This... this is the absolute best bug I've come across in the last couple of months. A quick summary.
-
Show this thread
-
We have a big blob job that fits volatility surfaces across a variety of markets. Pretty standard stuff.pic.twitter.com/labyi92pwe
1 reply 0 retweets 5 likesShow this thread -
In this case, "series" is an array option underlyings, and "dates" is an array of dates. QMapReduce permutes the two arrays to create a job stream of map functions. Cool.
1 reply 0 retweets 4 likesShow this thread -
This has worked without fail for the last year or so, but failed today. Well, it didn't actually fail, it just froze, not doing anything: I checked the backend... there were not actually 8 map jobs running.pic.twitter.com/41yG00J56z
2 replies 0 retweets 4 likesShow this thread -
Replying to @kchoudhu
This logging makes it look like you have a time.sleep(1) somewhere, plus some other code that takes 1-2 millis to execute
1 reply 0 retweets 1 like -
Replying to @macrocephalopod
Correct, but it's not the actual problem (I posted the actual controller code later on).
1 reply 0 retweets 1 like -
Replying to @kchoudhu
Oh yeah, didn't think I had diagnosed the problem. Just noticed because I'm one of those weirdos who always defines next_update_time to be the end of the next second/minute/whatever, and then time.sleep(next_update_time - http://datetime.now ())
1 reply 0 retweets 3 likes -
Replying to @macrocephalopod
Oh right, generally I do that too, it just didn't seem important enough to do here :D
1 reply 0 retweets 2 likes
I just really like round numbers.
-
-
-
Replying to @theemilyaccount @macrocephalopod
They're right though, I was sloppy with my sleep code.
1 reply 0 retweets 2 likes - Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.