Bryce Lewis died at this intersection on 9/7/2007:

The crash site, an intersection that sees 30,000 cars a day, is notoriously dangerous for cyclists ("scary as shit," one Point 83 poster called it). Although Seattle Department of Transportation (SDOT) spokesman Gregg Hirakawa says SDOT has no specific records indicating that the intersection of Eastlake and Fuhrman avenues is more dangerous than other trouble spots. (SDOT doesn't keep crash statistics by intersection.) "This is a major connection [for cyclists], and it needs improvement," says David Hiller, advocacy director for the Cascade Bicycle Club.

Update: Councilmember Jan Drago (chair, Transportation Committee) and Grace Crunican (Director of the Seattle Department of Transportation) are the most appropriate contacts to voice concern over this intersection.

Technorati Tags: cycle

(As an aside, learning how to ask smart questions can help avoid time wasted over an “it doesn’t work” exchange.)

A good admin will ask many questions, and try to answer all of them, either by asking, or if possible, running the commands. The exact order of steps is not important so much as checking a wide variety of symptoms. Another admin may check df to see if the filesystem is actually mounted before running sudo lsof | grep mountpoint for open files. These commands are (usually) quick to run, and (usually) additional terminals can be opened to continue should one command be slow or block.

Junior admins should spend as much time learning the system as possible. Books such as the UNIX System Administration Handbook should help, as will delving through the manuals or other available vendor documentation. Consider also reading Unix Debugging Tips.

Quick System Overview

Be aware that an error message may not be from the problem, but something affected by the real problem. This is why some initial time spent getting a feel for the system will almost always help:

• How long has the system been running? Use the uptime command, which also shows the system load. (Old RedHat 7.2 systems, if left running long enough, would eventually show 100% CPU usage due to a procps bug.)

• Abnormal CPU, disk, or memory use? top, free on Linux, vmstat 1 or iostat can all be consulted in a matter of seconds. Processes stuck in D state or high I/O wait times may indicate an overburdened disk system.

  Note: df and similar commands can hang on a broken NFS mount. With multiple SSH or a screen session, this probably is not a concern. If one only has a single console link, the blocked df process could hamper diagnostics.

• Anything strange in the system logs? This is more relevant should the system have recently rebooted. Check dmesg, and wherever else the logging daemon sends this information.

• RAID or SAN or LVM state? More relevant on database and file server systems than throwaway desktop or service nodes. Due to the plethora of different arrays and devices possible here, best to write a wrapper script that figures out what the system has, and emits appropriate diagnostics.

Monitoring can help collect and alarm on abnormal CPU, memory, or disk usage. However, these should not page, as a busy but perfectly operational system may show high CPU, and waking someone up for false alarms hurts their ability to work on real issues.

Level 52 is tricky the second time through, and requires cutting corners to reach the falling enemy in time. The third time through, a different strategy is required, as the enemy is now too fast to reach in time. Instead, one must dig to get past the upper two enemies, then fall into the usual pattern.

Curtains fall at level 357, where the now twice sped enemy always intercepts one before crossing the first ladder (only eight bits were allocated for the level counter, as following level 256, it wraps, hence level 101 being displayed):

Note: lode_runner1.po on the Asimov Apple //e archive is corrupt at level 130-something. I am using lode_runner.do.gz. If paranoid, advance through the levels with control+U or control+6 to confirm the levels look right.

Next, Championship Lode Runner.

Do not much like Polaroid film. To be fair, I am using a Holgaroid back to a Holga. Peeves: loading the film is tricky, though worse is extracting the film for exposure, as the paper pull tabs risk breaking, leaving the film stranded, and the ninth exposure usually pulls the tenth out along with it. Another problem: odd streaks down the film, which appear to line up with the pull tab paper: somehow this is getting compressed against the film through the rollers?

Film quality excellent, with a deep glossy black, though getting the exposure right was very tricky, as the Holgaroid metered nothing like my DSLR measured. A incident light meter should give better readings, but I do not own one.

Digital or regular film photography, combined with post processing and glossy printing, while slower, should produce similar effects. I could see using Polaroid to take test shots for medium or large format needs. However! Given that Polaroid is going the way of the Dodo, best to avoid the resulting high film prices until and if an instant film market emerges following 2009.

The Company

“Dire financial straits notwithstanding, Polaroid paid senior executives and directors a total of $6.3 million in bonuses, consulting fees, and lump-sum pension payouts in the months before the [bankruptcy] filing. Payments included $1.7 million in incentive comp to former CEO DiCamillo, while former CFO Judy Boynton got $300,000 in severance, a $510,000 stock award, and a $638,000 lump-sum pension payout. (Boynton, now the CFO of Royal Dutch/Shell Group, is listed as an unsecured creditor, for an additional severance of $600,000 she is still owed.)”

“Polaroid retirees had feared the results of a sale to OEP, and that fear was justified. After June 28, the company's cash balance plan was terminated and handed over to the federal Pension Benefit Guarantee Corp., meaning many retirees had their pension payments slashed. Employees on long-term disability received letters informing them that they would not be hired by the new company and that their benefits were being terminated. Indeed, the Massachusetts attorney general's office had difficulty convincing OEP, as owner of the new Polaroid, to sponsor the retirees' supplemental Medicare plan, even though that sponsorship costs nothing except time spent keeping the books.”

    — What’s Wrong with This Picture?

