I have observed a memory leak in 3.11.2 version similar situation which
Bobbie has mentioned, We faced corrupt sstable issue after disk space was
100% full (due to snapshots) in multiple nodes (we have default
disk_failure_policy)
After clearing the snapshots we ran repair and observed following i
Hi Bob ,
We did not use the reaper(yet) just trying to run tests against the new C*
v3 cluster .
The nodes keeps crashing all the time and these are the error we are
getting .
Any other ideas ?
Thanks!
Roy
On Wed, Jan 16, 2019 at 8:51 AM Bobbie Haynes wrote:
> Hi Roy,
> I don't think
Hi Roy,
I don't think the Memory Leak issue is related to MAP errors .I
was also using Reaper in our cluster.I have seen Memory Leak issue (ERROR
[Reference-Reaper:1] 2019-01-14 00:03:46,469 Ref.java:224 - LEAK DETECTED)
when some of SStables got corrupted because of disk space issue we
Hi ,
We are testing C* 3.11.3 and we have mapping issue and possibly leaked
memory.
It might be related to our configuration,any ideas would be helpful .
Cassandra version: 3.11.3
OS: CentOS Linux release 7.4.1708 (Core)
Kernel: 3.10.0-957.1.3.el7.x86_64
JDK: jdk1.8.0_131
Heap: same errors with
