Skip navigation
Steve Carl

RedHat Summit 2010

Posted by Steve Carl Jun 28, 2010

I am traveling back from the RH Summit 2010, and have a couple hour layover in Detroit, so I thought I would try to get some thoughts about it down here while they were still fresh. Never mind the pages upon pages of notes I took on the iPad, where next week I will be editing them for an internal presentation and going "What in the world did I mean when I wrote "rhev next advantage T.P."" (Oh yeah: RedHat Enterprise Virtualization version 2.2 (next version) will take advantage of transparent large pages in EL 6)


The Experience


If you have never been to a RedHat Summit it is probably worth talking about what it is like. In some ways, it is like many other of the very best tech conferences I have been too, reminding me favorably of SHARE and BMC Userworld, mostly because it is non-stop, never ending, drinking from a fire hose of information delivery. Like a good conference, one person can not cover it. In fact, two of us went and could not cover it. And we did not even try to get to anything about JBoss. Just the Redhat and and Cloud tracks were massed with things I wanted to attend, and I was very often choosing between two different talks that I very much wanted to be at.


The energy and the drive of the whole thing was palpable, and the geekery was also off the scale. Over and over I heard, when a GUI was being described that we should not worry because the command line was still there, and everything was still script-able. Competence and Confidence. There were also quite a number of decision makers there: It was an interesting melange of the corporate and the creative. IBM even sponsored dinner one night, which means IBM bought us all a beer! The HP booth was right next to the RedHat booth, and if there were any ill feeling about the fact that RedHat is pulling Itanium support with RHEL 6, it was not in evidence there. Of course, with Nehalem-EX and AMD's 6000 series, there is not much reason to miss Itanium either.




One of the cognitive dissonances for me came from all the references to the new memory and processor addressing limits of RHEV 6 over RHEV 5.5. I live in what they call the "upstream" kernel space so much on the desktop that I forget the RHEV 5.5 starts with kernel 2.6.18, and then starts backporting things from later kernels. In fact, the whole kernel number system really does not have much meaning in RHEV. One of the basic limits that appears to have been in place though were those surround how many processors, how much memory, and how well NUMA was handles. I imagine that would have just been way to hard to pull out of the upstream kernel because the changes are everywhere. With 2.6.32 as the new starting place, all sorts of new capabilities are enabled.


The last paragraph is *not* a criticism of RedHat, though I can see how it might read as such. That complication is the cost of providing Enterprise level, tested in every possible way code, so that it can be supported for the very most critical applications, be they medical or something that underlies the very heart of a stock market computing process. This is not just even about support per-se, but how long it is supported and supportable. RHEV 6 will have a seven year life span! Sure, it support 4096 cores now, and that seems like a lot, but seven years from now? One session I went to was also very careful to point out theoretical limits versus tested ones. In truth, we don't really know how many CPU's Linux can support in practice because the kernel does not really put a limit on it. Just money does, and I can not buy a 4096 core system from anyone yet... and probably not for a few years.


The "tickless" kernel is another example of a huge change for RHEV 6 that actually we have had upstream for a while. Sure, it changes all sort of things, not the least of which is how one might go about figuring out time related problems. The old days of the 1000-times-a-second loop are yesterday, and the quantum that was based off it is too. It's a brand new world of scheduled interrupts for everything. Sure, for mainframe or VMS folks that is old news, but it is a huge change with subtle effects all through the kernel. I went to two back to back session about problem determination in this new world, and it was barely able to scratch the surface of all the new things this, and other changes to RHEV 6 bring to the table.


One stat alone should tell the tale, at least for the permutation minded of you: RHEV5 was about 1500 or so packages. RHEV 6 is 3000+.


That is not a simple doubling of packages. That is orders of magnitude more permutations that need to be looked at and tested.


