2026/04/06
< 2026
Jump to navigation
Jump to search
|
Monday, April 6, 2026 (#96)
|
|
References |
- 07:55 Somewhere in the code between where the SQL is sent and where the loop tries to fetch the response, it stops working. Today's task is to figure out where.
- 08:38 The
PushBytes()call inDoCommand()is somehow not actually pushing the bytes right. - 08:57 It wasn't adding the newline. (How does that keep getting removed.)
- Still not retrieving the schema list, but there are some errors to fix...
- 09:07 ...and boom, there's the listing! Tidying up debug code...
- 09:35 ...and now I need to get exports working again...
- 12:04 ...and now a remote export seems to be happening! (Time for a commit.)
- 12:08 ...except that the output is going on far longer than it should -- I'm guessing that it's not detecting EoS correctly, and probably just writing the last bufferful over and over again.
- 12:14 This does seem to be the case -- the last table is getting written, and then the repeats begin.
- Oddly, what's getting repeated after that is a piece from the very start. I don't know if this is significant or not.
- 12:28 The local exports do not have this problem -- so it's specific to ssh2, and probably specific to ssh2 stream EoS-detection.
- 15:34 After much Lunar distraction, exports are working a little better but getting stuck on
performance_schema. I'm guessing this is probably because of the error this causes.