Older blog entries for jbowman (starting at number 24)

Mmm. Been reading Illuminatus a lot lately. That book, combined with sinfest are usually enough to bring me out of just about any funk. So far, it's working.

The updated version of openssh that RedHat released as an errata package works nicely now. It fixed the odd MAC corruption error-thingy that I was experiencing when connecting to my ssh-2.3.0 box at work, which is a plus. However, it still does the odd "keep the session alive" trick. Time to go Bugzilla something else.

Noticed my monitor was doing strange wavy/sweepy lines again today. Started getting worried that it was going bad, so I went through and tightened all the connections and unplugged the component cable from the back (my monitor has dual-inputs :). Those two actions combined seem to have fixed things. My screen is once again it's usual self, which is at a slightly slower refresh than I'd like (the default 60hz, when the monitor should be able to do at 70). If stare at large solid-colored portions of the screen (like Advogato's background while I'm typing this) I can see the monitor refresh itself. Ugh, eyestrain.

I need some way to generate custom modelines for Xfree86 4 for a 1920x1600 resolution. Anybody have any, or know of a program to generate them for me? The programs I've seen so far don't seem to support 1920x1600. Then again, from most of the people that have seen my desktop, 1920x1600 doesn't seem to be compatible with their eyes. And yes, I can read it just fine. :)

suso was grumbling was grumbling about his randomsig program not parsing newlines in his quotes file. Browsed through the thing a bit with him and help him find out where the heck it was eating the newlines. Turns out the module he uses to do linewrapping was the culprit. I suggested he just borrow the necessary functionality and implement it in randomsig, but he just sighed and shook his head. So I'm going to try and hack it in myself, just as soon as I find that copy of the latest randomsig I have lying around somewhere... Hrrrrm....

Today was a surprisingly productive day. Did a round of security upgrades at work (RedHat security announcements are always so much fun). Pondered really odd masquerading problems that one of our clients was having (still no clue what's going on... yet). Finally managed to get ahold of a technician and get a time scheduled for them to inspect my television, which has been doing this strange crackling thing with the picture. The tech was actually quite helpful, suggesting that something inside the tube itself might be arcing, thus explaining the "snap" sound I hear and the instant of screen flicker that I get. I'm glad I got the extended service plan on the thing now...

Tomorrow shall be more pokings at the strange masquerade problems, a nice meeting in the morning with some more clients about setting up another Linux colocated server (whoohoo! :) and more general fun. Ah, it's nice when a Monday gets off to a good start. Of course, this just means it's going to ambush me during the week sometime... Mondays always get their revenge.

Days RedHat has ignored my glibc bug report: 7
Days since SSH first promised to take a look at my patches and haven't actually done so yet: 24

[ "Save tonight and fight the break of dawn... Come tomorrow... Tomorrow I'll be gone." ]

9 Oct 2000 (updated 9 Oct 2000 at 21:30 UTC) »

Well, I had a nice long diary entry fired up, and then Netscape died a horrible death while I was moving my mouse across the window. I didn't have anything else open, mind you, it just.... died, which is odd since Netscape tends to be really stable for me. Guess I've just been lucky. I'm still waiting for the Mozilla nightly binary builds to include the security manager and SSL support. After seeing how nice the latest builds are, that's the only thing really keeping me from using Moz on a regular basis (I must have SSL suport for work. It's not optional).

crudman: Sent you a reply about my BP6 problems. Sorry for the delay, it's been a somewhat odd and confusing week.

My karma on /. has finally finished it's downward spiral to the 50-point cap. Watching about 10 surplus points of karma slowly burn off was amusing. The funny thing was, I lost it *all* to metamoderation. Guess those rampant metamoderator abusers are continuing to smash the system. It'd be nice to see a new face on the News for Nerds scene. Something with the slashdot style of story postings and a revamped and improved moderation system. Yes, I know about kuro5hin. That's not what I'm talking about/looking for.