Enterprise Virtualization


My main focus while at the summit was around KVM, and to a lessor extent, VDI. I had been looking at this for a while, but I viewed the summit as a chance to take a deeper dive. While there I got a chance to talk to some of the people that helped write KVM, and the management tools around it, and I am impressed. I have been a virtualization guy since 1980, and there are things here that both reminded me very much of my roots (such as AMD and Intel's microcoded page table assists) and leveraging the modern such as the data-de-dup like Kernel Samepage Management (KSM). It was also clear that we are still think in more mainframe like terms here with our virtualization servers, since we just switched to Dell R810's with 256GB of RAM, and folks I talked to at the conference were saying their sweet spot was 32 GB of RAM. And a PS about KSM: One person in RH mentioned that calling KSM "Kernel Samepage Management" was a bit of a struggle, and that others used different words like "Kernel Storage Management".


Either way, KSM is one of the big big wins of RHEL 6 / RHEV. It just makes too much sense. If the page is the same, just keep one copy. It does not matter what type of Guest VM it is. If the page matches, just keep one. This is huge for things like Windows, that zero memory at boot, so they have lots of pages with recent writes that are all zeros, but Windows isn't the only guest that can have a page like that, and it should not matter..  and to KVM and RHEV, it does *not* matter. It is simple, and it is elegant. It should also help safely overcommit memory without a lot of jumping through hoops with special memory managers and guest device drivers or modified guests, etc.


I will be starting on a test deployent of KVM in the near future, and that will be my next "Adventure" in Linux. Read about it here...

Steve Carl

Mint 9

Posted by Steve Carl Jun 3, 2010

Fedora 13's stay on my Dell D620 was short-lived. The T43 still has it, but the D620 needs to do actual work at the office and on the road. It needs to be a production ready desktop. If I was not interacting with MS Exchange, Fedora would be better, but since neither KDE's Kmail nor Evolution are really working all that well against MS Exchange 2007, I decided to move on with the D620.


My first stop one evening was OpenSUSE 11.3 Milestone 6. Interesting things happening there, but not stable (nor advertised as such) so I decided it was time to go back to a favorite and see what was new. Mint 9 recently came out, and a friend of mine had it running well on an Acer Revo. Why should they have all the fun?


In theory, this should not be all that different from Ubuntu 10.04, but Mint always seems to take that extra month and really polish things up.


I did not see anything really new during the install. The timezone map is nice now... everyone should copy this way of setting the default time zone. It was the usual 7 panels that Ubuntu has as far as I could tell, but with the colors being set up for Mint's themes.


The disk layout as usual required manual override, and here is a place where OpenSUSE is much smarter. It looked at my Fedora 13 install, and correctly guessed the way the disks should be done, I.E., P1 as NTFS / "/windows", P2 as "/" and formatted, P3 as swap, and P4 as "home" and not formatted. Neither Ubuntu nor Mint can figure that out, even though it is pretty simple and to me fairly obvious as dual boot configs go:


Disk /dev/sda: 80.0 GB, 80026361856 bytes

255 heads, 63 sectors/track, 9729 cylinders

Units = cylinders of 16065 * 512 = 8225280 bytes

Sector size (logical/physical): 512 bytes / 512 bytes

I/O size (minimum/optimal): 512 bytes / 512 bytes

Disk identifier: 0x00019fb7

   Device Boot      Start         End      Blocks         Id    System

/dev/sda1               1           4256     34179687+   7    HPFS/NTFS

/dev/sda2   *       4256        5468     9741313       5    Extended

/dev/sda3            5469        5711     1951897+     82  Linux swap / Solaris

/dev/sda4            5712        9729     32274585     83  Linux

/dev/sda5            4256        5468     9741312       83  Linux


/dev/sda2 is the extended partition that all of the Linux logical disks live inside.


Disk Partitioning Aside


I was having a recent conversation offline with someone that reads my blog, and they were asking me about the how and why of the way I partition my disks for Linux. As I noted, set up the way I am, OpenSUSE 11.3 correctly divined the exact way that I have to manually set up the disks when I install Mint, Fedora, or Ubuntu on the same system.


Some of what I do is just habit from days gone by, but there are reasons for some of it.


Keep "/home" separate: This allows me to serially install all sorts of OS's and not have any of my documents, photos, music, whatever be deleted. It also means all my settings as stored in various config files are kept. That last part is a mixed blessing.


