2016-08-13T00:49:34Z aries_liuxueyang quit (Ping timeout: 240 seconds)
2016-08-13T00:49:57Z aries_liuxueyang joined #ccl
2016-08-13T01:55:36Z pipping joined #ccl
2016-08-13T02:01:10Z Audrow joined #ccl
2016-08-13T02:02:08Z Audrow quit (Client Quit)
2016-08-13T02:12:21Z pierpa quit (Ping timeout: 244 seconds)
2016-08-13T03:36:55Z milanj quit (Quit: This computer has gone to sleep)
2016-08-13T04:20:11Z rme quit (Quit: rme)
2016-08-13T04:20:11Z rme quit (Quit: rme)
2016-08-13T08:24:54Z DavidGu joined #ccl
2016-08-13T08:46:19Z DavidGu quit (Quit: DavidGu)
2016-08-13T08:46:54Z DavidGu joined #ccl
2016-08-13T09:30:01Z DavidGu quit (Ping timeout: 250 seconds)
2016-08-13T11:20:24Z pierpa joined #ccl
2016-08-13T11:43:47Z milanj joined #ccl
2016-08-13T11:50:00Z rme joined #ccl
2016-08-13T11:56:01Z rumbler31 joined #ccl
2016-08-13T12:35:03Z rumbler31: good morning #ccl
2016-08-13T13:07:06Z vydd quit (Remote host closed the connection)
2016-08-13T13:16:32Z DavidGu joined #ccl
2016-08-13T13:20:55Z DavidGu quit (Ping timeout: 244 seconds)
2016-08-13T13:47:31Z milanj quit (Quit: This computer has gone to sleep)
2016-08-13T14:20:28Z Yan__ joined #ccl
2016-08-13T14:22:03Z Yan__ quit (Client Quit)
2016-08-13T14:22:41Z yang_ joined #ccl
2016-08-13T14:38:40Z rumbler31: is rme lurking?
2016-08-13T14:38:53Z rme: yes
2016-08-13T14:51:45Z rumbler31: ah, I'm the guy who filed the ticket about the fstat thing
2016-08-13T14:52:50Z rme: OK.  So, is it working after all?  Or is there still something going wrong?
2016-08-13T14:56:21Z rumbler31: every time I mount the filesystem, the results of those calls change, and it seems like when its a positive number, the error won't occur
2016-08-13T14:57:43Z rumbler31: I confess that I don't know exactly why it works sometimes and not others, or why random values aren't always catastrophic
2016-08-13T14:59:55Z rme: CCL uses (nth-value 6 ...), which is st_blksize, to size a buffer.  If that value is negative, CCL will try to make a buffer with a negative number of elements, which can't work.
2016-08-13T15:02:01Z rme: If the NTFS filesystem driver is returning junk for st_blksize, this is going to confuse CCL.
2016-08-13T15:02:52Z rumbler31: do you know what the field is supposed to be a size for?
2016-08-13T15:03:31Z rme: According to the stat(2) man page, "optimal file sys I/O ops blocksize".
2016-08-13T15:05:10Z rumbler31: hmm ok
2016-08-13T15:05:18Z rme: I could change ccl::optimal-buffer-size to ignore clearly ridiculous st_blksize values, I guess.
2016-08-13T15:05:58Z rumbler31: heh, sometimes it returns junk values, sometimes it returns 4096 like other volume mounts
2016-08-13T15:06:10Z rumbler31: ok, i'll check with the vendor
2016-08-13T15:06:24Z rme: Have you tried a simple C program to stat files on the NTFS volume?  If that shows the same sort of varying values, then that would certainly point to an issue with the NTFS driver.
2016-08-13T15:06:50Z rumbler31: no not yet but I plan to
2016-08-13T15:07:33Z rme: OK.  I think that would be a good idea.
2016-08-13T15:11:05Z rumbler31: i will update the ticket with the results from that
2016-08-13T15:11:53Z milanj joined #ccl
2016-08-13T15:11:55Z rme: If you see the same results from the C program that you see from CCL, then that will rule out CCL as the source of the problem.
2016-08-13T15:12:18Z rumbler31: yea that makes sense
2016-08-13T15:25:36Z pierpa quit (Ping timeout: 276 seconds)
2016-08-13T15:36:26Z rumbler31 quit (Remote host closed the connection)
2016-08-13T18:05:29Z DavidGu joined #ccl
2016-08-13T18:27:49Z DavidGu quit (Ping timeout: 244 seconds)
2016-08-13T20:43:26Z pierpa joined #ccl
2016-08-13T23:48:03Z aries_liuxueyang quit (Ping timeout: 240 seconds)
2016-08-13T23:53:12Z aries_liuxueyang joined #ccl