I'm becoming... annoyed... with RedHat and SSH both. Every time I've registered a bug in bugzilla, it's been ignored for weeks on end (not even getting assigned to someone), and then winds up being closed when someone else bitches about the problem on the mailing list without bothering to submit a bug report. And this is when bugzilla is supposed to be the end-all, be-all, if-you- don't-submit-it-we-won't-look-at-it thing. It's not like I'm asking much. I'd just like them to actually assign it to someone rather than leave it sitting there with a 'new' flag.

Then there's SSH, which emailed this week to let me know that they have, indeed, received my emails about the horrendous problems with their specfile for 2.3.0, and that rest assured, they will be looking at my submitted patches. Really they will. Yes, they're sorry they spent several weeks completely ignoring me. Feh.

Since I installed RH 7.0 at home, I've had a chance to play around with openssh. It's a little friendlier than the 'reference' ssh clients, with far, far superior ssh1 compatibility. It just has some oddities that I need to poke at to see if it's a bug or an intentional design thing. For example, when I log into my gateway, I su up to root, dial into work via "wvdial &" and then log out of root. When I go to log off the ssh connection, reference SSH 2.3.0 closes the connection like I'd expect. OpenSSH acts like I've left something being X-forwarded back, and logs out my session without fully closing it. Perhaps I'll submit it as a bug report to bugzilla tomorrow...

cdent leaving has brought back to mind the things I dislike about my current place at Kiva. The positives so far continue to outweight the negatives, but Chris' leaving has dredged them back up, and I'm still shaking them back off again. My motivation's been shot to heck this past weak as a result of that, and due to the continued rudderless drifting in regards to my future with the company. What exactly will I be doing there, now that about 3/4's of my workload has been lifted (it went to a newly-formed sister company the same day Chris left)? jlf has been trying to build some answers for me, for which I'm immensely grateful. The next few months will be "interesting" to say the least.

Wow, this was a long entry. I need to kick myself back into gear and get back to my usual self. Being depressed sucks, and leads to long rambling rants.

And I think I agree with squiggy. We need a giant space rock or two to take care of things.

Days RedHat has ignored my glibc bug report: 6
Days since SSH first promised to take a look at my patches and haven't actually done so yet: 23

[ "Now I whoop you with my Pimp-Ninja powers!" ]

Trimmed back my cert list a bit. I made a promise when I started on my cert rampage that I'd go back and check my certs from time to time to make sure people continue to earn them. Wound up dropping several people off my cert list, and wondering in one or two cases what possessed me to cert someone. Ah well. Tomorrow I'll continue checking up on my certs.

Pondering ways of automating the security audit processes that I'll be going through here at work soon on some of our colocated servers. The whole thing seems like it should be easily scriptable.

Cleaned up my desk at work today. I'm no longer hidden behind a wall of Coke cans and old editions of the New York Times. I think I scared a few people here in the office when they saw me removing the "holy wall of cans" from my cubicle.... :)

[ "The holy trinity are there in her purse: A condom, Obsession, and mace." ]

Hrm. It appears kernels just don't like me.
The 'default' RedHat kernels for 7.0, 6.2, and 6.1 all lead to random segfaults and sig11s on my BP6. 2.4.0-test5 leads to random sig11s. 2.4.0-test8 and beyond all work cleanly. I don't even get odd APIC errors anymore. Weird.

Had fun trying out redhat's "preview" version of glibc. It apparently has some broken things (like, ohhh, locale support for example) so I wound up rolling back to the 'normal' version they include with 7.0. Made a nice shiny Bugzilla entry for the RedHat folks too. Oooh, ahh.

Friday was the last day for a bunch of people at Kiva. It's sad to see them go, as I'll miss most of them terribly. It'll take a while to get used to not having them around (and to get used to the organizational changes that are happening as a result).

Need to get around to hacking some more on logwatchd. Making it into a nicely scalable log multiplexor looks to be a fun learning experience.

