Every five minutes cfdcollect logs this to syslog:
Dec 17 19:22:36 pecan cfdcollect[7233]: [ID 183499 local6.info] [I]
awakened by alarm.
Dec 17 19:22:37 pecan cfdcollect[7233]: [ID 671106 local6.info] [I]
connected to localhost:2056
Dec 17 19:22:43 pecan cfdcollect[7233]: [ID 130648 local6.info] [I]
localhost has data for -1 routers.
Dec 17 19:22:43 pecan cfdcollect[7233]: [ID 151931 local6.error] [E]
FdRead(4,ffbef810,4) failed: Interrupted system call
{CflowdCisco.cc:322}
Dec 17 19:22:43 pecan cfdcollect[7233]: [ID 655994 local6.error] [E]
cflowdCisco.read(4) failed {CflowdServer.cc:711}
Dec 17 19:22:43 pecan cfdcollect[7233]: [ID 556317 local6.info] [I]
sleeping for 293 seconds.
Then cflowd dumps core. cflowd continues to show up in a ps listing
though.
--cflowd.conf--
OPTIONS {
LOGFACILITY: local6
TCPCOLLECTPORT: 2056
PKTBUFSIZE: 2097152
TABLESOCKFILE: /var/run/cflowdtable.socket
FLOWDIR: /cflow/flows
FLOWFILELEN: 200000000
NUMFLOWFILES: 20
MINLOGMISSED: 1000
}
COLLECTOR {
HOST: 127.0.0.1
AUTH: none
}
CISCOEXPORTER {
HOST: x.x.x.x
ADDRESSES: { x.x.x.x }
CFDATAPORT: 2055
SNMPCOMM: 'public'
COLLECT: { flows }
}
------cfdcollect.conf-- system { logFacility: local6 dataDirectory: /cflow/data filePrefix: arts pidFile: /var/run/cfdcollect.pid }
cflowd { host: localhost tcpCollectPort: 2056 minPollInterval: 300 } ----
--/etc/system-- set shmsys:shminfo_shmmax=16777216 --
This is on a Netra T1 running Solaris8. It has 512mb RAM and 18G of disk on /cflow
Any ideas what's going wrong? -- cflowd mailing list cflowd@caida.org
This archive was generated by hypermail 2b29 : Mon Dec 17 2001 - 16:45:38 PST