Including event queue stats in live replication output
samongyr-sq opened this issue · comments
sam / song commented
When using the replicate
subcommand with --mode live
(or --mode liveonly
), there's no way to determine when the internal queue is full from the current output
Including information about the the size of the event queue relative to its capacity would make it easier to determine when --event-queue
and/or --read-queue
should be configured to prevent updates from being dropped
Version Info
riot v3.0.4
$ riot <source> replicate <target> --mode liveonly --progress ascii
Listening ? % [ = ] 20/? (0:00:20 / ?) 1.0/s
Julien Ruaux commented
🎉 This issue has been resolved in v3.1.1
(Release Notes)
Pamela Yin commented
@jruaux This seems to be fixed for above command at v3.1.1
but it doesn't seem to work when the source is elasticache and target is redis on-prem with --type ds
.
/riot-3.1.1/bin/riot <elasticache source> replicate <redis target> --mode live --type ds
Listening ? % [= ] 0/? (0:00:00 / ?) ?/s
Scanning 100% [=========================] 4000/4000 (0:00:06 / 0:00:00) 666.7/s