[ "Microprobe thingamajig goes right here..." ]

Did some more beating-of-the-head with my gateway machine last night. For whatever reason the default kernel image would fail to load (it would uncompress, say it was booting, and then just hang there without ever going into the usual bootup sequences). However, my backup kernels worked just fine. Figured the boot sector may have gotten corrupted some how, so i reran lilo. Same problem.

So, I rebuild a new kernel, thinking perhaps it was the kernel image itself that was the problem. After all, I discovered, it was exhibiting the same symptoms regardless of whether it was the default boot image or not. 2.4.0-test5 had been working flawlessly on this machine for over a week, and then all of a sudden it died a quick, painless death after a single reboot. Sigh again.

So, I wound up replacing the hard drive late last night and installing RedHat 7.0 on the darn thing. Things now work like they're supposed to, even with -test5 again, so I suspect it was the old 600mb seagate harddrive I'd been using.

On the plus side, it means I have more space to play with and thus have a system that's less a "well, what can I fit into 600mb of disk" exercise. Seeing 128mb of that 600 total get eaten by swap was less than helpful for determining what packages to install. :)

Iptables continues to impress me with the sheer raw power you have over your packets. I've been playing around with some of the kernel patches included in the patch-o-matic CVS stuff, and if/when they get finished up and included, things will be even nicer than they already are. I need to look up the way the various state flags like "RELATED" actually work with the different protocols, but that's one of the last things left for me to grok in the incredibly complex rules dancer sent me. The other thing left to examine is the tcp-flag checking rules, but that's mostly just working out the logic behind them.

I'm also getting tired of dealing with the assorted oddities of my BP6 motherboard. ABIT just announced their VP6, which is a second-gen BP6-style mobo that's FC-PPGA compatible, so I may go with that. I just plain dislike random segfaults in things that shouldn't segfault. Even a single non-overclocked CPU, it has issues. Sigh.

[ "This is your life, and it's ending one minute at a time." ]

Hrm. Installed RedHat 7.0 last night. Found it refuses to upgrade 7.0 beta installs due to some strange python segfault in the installer. I'd report it as a bug, but I know the response I'd get: "We don't support or recommend upgrading a beta release to the released version". Feh

jlf continues to poke and prod at me in an attempt to nudge me around some of the obstacles he's encountered in the industry. Thanks for being a mentor, Jeremy. In discussing my last diary entry with him, I think I may have described things a bit inaccurately. The "waiting game" I grow tired of isn't the wait time caused by waiting on customers to get answers back, but rather the wait time in finding out where the heck our end of things stands. It's the complete lack of information being handed out that irks me, not the actual act of waiting itself. To analogize it to forked processes, it's the difference between being a zombie child process that gets reaped and a zombie child process that doesn't and just sits around all day...

ahosey and I have been discussing other uses for spinwebd/logwatchd (I really do need to change the name soon). What we want/need at work in the not-to-distant future is a log multiplexor.
For example:
We have several thousand customers hosted on our webserver.
We want to be able to sort their webpage statistics out for them.
We can't do this because, well, the log traffic for that number of users is absolutely obscene.
So, what we can do with logwatchd is setup the appropriate regex and have it append the contents of the line it matched to a specific file based on the regex. This would give us the ability to record all of user foo's log entires into file bar on the system. And the nice thing is that unlike a bulky log-parsing/sorting script, this works "live" as the data comes in.

After saying that logwatchd could already do that with very little in the way of changes, Adrian pointed out that with as much traffic as is being generated, that's an awful waste of cpu cycles. We got into a discussion of LRU caching of open file descriptors, and how best to organize things. Now I just need to get my workstation at home up and working again, and I'll be able to add a new module to logwatchd.

Adrian also pointed out that my feelings of "Surely someone's done this before!" aren't all that unusual. His explanation lay (basically) along the lines of "Lots of people have ideas, few of them actually get implemented, and even fewer are ever shown to the world".I guess that's what I like the most about the spirit of open source software. You get the chance to reuse other peoples ideas (and, assuming the license is compatible, their code as well) to inspire new solutions to problems.

