| virt-manager and guest serial console |
[Jul. 2nd, 2009|02:55 pm] |
So maybe this is already well known, but I always have a heck of a time getting a serial console to one of my KVM guests managed by virt-manager. I don't know why this is, but I do. The UI doesn't seem to present an interface for it, and the hardware wizard doesn't present an option for adding a serial port to the guest hardware. So I sat down and figured out how to get it working today, at least with F11.
Most of my guests have a serial port configured in the guest.xml file already. Whether I added this long ago, or virt-install creates it, I can't tell. If your guests don't have this, the syntax seems to be:
<serial type='pty'>
<target port='0'/>
</serial>
I opened up the guest details GUI in virt-manager, went to the serial port entry and saw that Serial 0 was directed to /dev/pts/10. I started the guest up, edited the kernel command line boot parameters to include 'console=ttyS0', and booted.
Then I did what I would do normally for any real machine hooked up to a serial port and fired up the tried and true 'minicom' like this: sudo minicom -p /dev/pts/10
Amazingly enough, there was the serial output. A few quick configuration changes later, and I could even login to the guest via that.
This may be really really simple. Some of you may even be laughing that I bothered to write this up. I don't care. My googling skills either suck, or this isn't very widely known. If nothing else, I wrote it up so I could remind my tiny brain how I did it before. Maybe someone else will find it helpful too. |
|
|
| Random Fedora Observation of the Day |
[Jun. 26th, 2009|11:00 pm] |
The PackageKit applet. Leaving ugly red zits on my panel since Fedora 10.
Seriously. It is the most excited and overzealous software I have seen in a long time. It continues to signify updates available, regardless of whether or not there are. Even when I click on it and it then reports none available.
I would care or rant or file a bug, but it's persistence is so astounding that I can't help but find it cute. Perhaps it knows that I'm doing updates pushes as fast as I can, so it doesn't bother turning itself off.
So I applaud you PackageKit applet. Even in your utter failure to function properly, you still seem to get it right anyway. |
|
|
| Updated Fedora 11 PPC iso |
[Jun. 24th, 2009|03:04 pm] |
Due to a bug found really late in the F11 cycle (as in 3-4 days before release), the GA boot.iso and CD/DVD iso images that came with Fedora 11 are broken on ppc64 machines. The bug wound up being in yaboot, and thus was not easy enough to fix for the release since the compose and media use yaboot to boot things. Sad day.
But wait!
I bugged Jesse Keating about possibly spinning an updated boot.iso with the yaboot from F11-updates. To which he replied, "That's just a pungi compose with updates enabled. Can't you do that?" And behold, he was right. So I did that. And then I had a couple of awesome people test it. And it worked.
You can find the updated boot.iso here:
http://jwboyer.fedorapeople.org/boot.iso
There is no warranty, it is not guaranteed to work, and if it makes your system catch on fire it is not my fault. But hopefully people find it useful and we get a massive explosion of ppc64 systems installed with F11 because of it.
Moral of today's story: Ask not what Fedora can do for you, but how you can use Fedora to scratch your own itch! |
|
|
| Fedora Updates Status |
[Jun. 23rd, 2009|04:21 pm] |
I hate microblogging. I don't understand it, it seems like a big waste of time, and I'm busy enough as it is. However, I get asked often enough about 'When is the next updates push going to happen' or 'Where is the push at?' that I have decided to do something about it.
I created an identi.ca account for myself. I'll post updates push information on there, and that's pretty much all I'll use it for. So if you're interested in know what is going on with the updates and updates pushes, follow 'jwboyer' on identi.ca. THIS IS NOT FOOL-PROOF AND NOT DEFINITIVE, but I will try and update it regularly.
That is all. I now return you to your regularly scheduled fu. |
|
|
| Fedora PPC |
[Jun. 19th, 2009|09:30 am] |
I realized that I haven't blogged about the change in store for PowerPC when Fedora 13 comes about. Contrary to some beliefs, it's not because I am bitter or busy sobbing into a beer. Mostly, I've just been very busy lately.