Having all my documents stay without going to the trouble of having to always move them offline to a USB disk or something is nice, but sometimes when I change releases, Gnome or KDE stop working correctly because the config files that live on in the home directory are there, and trying to make things be the way they were, and that does not always work. From time to time, I have to do some research, and figure out what config file is misbehaving, and delete it, and usually that just means shotgun blasting something like .gconf rather than messing around with it.


Having my config survive past an OS install also means that if the distro did something to the look and feel, I miss it half the time because my desktop goes back to the way it was, not the default. I'll read in a trade about the new theme's nicely smoothed corners or something and I won't see it. Again, easy to erase a config file, or even easier, set up a temporary userid to look at what the default looks like.


The good news is that, I keep my top task bar, and most of the things I have installed on it, like the applications pull-down menus. Mint has its own "start" button application menu, and it follows the trend started back a few years ago by OpenSUSE to having a sort of Windows XP'ish set of menus, try's to hide not commonly used things, integrates a search bar, etc. I dislike them all, though Mints is better than average. I like the classic Gnome menu bar. Having /home separate means I do not often have to re-set that up when I install something new.


Dual Boot / Windows at the front: I do not know if it is still true, but there was a day when your Windows partition had to be the first one. For a system that comes with WIndows already installed, and Linux is going on second, all the partition shrinking programs seem to still run everything to the front for MS WIndows. It may be that this is not the case, but it is easy, it works, and so far I have not really needed to put Linux up front. The old days of /boot needing to be within a certain distance of the front of the disk for addressing reasons are gone, so Linux is happy pretty much anywhere.


Four Partitions: for whatever reason I have never tried to set up triple boot, with two different Linuxii and an MS WIn. That means I always have four partitions: the NTFS one for MS WIn, then one each for "/", swap, and "/home". That means I could get by without an extended partition. For some reason, something somewhere along the way set up this computers with two partitions, and dropped the three logical partitions into the second real one. I guess it was trying to conserve real partitions for something or the other. Had I laid it out, it would just be four real partitions. It does not matter that much, and it would be a pain to fix, so I leave it like this.


Swap: Here is a case where I am really living in the past to some degree. I have 2GB of RAM on the laptop: That is plenty for Linux and most of the applications that I run at the office. Unless I am running a virtualization solution, 2GB at the time of this writing leaves me with 44% of the RAM being allocated as cache, and the swap not in use at all. I have OpenOffice, Evolution(MAPI), Tomboy, and Chrome with 8 tabs all running.


swapon -s

Filename    Type      Size            Used      Priority

/dev/sda3    partition 1951888      3716     - 1


Clearly nothing happening in swap-land. I could just not define it, but it always feels to me like cheap insurance. 2GB in disk space is not much anymore, and 2GB of RAM, even slow virtual RAM is still something. If I was disk constrained, and had lots of RAM, and was not going to running any Virtual OS's as a guest, I might decide not to define a swap space... but that corner case has not happened to me yet.


Evolution 2.28.3 is ... working?


Color me confused. Part of the reason for the side trip into disk layouts is to mention that this systems has sequentially had Ubuntu 10.04, Fedora 13, OpenSUSE 10.3, and now Mint 9 on it. I have installed and tested Evolution on each version, and that meant for Fedora 13 and OpenSUSE 10.3, I was at 2.30.1, and for Ubuntu and Mint I am at 2.28.3:


dpkg -l | grep -i evolution

ii  evolution                                      2.28.3-0ubuntu9                                       groupware suite with mail client and organiz

ii  evolution-common                      2.28.3-0ubuntu9                                       architecture independent files for Evolution

ii  evolution-data-server                                           evolution database backend server

ii  evolution-data-server-common                                   architecture independent files for Evolution

ii  evolution-mapi                            0.28.3-0ubuntu1                                       Evolution extension for MS Exchange 2007 ser

ii  evolution-plugins                         2.28.3-0ubuntu9                                       standard plugins for Evolution

ii  evolution-webcal                         2.28.0-1                                                     webcal: URL handler for GNOME and Evolution

ii  libcamel1.2-14                                                       The Evolution MIME message handling library

ii  libebackend1.2-0                                                  Utility library for evolution data servers

ii  libebook1.2-9                                                        Client library for evolution address books

ii  libecal1.2-7                                                            Client library for evolution calendars

ii  libedata-book1.2-2                                               Backend library for evolution address books

ii  libedata-cal1.2-6                                                   Backend library for evolution calendars

ii  libedataserver1.2-11                                            Utility library for evolution data servers

ii  libedataserverui1.2-8                                           GUI utility library for evolution data serve

ii  nautilus-sendto                           2.28.4-0ubuntu1                                        integrates Evolution and Pidgin into the Nau

ii  python-evolution                         2.30.0-0ubuntu1                                       Python bindings for the evolution libraries


