diff options
| author | Garrett D'Amore <garrett@damore.org> | 2017-01-16 10:08:24 -0800 |
|---|---|---|
| committer | Garrett D'Amore <garrett@damore.org> | 2017-01-16 10:08:24 -0800 |
| commit | 656f091858b26e6fb61e5a9dd8395268dbee2447 (patch) | |
| tree | 1e8685955a75b5989a392dba1a86492c4a8dce34 /tests/convey.c | |
| parent | 70d65fed3a230dcf17939786b5ac941423e29216 (diff) | |
| download | nng-656f091858b26e6fb61e5a9dd8395268dbee2447.tar.gz nng-656f091858b26e6fb61e5a9dd8395268dbee2447.tar.bz2 nng-656f091858b26e6fb61e5a9dd8395268dbee2447.zip | |
Fix pipeline test failure.
I think there are a couple of linked issues; it really comes down to
the fact that various threads start up asynchronously, and might not
have gotten to the point where they are ready to receive. Making matters
worse is the fact that when the message queues are unbuffered, the absence
of a reader *right then* can cause pushback, which causes the load balance
test to pass over a queue it really shouldn't.
Diffstat (limited to 'tests/convey.c')
0 files changed, 0 insertions, 0 deletions