I understand the reasoning behind the Board and FESCo decisions. I do believe that having a non-x86 architecture as a primary arch is a benefit to not only Fedora, but also to the upstream projects we consume. It helps keep code bases robust and portable and exposes maintainers to something a little different. However, given the statistics we have I can't really fault the committees for voting the change.
Instead I'll take a second to point out that if we want Fedora to continue to work well on PowerPC, we'll need people to step up and help out. While the change doesn't switch over until Fedora 13, I would almost like to get things in place and ramped up starting _now_. That way when the primary koji instance switches off, it should be fairly seamless.
So what do we really need in place? Well, we need to flesh out a SIG for starters. And behold, we already have a SIG wiki page. Though I will note it is very out of date and probably needs redoing completely. Aside from that we need:
- Koji builders. The Board has a request in to reuse the existing Fedora ppc builders. This will require some coordination, given that the builders will still need to be used for F10, F11, and F12 as primary builders and they can only talk to one koji hub (I think). So assuming the Board approves, we'll likely split a small number out for F13 and add more as other releases expire. Long term though, we'll need to come up with replacement hardware.
- A Koji hub. This would be the hub the builders talk to for the ppc secondary arch. It will also need a larger amount of storage to store all the packages being built.
- People. We so need people. People to fix bugs, people to run the hubs, people to TEST. People to TEST. Oh, and people to TEST.
Things we can work towards after we get some of the above in place:
- Releases. If we're at the point we can build the package set fairly regularly and test it out, then we can focus on some rel-eng type issues like composing releases.
- Updates. We'll need a bodhi instance running somewhere to manage this, and we'll probably have to sit down and think about how it's really going to work.
Sound like a challenge? You bet. But if it also sounds like fun and you want to help out, then let me know. Sign up for the fedora-ppc list. Subscribe to the SIG wiki page. Hopefully we can take our users (and yes, we do have users) and make them contributors. I know I'm at least going to try. |
|
|
| FAD Day 2 |
[Jun. 9th, 2009|08:51 pm] |
We picked up today where we left off. Starting off the day with a wonderful literal video translation of the horrible Bonnie Tyler 'Turn Around' music video was a bit odd, but hilarious nonetheless.
We brainstormed a bit more on the topic we had intentionally avoided yesterday, which is how we manage freezes, why they suck, etc. It's been fairly well known that freezes are painful, and the final freeze impacts both our developers and release engineering team pretty hard. This is arguably one of the reasons we had the FAD to begin with, so it was great to get the issues out in the open.
After a break, we moved from brainstorming about the problems to thinking about some of the actual solutions that we could reasonably implement. After we had been working in the issues for a while, it was starting to become apparent which were fairly easy, which were really hard, and which we should spend the effort on. I was pleased when we were able to identify a few key issues that if we really fixed them, they would make yet a wider set of issues much more simple.
The next steps for tomorrow, are to sit down with the ideas and start drafting up some proposals to present to the community for discussion. I do want to stress that we're drafting PROPOSALS, and not policies. Unfortunately, I won't be able to attend tomorrow's session.
Towards the end of the day, Jesse asked whether we thought the overall FAD was worthwhile. To me, it was a two fold answer. I think it was a huge success in that the concentrated, uninterrupted all-day sessions allowed us to accomplish in 2 days what would have taken at least 3 weeks on the various rel-eng and QA lists. Working on lists and IRC is how we do business, but at times there is no replacement for face-to-face communication.
The second, and arguably most important, part of my answer depends on you, the contributors. The return on investment really comes from the submission, discussion, and hopefully positive feedback and approval of the proposals that are being worked on. I have high hopes that we'll submit some pretty good proposals with good background and information about what, how, and why the changes are being proposed. If we get a number of these approved, then I really think we can help our contributors and make the Fedora project that much better of a project to work on.
I am pretty excited about this and hoping the outcome is positive.
Unrelated: The Red Wings better win tonight, or my face-to-face taunting of Greg will be slightly wasted. There is no substitute for face-to-face taunting when you have won a bet. GO WINGS |
|
|
| FAD Day 1 |
[Jun. 8th, 2009|10:33 pm] |
As I'm sure you've seen by now, today was Day 1 of the development cycle Fedora Activity Day. Overall, I think this was a really productive day.
