light code cleanup and obsolete commented-out-stuff removal |
|
More...
|
over 1 year ago
|
handle out-of-band UNSUBSCRIBEs for redis chanheads |
|
More...
|
over 1 year ago
|
cleanup setting redis chanhead pubsub status |
|
More...
|
over 1 year ago
|
relaxed timeque finishing doesn't need to verify the cmdtag -- we need this because Redis might send out-of-band UNSUBSCRIBE/SUNSUBSCRIBE commands |
|
More...
|
over 1 year ago
|
track more command stats |
|
More...
|
over 1 year ago
|
track batch unsubscribe command stats, too |
|
More...
|
over 1 year ago
|
properly reuse node stats when id doesn't match or is empty |
|
More...
|
over 1 year ago
|
cluster check, cluster recover, and channel keepalive redis command stats |
|
More...
|
over 1 year ago
|
better way to handle tag mismatch |
|
More...
|
over 1 year ago
|
split command and pubsub command queues for stats |
|
More...
|
over 1 year ago
|
fix: GET publisher location request may return incorrect subscriber count when using Redis |
|
More...
|
over 1 year ago
|
oh the valgrindity! |
|
More...
|
over 1 year ago
|
timequeue oops |
|
More...
|
over 1 year ago
|
get redis stats into a tentatively working state |
|
More...
|
over 1 year ago
|
initialize the accumulator completely, and fix the glaring issues |
|
More...
|
over 1 year ago
|
fix: possible to log invalid string during very rare error logging |
|
More...
|
over 1 year ago
|
redis stats should be datachable from the nodes so that data about intermittently disconnected nodes doesn't get lost |
|
More...
|
over 1 year ago
|
list size is useful |
|
More...
|
over 1 year ago
|
nchan_redis_upstream_stats_enabled and nchan_redis_upstream_stats_disconnected_timeout |
|
More...
|
over 1 year ago
|
we're getting there |
|
More...
|
over 1 year ago
|
"is this off-by-one-error?" |
|
More...
|
over 1 year ago
|
Redis node stats config directive |
|
More...
|
over 1 year ago
|
redis stats output and IPC-transfer |
|
More...
|
over 1 year ago
|
keep workin' on stats |
|
More...
|
over 1 year ago
|
nodeset names should be just the names, not full titles. They way they're more easily searchable |
|
More...
|
over 1 year ago
|
more redis stats work |
|
More...
|
over 1 year ago
|
working on Redis stats output and IPC |
|
More...
|
over 1 year ago
|
prepare to add redis stats strings and such |
|
More...
|
over 1 year ago
|
data accumulators -- we got exponentially decaying averages, and boring but atomic-ish sums |
|
More...
|
over 1 year ago
|
timequeue is a paged queue that stores start times for things with a tag |
|
More...
|
over 1 year ago
|