Details
-
Bug
-
Resolution: Fixed
-
Major
-
None
-
1.0.0
-
Security Level: Public
-
None
-
CentOS 6.4, libcouchbase 2.0.7, there 4 nodes in target cluster.
Description
I start multiple clients so that they perform about 2000 queries per second.
Eventually I receive lots of errors:
[09/Aug/2013 02:18:20] WARNING - <RC=0x18[Connection failure], There was a problem scheduling your request, or determining the appropriate server or vBucket for the key(s) requested. This may also be a bug in the SDK if there are no network issues., C Source=(src/http.c,317)>
Apparently I see many connections in TIME_WAIT state (60000-80000 connections):
tcp 0 0 172.23.96.10:52484 172.23.96.18:8092 TIME_WAIT -
tcp 0 0 172.23.96.10:52996 172.23.96.16:8092 TIME_WAIT -
tcp 0 0 172.23.96.10:52997 172.23.96.16:8092 TIME_WAIT -
tcp 0 0 172.23.96.10:52994 172.23.96.16:8092 TIME_WAIT -
Attachments
Issue Links
- depends on
-
CCBC-226 Support HTTP keepalive semantics for views
- Resolved