How does the Node determine what nodes to delete from its routing table to make room for a new noderef?
 
I noticed my node seemed to delete one of the only 10 nodes that I had open connections to.
 
Why would my node delete a working noderef (it had an open connection to it) when it had 30 failed/unconnected nodes it could  have deleted instead?
 
This might explain the "disappearing routing table specialization" I have heard complained about...
 
This is on my node after I deleted the node's routing data files, so it has almost no routing knowledge and therefore is not deleting noderefs based on their NGrouting performance.
 
 
 
build 7016
 
 
Architecture and Operating System
Architecture x86
Available processors 1
Operating System Windows XP
OS Version 5.1
Java Virtual Machine
JVM Vendor Sun Microsystems Inc.
JVM Name Java HotSpot(TM) Client VM
JVM Version 1.4.2-b28
Memory Allocation
Maximum memory the JVM will allocate 130,112 KiB
Memory currently allocated by the JVM 15,524 KiB
Memory in use 7,736,488 Bytes
Estimated memory used by logger None
Unused allocated memory 8,158,192 Bytes
Data Store
Maximum size 750 MiB
Free space 348,792 KiB
Used space 419,208 KiB
Percent used 54
Total keys 2769
Space used by temp files 260 KiB
Maximum space for temp files 262,144,007 Bytes
Most recent file timestamp Thu Aug 21 23:11:44 EDT 2003
Least recent file timestamp Sun Jul 20 04:24:44 EDT 2003
Transports
Current IPv4 address 198.92.73.44
Current IPv4 port 44744
ARK sequence number 149
Last ARK sequence number inserted 149
Thread Pool
Total pooled threads 43
Available pooled threads 11
Pooled threads in use 32
Pooled Thread Consumers
Class Threads used
Checkpoint: Connection opener 22
Checkpoint: Garbage Collection Checkpoint 1
freenet.Message: QueryRejected 1
freenet.Message: QueryRestarted 1
freenet.interfaces.LocalNIOInterface$ConnectionShell 4
freenet.node.states.data.DataStateInitiator 2
freenet.node.states.request.RequestInitiator 1
 

Reply via email to