java - JCS, recovering from a dead node using in memory caching only -
for example, have server acts file store (jackrabbit end). data pushed on start in memory jcs cache. seems work fine.
however, our software can multi-noded, need have jcs replicate data other node. test used following configuration files (with ports flipped each node).
# default cache region jcs.default=ltcp jcs.default.cacheattributes=org.apache.commons.jcs.engine.compositecacheattributes jcs.default.cacheattributes.maxobjects=1000 jcs.default.cacheattributes.memorycachename=org.apache.commons.jcs.engine.memory.lru.lrumemorycache jcs.auxiliary.ltcp=org.apache.commons.jcs.auxiliary.lateral.socket.tcp.lateraltcpcachefactory jcs.auxiliary.ltcp.attributes=org.apache.commons.jcs.auxiliary.lateral.socket.tcp.tcplateralcacheattributes jcs.auxiliary.ltcp.attributes.tcpservers=localhost:1987 jcs.auxiliary.ltcp.attributes.tcplistenerport=1986 jcs.auxiliary.ltcp.attributes.allowget=true
this seems initially, work, if load node 100 files (and keeploading @ 1 every 10 seconds), gets pushed across node b, , keeps updating.
however if pull down node b, , restart node b, pull in newly created files.
how can make consistency check on start, when node re-finds node b (which seems to) resyncs , sorts self out.. i.e pushes 100 + x node b again?
i want use in memory caches, disk cache jackrabbit backend (slower know, hence need in memory cache)
thanks
Comments
Post a Comment