We started off by doing something called 'Design Thinking'. This was new to me, and sounds pretty buzzwordy, but the general premise is to identify the problems you are facing without focusing on the negative aspects of this or devolving the discussion into details about stuff that generally doesn't matter. To be honest, it sounded a bit too Zen for me at first, but I think it worked rather well. The irony of staying positive while identifying all the fail was no lost on me though.
We had two whiteboards full of various topics that were deemed to be problems in one form or another. Many of them were sort of inter-related, but a surprising number didn't overlap. We took these, did a BarCamp style voting on them, and came up with a set of items to start discussing.
We cherry-picked from the list and identified a number that just seemed outside the scope of the FAD, which helped focus things a bit. Then we started to delve into the remaining items and flesh out some of the details of the problems. We got through a good number of them, but the harder ones are remaining for tomorrow.
I was quiet pleased with the level of participation in the discussions, both in the room and in the IRC channel and VoIP session. We also had an active gobby session going to log our notes to place into the Wiki as we work. I'm sure tomorrow will bring even better results. |
|
|
| The Elections are actual races now! |
[Jun. 1st, 2009|09:35 am] |
I was a bit surprised to see that the FESCo election nomination page had an explosion of candidates over the weekend. There are 5 seats open at the moment, and 12 candidates. Out of those 12, there are 7 people who have never served on FESCo before, 1 coming back to the fold, and 4 FESCo members that have their seats up for election. I think this will make for a great race, and we'll get some new blood into the engineering leadership as well.
The Board election is also a race between 5 candidates for 3 seats.
Looking forward to some of the town-hall meetings and such coming up. Remember, if you have questions for any candidate in the FESCo or Board election, feel free to ask them. Or you can add your question to the Election Questionnaire. |
|
|
| Victory! |
[May. 27th, 2009|01:09 pm] |
By all kinds of random math I can think of, I seem to have won the Fedora Hockey Challenge. I've yet to figure out how I'm going to collect all my prizes, but I look forward to a fine sampling from all of the LOSERS!
I'm so hoping my team completes the challenge and wins the Stanley Cup. Go Wings! |
|
|
| Names |
[May. 19th, 2009|08:09 pm] |
Do you like the Name game for Fedora releases? Do you think you have an awesome name for F12? Great! Want to make sure it actually has a chance of being a candidate? I thought so. So here is how to do that:
1) Read the full instructions on the wiki page http://fedoraproject.org/wiki/Name_suggestions_for_Fedora_12 This is important for a number of reasons. Primarily, if you don't follow the rules then your name has no chance at all of being a candidate. Secondly, if you put down a name "because it's cool" or "sounds awesome" you are just wasting the time of the people trying to actually get the name selection done. This is more than just me, this is also the Board, Red Hat's legal department, and everyone else that has to read through the name list to avoid duplicates
2) Try to be creative, and also give a way to get to something after F12. This is why we do this whole name game to begin with. It is supposed to be a fun semi-brain teaser type of exercise. The more creative the link, the more possibilities you open up. Just make sure the link fits the rules.
3) For bonus points, think of things that fit 1 and 2, and then think 'is this art themeable?' The only reason we now do the name selection so early is so the art team actually has time to try and come up with a theme for the artwork that matches the name. There is no rule that says the artwork has to match the name, but the art team does a really great job at trying their best to match it. So reciprocate a bit, and help them out too.
If everyone tries their best and stick to the rules, the name game can be actually fun. |
|
|
| Translators rock |
[May. 13th, 2009|08:29 pm] |
In the free software world, people often think of 'translators' as those that have the sometimes tedious job of taking the typically horrible error messages, dialog boxes, and other interactive messages that we code in our apps and translating them to other languages. And that is certainly true. It's a thankless job, and the Fedora community has an awesome translation SIG. They have even created Transifex which should help upstreams and other distros with their translation tasks. I find this simply amazing.
