2007/12/05
IRC [04:37] *** IEF parted the chat.
IRC [04:37] *** IEF joined the chat.
IRC [10:00] *** Alethes joined the chat.
IRC [10:04] *** tino__ joined the chat.
IRC [10:04] *** tino__ parted the chat.
IRC [10:14] *** eth01 joined the chat.
IRC [10:14] *** eth01 parted the chat.
IRC [10:49] *** Alethes parted the chat.
IRC [14:05] *** jbellis parted the chat.
IRC [21:13] *** asteets joined the chat.
IRC [21:13] <asteets> hello
IRC [21:14] <asteets> ns_adp_stats is giving me strange results
IRC [21:14] <asteets> namely, everything seems to be off by one
IRC [21:16] <asteets> in other words, the size is wrong
IRC [21:16] <asteets> but blocks gives the actual size (as reported by stat, ls, etc)
IRC [21:16] <asteets> very strange
IRC [21:17] <asteets> erh, brb
IRC [21:17] *** asteets parted the chat.
IRC [21:17] <Dossy> strange ...
IRC [21:18] *** asteets joined the chat.
IRC [21:18] <asteets> ok
IRC [21:18] <asteets> just curious if anyone is seeing the same weirdness (4.0.10)
IRC [21:19] <Dossy> honestly, I haven't looked
IRC [21:19] <Dossy> ns_adp_stats is off-by-one?
IRC [21:19] *** Dossy parted the chat.
IRC [21:19] *** Dossy joined the chat.
IRC [21:20] <asteets> oh sweet hey Dossy
IRC [21:20] <asteets> at least through nscp
IRC [21:20] <asteets> haven't tried actually embedding it in an adp
IRC [21:23] <asteets> do you have an aolserver instance up that you can look at? I want to know if its just something weird for me or if it's a real issue
IRC [21:23] <Dossy> LOL, yeah I think I see what you're talking about
IRC [21:23] <Dossy> like here, the value for mtime is actually the inode. LOL
IRC [21:24] <Dossy> everything says "ino 0" -- that's broken.
IRC [21:24] <Dossy> wow, nice bug find :)
IRC [21:24] <Dossy> you want to fix and commit?
IRC [21:24] <asteets> thats the thing
IRC [21:24] <asteets> the code looks right
IRC [21:24] <Dossy> really? oh grr.
IRC [21:24] <asteets> nsd/adpeval.c
IRC [21:24] <asteets> NsTclAdpStatsCmd
IRC [21:25] <asteets> I thought maybe I had a busted build or something
IRC [21:26] <Dossy> the stats cmd looks right
IRC [21:26] <Dossy> but the code that puts the entry in the hashtable must be wrong
IRC [21:26] <Dossy> look around line 1262
IRC [21:26] <Dossy> (in CVS HEAD)
IRC [21:27] <Dossy> er, line 467
IRC [21:27] <Dossy> sorry.
IRC [21:27] <Dossy> 441 pagePtr = ParseFile(itPtr, file, &st, flags);
IRC [21:30] <Dossy> wow, you're right, all of this looks right. this is _strange_
IRC [21:30] <asteets> but if that code was wrong, I think it would bust all of the adp parsing logic
IRC [21:30] <asteets> yeah
IRC [21:30] <asteets> real strange
IRC [21:31] <asteets> I started out trying to figure out how the adp cache / size limits worked, but then I got here :-)
IRC [21:38] <asteets> anyways I need to run, but I will check back in later, probably tomorrow
IRC [21:47] *** jhavard joined the chat.