[c-nsp] 3825 memory issue

Tseveendorj tseveendorj at gmail.com
Tue Jun 16 20:42:32 EDT 2009


Hello,

I'm using 3825 router to provide internet for ADSL customers. Last week 
I configured policy-map to provide different bandwidth usage. Since this 
time my routers memory usage goes up.

border-r#sh processes memory sorted allocated
Processor Pool Total:  143619888 Used:  118621160 Free:   24998728
      I/O Pool Total:   36699648 Used:   10651088 Free:   26048560

 PID TTY  Allocated      Freed    Holding    Getbufs    Retbufs Process
   0   0 4186877216 4187728860    1501456     256604          0 *Dead*
 104   0  110901912    6213036   95681064          0          0 PPP IP Route
 224   0   79114360   79153804       8752          0          0 crypto 
sw pk pro
   0   0   45088716   14947936   24508036          0          0 *Init*
 219   0   26739468   26964080      13060          0          0 SNMP ENGINE
 217   0   17194704   16959140      24012          0          0 IP SNMP
   1   0   16815572   15676464    1146168          0          0 Chunk 
Manager
 225   0   12566668   12602840     170956          0          0 
VTEMPLATE Backgr
 213   0    5154176   13760992      38104          0          0 PPP Events
  18   0    4428420    4428088       7392          0          0 ARP 
Background
 210   0    3901896    1455596     422460          0          0 PPPoE 
Discovery
  88   0    3291920    3138964      82928          0          0 IP Input
 109   0    2823624          0      13060          0          0 TCP 
Protocols
 209   0    1489464     320004     606848          0          0 PPPoE 
Background
  83   0    1131888          0       7060          0          0 AAA ACCT 
Proc
 222   0     969124    4249452      67108          0          0 RADIUS
  82   0     953112        252      13664          0          0 AAA Server
 207   0     750732       4252     775844          0          0 L2TP 
mgmt daemon
  59   0     648724       1328     629396          0          0 USB Startup
 PID TTY  Allocated      Freed    Holding    Getbufs    Retbufs Process
  95   0     439864     305652     243420          0          0 SSS Manager
  77 578     413692     401112      24988          0          0 Virtual Exec
 178   0     350580    1918340       7060          0          0 AAA SEND 
STOP EV
  25   0     263208          0     272656     113400          0 EEM ED 
Syslog
 198   0     250300     250300      13060          0          0 Syslog
 167   0     247992          0     255052          0          0 
QOS_MODULE_MAIN
 188   0     119396       7344     125844          0          0 EEM Server
 122   0     108244        252     115052          0          0 SCTP 
Main Proces
 120   0     104664        252     103352          0          0 DHCPD 
Receive
  93   0      98112        504      13060          0          0 PPP Hooks
 106   0      73808          0      73808          0          0 CEF process
 142   0      73236        252      72648          0          0 FLEX 
DSPRM MAIN
  17   0      72948      32844      47164          0          0 ARP Input
  24   0      72748          0      79808          0          0 Entity 
MIB API
  40   0      67568       4424       7060          0          0 TTY 
Background
 103   0      66736          0      76796          0          0 IP RIB 
Update
   4   0      65588          0      90648          0          0 EDDRI_MAIN
 152   0      59568      73276       7060          0          0 LOCAL AAA
  54   0      57904        252      64712          0          0 VNM 
DSPRM MAIN
 206   0      56660          0      15868          0          0 SSH 
Event handle
  84   0      49384          0      56444          0          0 ACCT 
Periodic Pr
 226   0      47236      16108      16152          0          0 BGP Router

I have several questions.

1. What is *Dead* process ? it takes many memory why ?
2. Is there any unknown process working ?
3. How do I decrease memory usage ?
4. What is column of memory real usage ?

Sincerely,
Tseveen.


More information about the cisco-nsp mailing list