I can't help but notice though, that when I read Planet Fedora there are other types of translations that are also valuable that don't get much mention. Recently I blogged about the return of deltarpms to F11. LWN picked this up, which helps our PR a bit after somewhat of a small black eye about them not being in place a while ago. But there is a whole class of users that don't speak English and needed to know. Over the past few days, I have seen a number of blog posts in other languages that are talking about my post and yum-presto. I think the simple act of doing that is really helping distribute news to our global community and I just wanted to say Thanks! to those of you that are doing this.
Full disclosure: I have no idea if those blog posts are actually reporting the correct information, since my non-English skills primarily consist of some very rudimentary Spanish, enough German to be polite but not really say anything, and Japanese swear words. For all I know, they could be saying "Josh Boyer is an idiot for talking about DeltaRPMs and yum-presto." But I trust they aren't and even if they are maybe this post will encourage others to step in and help here too. |
|
|
| Super Saiyan Jack |
[May. 12th, 2009|08:14 pm] |
Is my son a Super Saiyan? The world will never know.
note: That is his real hair. There is no gel or mousse or hairspray in it. It's naturally curly and just happened to look like that today. He was kind enough to indulge Dad with a 'grumpy' face for the picture though. He thought it was hilarious when I showed him the 'glowing guys'.
Edit: Removed the hotlink to the goten picture. Sorry about that. |
|
|
| Fedora 11 Deltarpms: The Doom That Wasn't |
[May. 9th, 2009|08:56 am] |
Recently, Paul blogged about the awesomeness that is yum-presto and Deltarpms. I clarified that we weren't quite ready for F11 updates to have deltarpms enabled, but that we were working on it but wasn't sure if we would have the work completed. LWN picked this up, and apparently so did many other blogs and community members.
So today I would like to draw your attention to:
http://download.fedora.redhat.com/pub/fedora/linux/updates/11/i386/drpms/
Oh, yeah. That's right. What you see there is indeed deltarpms for the first Fedora 11 updates push. So Paul, you can un-edit your blog post now because we should be ready to go for Fedora 11 GA. We'll probably still have a few hiccups here and there, but the infrastructure is now in place.
I'd like to highlight that Seth Vidal, Luke Macken, and Bill Nottingham were the primary factors to getting this complete. They did the createrepo, bodhi, and mash changes needed. I'm always pleasantly surprised when people care enough to start digging in and fixing an issue they care about.
Fedora 11 is now made of that much more awesomeness. |
|
|
| The updates conundrum |
[Apr. 21st, 2009|09:15 pm] |
I have been the primary update pushed for Fedora for about 2 months now. Having done this every week day for that long has led me to some completely subjective opinions on updates, release lifecycles, and the health of our update system (from an admin point of view).
Overall, we have a lot of updates. The metrics page shows a decreasing trend the longer the release has been available. This is generally good, but there are still a very large number of updates that make me scratch my head when I see them, either because the update lacks any useful information at all, or because the update itself seems to make little sense in a stable releaes. Jesse Keating tried to highlight some of these questionable updates but it seems he was either ignored or berated for being "heavy handed".
When I see a package update submitted that just takes a package to the latest upstream release, I always question it in my head (and sometimes in the update). I realize that upstream releases often fix bugs that effect users, however the update should say that at a minimum and it generally doesn't. Many times there is an update like this that seems to just be 'because it's the newest!' Which leads me to my next annoyance.
Fedora doesn't have a rolling release. We have a large number of contributors that work rather hard at keeping with the various package freezes, mass rebuilds, and other administrative work that goes along with creating a release from rawhide. Yet after that release has GAed, we seem to sort of throw that out and just submit updates for updating's sake. It is not uncommon for a package to have identical versions in the three active releases. It is not uncommon to have a brand new package submitted to all active releases at the same time. While I'm quite aware of some of the reasoning behind this, I still question the practice of doing that. And it boils down to this: What value does that add?
Honestly, I have found myself running yum update not because I have problems. Not because I need some new package, but just because 'yum update' is habit (security updates aside).
So... why do I care? Because it adds churn that doesn't seem to be needed. It adds churn to the mirrors. It takes longer and longer to mash the updates repos. It requires more time to sign all the updates. And honestly, it causes churn for our users.