• Dubliners
• A Portrait of the Artist as a Young Man

Based on conversations with our Dublin staff, Dublin is very much so not the town James Joyce wrote about.

if ( $line =~ m/^ { $/x ) { $rule_target[-1]->[-1]->{_subrules} = []; push @rule_target, $rule_target[-1]->[-1]->{_subrules}; }

First, consider instead providing an Object Oriented interface, thus hiding the Big Blob. However, this may defeat a “well, I’m just trying to mangle X into Y, not waste time with class struggles” coding effort. Whether OO makes sense depends on the project. A standalone data conversion script probably does not justify OO. Code that other code will use, or a service interface, especially one used by other groups or users, will likely benefit from OO.

Secondly, Big Blobs could be a solution looking for a problem. The coder knows how to parse data into the blob, then iterate over the mess, but never considers whether a blob should have been used.

my %big_blob = load_from_file($filename); upload_to_database(\%big_blob);

In many cases, the entirety of data need not be loaded into memory, and instead only the minimum necessary data retained in memory before acting on it:

while <$fh> { my %line_data; # ... parse line into line_data hash # upload line_data contents to database $db->... } continue { if ($. % 1000) { $db->commit(); } }

This method scales better, as it no longer is bound by memory, and will not require DB_File or refactoring should the data set exceed available memory. The question: “do I really need to parse all the data to memory, or is there a more efficient solution?” will help prevent inappropriate use of Big Blobs.

Technorati Tags: Perl

# run just past the top of the hour # as many other things run then 7 * * * * rsync -e ssh -az --delete /source desthost:/dest

However, this solution has a major edge case that can bring down the system. Worse, simplistic attempts to fix this fault can result in rsync not running.

Technorati Tags: cron

Resource Usage Spiral

rsync, run directly from cron, will run until the file transfer has completed, or a bug causes the rsync to hang, or some fault causes rsync to terminate unexpectedly: bad memory, a system reboot, a cowboy admin getting frisky with kill -9, and so forth. The most worrying is when rsync operates normally, but takes too long to complete. In this case, cron will launch subsequent rsync, and over time, if the other rsync never exit, the system will eventually fail.

The rsync --timeout=999 option is useful, but not complete. This option ensures rsync will eventually exit. However, it will not help when crond launches the next instance while rsync is still transferring files. A wrapper script around rsync is necessary to prevent multiple instances from running.

Naïve Locking Schemes

< Lovecraft> thrig: its a matter of making a script that looks for a lock file. if exist <lockfile>, don't start. Ifnot exist <lockfile> make one and start. < thrig> Lovecraft: and what else? < Lovecraft> Thats it

File locking requires slightly more thought than looking for a lock, and not running if it exists. Lovecraft’s incomplete solution will cause problems should the system crash, should the system restart normally without the script handling the shutdown signal properly, should the script be terminated by kill -9 (even if it trapped other signals properly), should a hardware fault cause the script to exit, should a system configuration issue prevent the lock file from being created. Worst, if poorly written and poorly monitored, nobody may know the rsync process is not running—until some need reveals the lack of recent files on the destination server, which could be weeks or months since things went awry.

#!/bin/sh # Skeleton locking with signal handling. # Much more sanity checking required! PID_FILE="/var/lock/foopid" cleanup () { rm -- "$PID_FILE" } trap "cleanup" 0 1 2 13 15 # Race condition probably not a concern, # due to the infrequency of rsync runs. [ -e "$PID_FILE" ] && exit touch "$PID_FILE" || exit rsync --timeout=999 -e ssh -az \ /sourcedir/ desthost:/destdir cleanup

Lock files introduce a new problem: the lock file suggests—but by no means proves—the associated process is actually running. That is, depending on the implementation, the rsync may be running, and no lock file created—a permissions problem coupled with an “ignore errors creating lock file” implementation—or the rsync process may not be running, and a lock file exists, for the various reasons outlined above. Consider locking against the process name, not a file on disk.

Ideas for Improvements

Software besides crond, such as CFEngine, provide locking functionality. If possible, use these solutions, as they are likely better tested than an in-house shell script. CFEngine or similar configuration management software can also ensure the lock directory exists and has the correct permissions, if a lock file scheme is used.

With Perl, one solution is to lock the script itself via the special __DATA__ filehandle, which will avoid the various problems of an external lock file. I generally prefer Perl over the shell, as the shell lacks the equivalent of perl -c, makes writing unit tests difficult, and has a number of scary edge cases that can delete entire disks.

Another implementation wraps the rsync inside a loop. This prevents multiple rsync from running, but pushes the locking and monitoring to the wrapper script instead. This runs the process an hour after the previous one completes, not once every hour.

while sleep 3600; do rsync ... done

Monitoring whether rsync actually did anything is another can of worms. This monitoring should not be an e-mail, as if frequent it will become cron spam, filtered and ignored. Monitoring should also not report transitory errors, where the target is temporarily unavailable, as investigating false alarms waste time.

1. Thou shalt grok Unicode!
2. Unicode-processing issues in Perl and how to cope with it

More Unicode.

Technorati Tags: characters, utf8

• NWS Seattle
• Pacific NorthWest MM5 Weather Forecasts (neato technical detail)
• Hour-by-hour precipitation
• Column-integrated cloud water

The Japanese Garden in Seattle offers several photography sessions each year. These let you use a tripod and other gear not normally permitted in the garden. Well worth the money, as there were only three other photographers, and the weather mostly cooperated. Most useful lenses were the 18-70 kit zoom and 105mm macro.