You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@ignite.apache.org by shivakumar <sh...@gmail.com> on 2020/12/01 08:47:21 UTC
Re: connecting to visor shell permanently stops unwanted WAL
clean-up
I with 100 GB of data in persistence store this is easily reproducible not
sure what is causing this,
any one please let me know how visor gets number of records from cache and
what could be the issue of blocking wal cleanup. Iam adding wal usage graph
where without connecting to viros the wal usage was max 6GB but once i
connect visor it started growing until i stop data ingestion and it will not
even comeback from the max.
<http://apache-ignite-users.70518.x6.nabble.com/file/t2244/ignite_wal_usage.png>
--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: connecting to visor shell permanently stops unwanted WAL
clean-up
Posted by vbm <bm...@gmail.com>.
Even I had seen the same issue in ignite 2.9.0 cluster with jdk11.
We have tried disabling WAL archive but still the same issue is seen.
When we had tried in older release 2.7.6 we had not seen such issue.
Is this a known issue in 2.9 wrt WAL (FSYNC) ? Any configuration change has
to be done wrt to java 11 ?
--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/