The one benefit (and I say that in a sort of silver lining way) that has come from this is that it has helped me understand some of the weak points we have in the updates process, both in the bodhi backend code and just the actual process. Things like
1) We need more people that understand the bodhi code from front to back. Luke Macken is amazing at fixing things, but I think he would agree that it sucks when I ping him on IRC because bodhi failed a push. I'm getting better about this, but it's an example of a 'single point of fixing'.
2) We need to make the bodhi code a bit more resilient and/or verbose in error messages. Things like revoked updates during certain stages of an updates push can cause hassles and delays.
3) We need less updates. Oh, I already said that.
Well, I see a pile of updates to go sign. Guess I should stop ranting and get back to being part of the problem ;) |
|
|
| Not so posty |
[Apr. 20th, 2009|08:32 am] |
I haven't posted much lately. Neither here, nor on many of the mailing lists I subscribe to. There are a number of reasons. The first is that work is incredibly busy right now. I won't go into more detail there, but it's been sucking up pretty much all of my time.
However, there is a not so mundane reason too. These days, I'm finding that there are very few discussions that are actually worth joining. Now, I still read almost all of them. I still follow things, and I still look for pain points in communities that could actually be larger problems. But these are getting harder and harder to find amongst all the random complaints, mini-flames, sardonic replies, and general abundance of whining.
Too often these days I see posts that start out well intentioned enough, only to devolve into petty bickering and endless example/counter-example. Questioning a change, or making a proposition for a change are usually how it begins. There might be a few emails generated to get a better understanding of why something changed, or what goals someone has. Maybe a few more explaining 'the foo default changed, but you can do X,Y, and Z to set it back'. All of that is generally fine. It seems inevitable though that after the basics of the issues are hammered out, people continue to rant against whatever change happened or was proposed. It can actually be quite draining.
So, I just avoid adding to the noise and refrain from posting at all. In the odd case where something is explicitly addressed to me (or to FESCo, etc), I will certainly comment. I try to do so without being overly sarcastic or inflammatory, which sometimes works and sometimes doesn't. Thankfully, situations like that seem pretty rare. (It is nearly impossible to do all posts without sarcasm. It seems to be the single common trait among all of us.)
Btw, I don't think any of this is actually new. It is perhaps just hitting with greater frequency than in the past. Possibly due to overall community growth, better participation from existing users; I dunno. Maybe the rants will ebb a bit in the future and make the various lists a bit more constructive place to participate in again. At least I hope so.
(And NO. I do not think Fedora needs a Code of Conduct. In fact, I think that would actually make it worse.) |
|
|
| Ubuntu: It just works! ... except when it doesn't |
[Apr. 1st, 2009|02:49 pm] |
I got my Dell Mini 9 today. Liking it very much. I decided I dont have time to reinstall the OS right now, so I'm using the Ubuntu install that came with it. Evil, I know.
Overall, the OS isn't horrible. But I did find it odd that I couldn't ssh into machines. It turns out that the binary wireless driver (ew) has a bug in it that causes ssh sessions to hang. There is a workaround for it, and apparently a fix in a newer driver package that the Dell repo doesn't have yet.
Oh, and the package/update manager thing refused to acknowledge that there were updates available now matter what I did. So I ran the apt-get update command by hand, and then it decided I was serious or something and popped up a dialog. Dunno what that is about either.
Also, I find it odd that gcc is installed on this netbook. Because so many people want to compile by default on a netbook... |
|
|
| netfad |
[Mar. 15th, 2009|07:46 pm] |
I gave into the fad and bought a netbook today. A Dell Mini 9" with 1GiB of DRAM, and a 32 GiB SSD. I saw Kyle McMartin's at FUDCon and thought it was pretty neat looking, and a rather sharp little machine.
I've been asking various people about them for a while now, and trying other avenues before purchasing. I tried to use the OLPC XO as a netbook, but eventually it was just either just too slow or too buggy (running rawhide builds my have contributed to that). The keyboard was also a bit uncomfortable to use.
