only one client computer all the time.
So next time, given the choice, I will opt for IMAP over POP.
Well after reviewing this article, I've learned that IMAP may in fact be the superior protocol, at least for my purposes. IMAP, unlike traditional POP, supports online processing, using a more interactive client-server model. The obvious advantage to IMAP is that it allows users to access their email from different machines at different times (similar to just using a web mail client), wheras the POP paradigm is best suited for individuals that use only one client computer all the time.
So next time, given the choice, I will opt for IMAP over POP.
Subscribe to:
Posts (Atom)