Memory foot print of the ignite nThe background cache eviction process was unable to free [10] percent of the cache for Context

[Naveen]

HI

am using 2.3

Doing PT for a service exposed thru REST

Here is a node stats for cache called Customer, which has 10M and s nodes as
part of the cluster, so each node is storing 3.3M odd records, which are on
off-heap.

Cache 'Customer(@c1)':
+--------------------------------------------------------------+
| Name(@)                     | Customer(@c1)                  |
| Nodes                       | 1                              |
| Total size Min/Avg/Max      | 3330161 / 3330161.00 / 3330161 |
|   Heap size Min/Avg/Max     | 0 / 0.00 / 0                   |
|   Off-heap size Min/Avg/Max | 3330161 / 3330161.00 / 3330161 |
+--------------------------------------------------------------+

Nodes for: Customer(@c1)
+=================================================================================================================+
|        Node ID8(@), IP        | CPUs | Heap Used | CPU Load |   Up Time   
|         Size         | Hi/Mi/Rd/Wr |
+=================================================================================================================+
| FC27563D(@n1), 10.144.114.114 | 8    | 27.94 %   | 0.17 %   | 47:01:44.189
| Total: 3330161       | Hi: 0       |
|                               |      |           |          |             
|   Heap: 0            | Mi: 0       |
|                               |      |           |          |             
|   Off-Heap: 3330161  | Rd: 0       |
|                               |      |           |          |             
|   Off-Heap Memory: 0 | Wr: 0       |
+-----------------------------------------------------------------------------------------------------------------+

Question: Above, Off-Heap: 3330161 meaning its holding these many records,
Off-Heap Memory: 0 - what does this indicate ???

[12:34:30,163][INFO][grid-timeout-worker-#23%IgnitePOC%][IgniteKernal%IgnitePOC]
Metrics for local node (to disable set 'metricsLogFrequency' to 0)
    ^-- Node [id=fc27563d, name=IgnitePOC, uptime=47:01:18.341]
    ^-- H/N/C [hosts=3, nodes=4, CPUs=24]
    ^-- CPU [cur=0.17%, avg=0.13%, GC=0%]
    ^-- PageMemory [pages=2043040]
    ^-- Heap [used=282MB, free=72.18%, comm=1017MB]
    ^-- Non heap [used=71MB, free=95.31%, comm=73MB]
    ^-- Public thread pool [active=0, idle=0, qSize=0]
    ^-- System thread pool [active=0, idle=6, qSize=0]
    ^-- Outbound messages queue [size=0]

Question: Non heap [used=71MB, free=95.31%, comm=73MB] - does it mean this
node us using only 71MB to store all the data held by this node??
        
2017-12-13 12:27:04.916  WARN 33362 --- [o-9090-exec-192]
org.apache.catalina.webresources.Cache   : Unable to add the resource at
[/customers/P100043683] to the cache for web application [] because there
was insufficient free space available after evicting expired cache entries -
consider increasing the maximum size of the cache
2017-12-13 12:27:06.951  INFO 33362 --- [Engine[Tomcat]]]
org.apache.catalina.webresources.Cache   : The background cache eviction
process was unable to free [10] percent of the cache for Context [] -
consider increasing the maximum size of the cache. After eviction
approximately [9,858] KB of data remained in the cache.
        
Question: Why I am I getting this error, I am just trying to do a GET with
Java API, here is the code snippet, exposed as Rest service and running a
jar on the server

@RequestMapping(value = "/customers/{custid}")
        public String getCustomerData(@PathVariable String custid)throws
IgniteException {
                // TODO Auto-generated method stub
                //IgniteCache<String, Customer> cache = 
ignite.cache("Customer");
            String res ="";
        res = cache.get(custid).toString();
        return res;
        }

I have still 20GB RAM available on these server and also allocated 12GB for
the JAR

Any clues why I am getting these errors



--
Sent from: http://apache-ignite-users.70518.x6.nabble.com/