As can be seen here, these are not even Mint special packages. These are the ones Ubuntu provides. But Evolution, with MAPI, is up, and running, and moderately stable. No Calendar, but email and GAL are working well, if slow. I don't get it. This was not working at all well Under Fedora 13 and even worse under OpenSUSE 11.3. With Ubuntu 10.04 it worked some, but was prone to crashes and locking out my userid in the AD domain. I am speaking here only of MAPI: I was able to get IMAP to work in all cases, although IMAP+ was not working well, so I used the original IMAP.


But Evolution with MAPI is working under Mint 9. I just don't get it. More Evolution fragility I guess, but working in my favor this time.


OpenOffice 3.2


Mint 9 comes with Openoffice 3.2:


It is interesting to see the it is:


1) Already Oracle branded

2) From Debian, through Ubuntu, to Mint.


I don't see any special Mint massaging here. There is one thing I wish that Distros would do, and that is provide access to the HTML/Web editor mode from the menu, rather than making me launch the writer, switch it HTML, and then start to work.


OpenOffice works well, launches quickly, and is extremely usable these days in mixed Linux / non-Linux environments. I get files from non-Linux users all the time, and so far there has not been anything that OpenOffice has not handled. There have been files that were corrupted that *only* OpenOffice could deal with though.




Something I have always been unable to do with Linux is talk to people who use MS's so-called "Communicator". I say so-called because as a closed platform, it was clear one could only communicate with others on the system. I wondered how the SIPE project was doing these days, and looked in Synaptic, saw pidgin-sipe was available:



dpkg -l | grep pidgin

ii  pidgin                                1:2.6.6-1ubuntu4                              graphical multi-protocol instant messaging c

ii  pidgin-data                       1:2.6.6-1ubuntu4                              multi-protocol instant messaging client - da

ii  pidgin-facebookchat       1.64-1                                                Facebook Chat plugin for Pidgin

ii  pidgin-libnotify                  0.14-1ubuntu14                                display notification bubbles in pidgin

ii  pidgin-sipe                       1.8.0-1ubuntu1                                  Pidgin plugin for connect to LCS/OCS


Installed it. Configured it. Now I can IM non-Linux users inside the office! Wow... I wonder when that happened?




Mint 9 comes with Firefox 10.6.3.  I like Firefox still, but these days for speed reasons I mostly use Google Chrome. Currently I am at 6.0.422.0 dev. I run Chrome on Linux, the Mac, and WinXP/7. On resource limited, slow computers, it is just the thing to make it feel faster. On a fast computer the differences are less obvious, but still there. the other thing I like about Chrome is that it has a trick Opera has, where I can resize some text input fields to be larger. Some web forms default input boxes are stupid-small, and this lets me get around that.


And yes, I always run the developer channel version. In early days that was where you had to be to get feature / function comparable to Firefox. these days I mostly see the speed improvements first. While there is no guarantee about it, I have only had a crashing browser once in several years of using the developer channel, which the next update fixed.


One nice trick Chrome does is that once I install it's .deb, it updates the repository list so I keep getting fresh updates as part of a standard updating process. I don't have to do anything special to stay current.


It's like coming home...


After all the bleeding edge stuff: the early releases of Ubuntu 10.04, the betas and GA of Fedora 13, and the Milestone 6 of OpenSUSE 11.3, Mint 9 just feels like a warm cozy fireplace on a cold winter day. Everything is stable. Everything just works. No messing around.


Too bad I'll probably load Ubuntu 10.10 Alpha up soon... Can't stay comfortable too long.

Filter Blog

By date:
By tag: