Philip, if you can bear with so huge JVM pauses, then there is no use to investigate stacktraces anymore. Just increase systemWorkerBlockedTimeout parameter of IgniteConfiguration appropriately, as described in https://apacheignite.readme.io/docs/critical-failures-handling#section-critical-workers-health-check and Ignite 2.7 won't report these failures.
чт, 21 мар. 2019 г. в 20:42, Philip Wu <[email protected]>: > hello, Andrey - > > for your 2nd question, in Ignite 2.5, we have 15 mins + JVM paused as well, > but no IgniteException, was working fine. > > 2019-03-15 19:08:46,088 WARNING [ (jvm-pause-detector-worker)] Possible too > long JVM pause: 1001113 milliseconds. > > 2019-03-15 19:08:46,280 INFO [IgniteKernal%XXXGrid > (grid-timeout-worker-#71%XXXGrid%)] > Metrics for local node (to disable set 'metricsLogFrequency' to 0) > ^-- Node [id=1dc0de55, name=EnfusionGrid, uptime=01:49:39.992] > ^-- H/N/C [hosts=1, nodes=1, CPUs=32] > ^-- CPU [cur=100%, avg=39.77%, GC=1042.83%] > ^-- PageMemory [pages=2300496] > ^-- Heap [used=295123MB, free=14.48%, comm=345088MB] > ^-- Non heap [used=442MB, free=-1%, comm=463MB] > ^-- Outbound messages queue [size=0] > ^-- Public thread pool [active=0, idle=0, qSize=0] > ^-- System thread pool [active=0, idle=1, qSize=0] > 2019-03-15 19:08:46,280 INFO [IgniteKernal%XXXGrid > (grid-timeout-worker-#71%XXXGrid%)] FreeList [name=XXXGrid, buckets=256, > dataPages=1, reusePages=0] > > > > > > -- > Sent from: http://apache-ignite-users.70518.x6.nabble.com/ > -- Best regards, Andrey Kuznetsov.