[ "We are the all-singing, all-dancing crap of the world." ]

Hm. It's been a while since I last posted. Ah well. It's been a mind-numbing week.

I'm convinced some of my coworkers are clueless. It's to be expected, really, when you consider the general statistics. But why is it that the people that I'm forced to rely on for beta-testing things are also the ones that cannot remember what they were told the week before? *Sigh*

Spent most of last week poking and fiddling with my account-tracking system. For unknown reasons, last week was completely mind-crushing and depressing. I really didn't accomplish all that much, which is disappointing.

Feeling better and hopefully out of last week's funk today. I spent the weekend completely slacking off and sleeping, which helped a lot I think. Poked and fiddled with my iptables firewall rules last night to get a silly MMOG from Micro$oft working through my NAT setup. dancer provided me with the shell script I use to create the rules from, and it's been nice walk into some of the more advanced things you can do with iptables.

Getting tired of the coporate waiting game here at work. "So, how are we progressing with asking our colo customers to upgrade their RedHat 4.2 machines to something more reasonable (and supported by RedHat?' -> "Yeah, we're working on it." Feh. The portion of my brain that likes to champion lost causes is starting to come around to the fact that I may need to lower my expectations a bit...

Need to get back to hacking on spinwebd. Tonight sounds good, yes. I had a nice idea for how to handle targets over the weekend, I'll see how that pans out...

[ "It's the perfect time of year, somewhere far away from here. I feel fine enough I guess, considering everything's a mess." ]

Hrm. It's been a couple of days. Spent most of Saturday recovering from Friday night's escapades. No real hangover to speak of (lucky me!), but the entire event was physically exhausting for some reason. The rest of the weekend was nice and slow, a very relaxing break. Accomplished surprisingly little of value this weekend. Ah well.

Spent a chunk of today hunting down problems with my scheduling software. It just doesn't want to go away and leave me alone it seems...

Found wetdog had gotten himself an account on here. Cert'd him to apprentice, since he's got a lot of potential. Now if he'd just GPL that tic-tac-toe program... ;)

Got in touch with not one, but two old friends in the past 24 hours. Wow, talk about flashbacks.

[ "Do you remember the time we first met, girl?" ]

The Morning After Report

Oog. That was an amazing evening. Definately a night to be rememered always. Fortunately, I didn't get so hammered as to not be able to remember it. :) Discovered I have a particular fondness for cinnamon schnapps. Definately need to get some of that stuff for around the house.

Realized I hadn't certified dancer yet, so I did. Dancer, my personal -test8 expereince here at home has been quite stable on my workstation. No crashes/lockups/errors in 3 days straight so far. No iptables though, but that's what the gateway machine is. YMMV, I guess.

[ "We are the Cartoon Heroes..." ]

15 Sep 2000 (updated 15 Sep 2000 at 07:01 UTC) »

Got my dialup gateway machine finally up and running this afternoon (Thursday). The advanced features of iptables, like DNAT and TOS mangling, are just an absolute blast to play with. I wish 2.4.x would hurry up and get here so we can roll it out at work or something. :)

I worry that I'm becoming something of a work-a-holic. I think part of the "problem" is that I love the work I do. I should clarify this though. I love the *work* that I do, but not neccesariy the job that I do it for. While my working environment may not always be the greatest, the past few months have been filled with projects and things that, when you substract the "job" away from them, are things that I really enjoy doing. I wonder if loving the work and hating the job is the same as hating the work and loving the job. Or are they two completely opposite ideas?

Then again, maybe my problem is that I'm masochistic. A psychologist would have a field day with my brain, I'm sure...

So, it's Friday now, so I'm 21 now. Where are the flying cars? I was promised flying cars! Or at least omnipotent power or something, darn it. It's the least the world could do.

[ "You load 15 tons, whadda ya get? Another day older and deeper in debt..." ]

15 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!