Then I looked at just using the T60p, which is a great machine. However, all I'm really doing on a day to day basis after work hours is browsing, reading email, and the occasional non-computationally expensive rel-eng task for Fedora. The T60p is large, and really really hot. It makes my legs sweat if I have it on my lap, which can be really irritating.
My biggest hesitation before purchasing the Mini 9 was whether it was a "toy" or something I'd actually use. I'm pretty convinced it's the latter. As Kyle put it, "It's a great couch computer."
I do have a few questions in my head still. Like:
Will running rawhide on this result in earlier SSD failure due to the large number of daily updates, etc?
Will the SSD random read performance make loading applications noticeably faster?
Will I find myself storing lots of stuff to the SSD out of habit? (I don't think so).
Time will tell, and I think it's going to be a great purchase. |
|
|
| Linux is my power drill |
[Mar. 13th, 2009|12:09 pm] |
In the movie Pi, a little girl asks a mathematical genius to do complex math in his head and he can answer without really thinking too hard. During the course of the movie, he becomes obsessed with finding a particular number and eventually uses a power drill to destroy what he thinks is the mathematical center of his brain. At the end of the movie, the girl asks him to do a rather simple division problem and he smiles at her and says he doesn't know the answer.
I used to be fairly good at fixing Windows problems, or recommending Windows applications, etc. This was back in my younger days before I was introduced to Linux. I was by no means a wizard, but I knew it well enough to be able to know what services did what, etc. Then in 2002, I started working with Linux full time and haven't really used Windows since. Yesterday a co-worker asked a Windows question that I'm sure I knew the answer to at one time, but I couldn't for the life of me tell him what to do. I simply smiled (to myself, since this was via IRC but whatever), and said I didn't know. It made me think of Pi. |
|
|
| OMGWTFBBQ |
[Mar. 2nd, 2009|11:47 am] |
Ok, this is really about BBQ. Our grill died earlier this winter. We had been regularly replacing the burners and heat spreaders as they rusted out, but this time the control knob seems to have rusted to an on position somewhere between high and low. This makes it somewhat difficult to actually control the temperature of the grill and is possibly unsafe.
We had been eyeing grills for a while now but were sort of torn on what to get. Charcoal grills make better tasting food, but take longer to start, require a bit more maintenance, etc. Gas grills are great for convenience, but can make grilling some foods (ribs for example) slightly more of a pain. Then my awesome wife happened upon a dual fuel grill at the local store in town. The Char-Griller Duo #5050. We bought one yesterday.
Assembled it over the next couple of hours and was really please with the quality of construction on it. Really it's just two grills mounted on the same frame, but that is exactly what I wanted anyway. The side burner attached to the LP tank is awesome for starting the charcoal with in a charcoal chimney. I was so excited to have a working grill that we cooked up some burgers on it as soon as I had it assembled. Yummy.
I'm looking forward to lots of more traditional BBQ from now on.
(And I promise to write some more Fedora related posts soon, but who can resist blogging about BBQ!) |
|
|
| rsync hates me |
[Feb. 24th, 2009|01:06 pm] |
Went to login to my x86_64 box this morning and got asked for a password. Which is odd, given that I have ssh keys. So I logged into it, and got a funny looking prompt. Scratched head a bit, did ls. Noticed nothing that is normal there was there. Ran dmesg, noticed oops from ext3 on sdb2. Began to get that feel of dread. Looked more closely at home dir, noticed it now contained i386, ppc, ppc64, and x86_64 directories. Brief flash of panic. Tried 'cd /media/disk', returned error. Realized what happened. Swore a little. Turned the machine off.
Apparently sometime during the night ext3 decided to crap itself on my external USB drive. This caused the mount point to not be accessible. That caused my fedora mirror script (which is obviously a bit buggy in hindsight), to not be able to pushd to the mount point before running rsync to suck down the latest rawhide. Since the rsync is run with the --delete options, it dutifully pulled down rawhide, and then it deleted the rest of my home directory. Awesome. Wonderful day so far. |
|
|
![[info]](http://l-stat.livejournal.com/img/userinfo.gif){kind=small}
jwboyer's journal